Agriculture Economic and Sector Work, Report No. 96308-KH CAMBODIAN AGRICULTURE IN TRANSITION: OPPORTUNITIES AND RISKS May 19, 2015 Agriculture Global Practice East Asia and Pacific Region ________________________________________ i © 2015 The International Bank for Reconstruction and Development / The World Bank 1818 H Street, NW, Washington, DC 20433 All rights reserved: This volume is a product of the staff of the International Bank for Reconstruction and Development/ The World Bank. The findings, interpretations, and conclusions expressed in this paper do not necessarily reflect the views of the Executive Directors of The World Bank or the governments they represent. The World Bank does not guarantee the accuracy of the data included in this work. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgment on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Rights and Permission The material in this publication is copyrighted. Copying or transmitting any of this work without permission may be a violation of applicable law. The International Bank for Reconstruction and Development/The World Bank encourages dissemination of its work and will normally grant permission to reproduce portions of the work promptly.  For permission to photocopy or reprint any part of this work, please send a request with complete information to the Copyright Clearance Center, Inc., 222 Rosewood Drive, Danvers, MA 01923, USA, telephone 978-750-8400, fax 978-750-4470, www.copyright.com.  All other queries on rights and licenses, including subsidiary rights, should be addressed to the Office of the Publisher, The World Bank, 1818 H Street NW, Washington, DC 20433, USA, fax 202-522-2422, e-mail pubrights@worldbank.org. ii Table of Contents ACRONYMS AND ABBREVIATIONS .................................................................................... X ACKNOWLEDGMENTS .......................................................................................................... XI EXECUTIVE SUMMARY ...................................................................................................... XII 1. INTRODUCTION................................................................................................................. 1 2. A DECADE OF AGRICULTURAL TRANSFORMATION IN CAMBODIA .............. 3 2.1. AGRICULTURAL GROWTH ................................................................................................ 3 2.2. AGRIBUSINESS DEVELOPMENT ........................................................................................ 8 2.3. AGRICULTURAL GROWTH AND POVERTY REDUCTION ................................................... 10 2.4. AGRICULTURAL TRANSFORMATION ............................................................................... 13 3. CHANGES IN AGRICULTURE: EVIDENCE FROM THE FIELD ........................... 18 3.1. INTRODUCTION .............................................................................................................. 18 3.2. CROP PRODUCTION ........................................................................................................ 19 3.3. FARMLAND .................................................................................................................... 25 3.4. AGRICULTURAL LABOR FORCE ...................................................................................... 29 3.5. USE OF AGRICULTURAL INPUTS ..................................................................................... 32 3.6. USE OF AGRICULTURAL MACHINERY ............................................................................ 40 3.7. AGRICULTURAL SERVICES ............................................................................................. 42 3.8. FARMERS’ ORGANIZATIONS ........................................................................................... 50 3.9. AGRICULTURAL INFRASTRUCTURE ................................................................................ 52 4. CHANGES IN FARM BUDGETS AND FARM PROFITABILITY ............................ 55 4.1. INTRODUCTION .............................................................................................................. 55 4.2. RESULTS OF THE 2013 SURVEY ...................................................................................... 58 4.3. COMPARISON OF THE 2005 AND 2013 SURVEY RESULTS ............................................... 63 5. COMPETITIVENESS OF CAMBODIAN FARMS ....................................................... 69 5.1. INTRODUCTION .............................................................................................................. 69 5.2. DOMESTIC RESOURCE COST FOR RICE ........................................................................... 70 5.3. DOMESTIC RESOURCE COST FOR CASSAVA AND MAIZE ................................................ 72 5.4. SUMMARY ...................................................................................................................... 73 6. DRIVERS OF PAST GROWTH ....................................................................................... 74 7. DRIVERS OF FUTURE GROWTH ................................................................................. 80 8. A LONG-TERM VISION FOR CAMBODIAN AGRICULTURE ............................... 86 9. OPTIONS FOR SUPPORTING A LONG-TERM VISION FOR AGRICULTURE .. 93 iii 9.1. INTRODUCTION .............................................................................................................. 93 9.2. SCENARIOS FOR ENVIRONMENTAL SUSTAINABILITY ..................................................... 93 9.3. SCENARIOS FOR PRODUCTIVITY GROWTH...................................................................... 98 9.4. SCENARIOS FOR IMPROVED COMPETITIVENESS ............................................................ 105 9.5. SCENARIOS FOR INCOME GROWTH ............................................................................... 108 9.5 SUMMARY OF SIMULATIONS’ RESULTS ........................................................................ 113 10. POLICY AGENDA TO SUPPORT A LONG-TERM VISION FOR CAMBODIAN AGRICULTURE....................................................................................................................... 115 10.1. MAINTAINING A PRIVATE SECTOR FRIENDLY AGRICULTURAL POLICY ENVIRONMENT 115 10.2. STRENGTHENING THE ENVIRONMENTAL SUSTAINABILITY ........................................... 117 10.3. IMPROVING THE AGRICULTURAL “PUBLIC GOODS” INVESTMENTS .............................. 118 10.4. HELPING DEVELOP THE AGROPROCESSING AND AGRIBUSINESS INDUSTRY ................. 127 11. CONCLUSIONS AND POLICY RECOMMENDATIONS ..................................... 132 12. REFERENCES .............................................................................................................. 139 ANNEX 1: ANALYTICAL WORK FROM THE WORLD BANK AND AUSAID PARTNERSHIP ........................................................................................................................ 143 ANNEX 2: METHODOLOGY FOR THE 2013 FARM SURVEY...................................... 144 ANNEX 3: DETAILED FARM BUDGETS BY CROP ........................................................ 152 ANNEX 4: PROJECTION OF SELECTED ECONOMIC INDICATORS, CAMBODIA184 ANNEX 5: RESULTS OF POLICY SIMULATIONS .......................................................... 186 iv Boxes Box 1: Rural Population in Cambodia .......................................................................................... 14 Box 2: Rice Yields in Cambodia .................................................................................................. 24 Box 3: Gender Dimensions of Cambodian Agriculture ................................................................ 31 Box 4: Farmers’ Organizations in Cambodia ............................................................................... 51 Box 5: Factor Analysis and Modern versus Traditional Farmers................................................. 57 Box 6: Gross Margin versus Value Added ................................................................................... 58 Box 7: Driving Forces Underpinning Future Agricultural Transformation in Cambodia ............ 85 Box 8: Indicators of Agricultural Transformation ........................................................................ 87 Box 9: Importance of Fiscal Sustainability for Pro-Poor Agricultural Growth .......................... 119 Box 10: Fertilizer Subsidy in Sri Lanka ..................................................................................... 126 Box 11: Food Safety Investments in Thailand ............................................................................ 128 Box 12: Requirements for an Increase in Cambodian Rice Exports to China ............................ 129 Figures Figure 1: Agricultural GDP growth (constant prices), Cambodia, 2002-2012 ............................... 3 Figure 2: GDP composition by sector (constant prices), Cambodia, 1995-2012 ........................... 4 Figure 3: GDP composition by sector (current prices), Cambodia, 1995-2012 ............................. 4 Figure 4: Growth of agricultural subsectors (constant prices), Cambodia, 2002-2012 .................. 5 Figure 5: Share of subsectors in agricultural GDP, Cambodia, 2002-2012.................................... 6 Figure 6: Contribution to turnover of different types of agribusinesses, Cambodia, 2011 ............ 9 Figure 7: Poverty rate by region, Cambodia, 2004-2011.............................................................. 11 Figure 8: Drivers of poverty reduction in Cambodia, 2004-2011 ................................................ 12 Figure 9: Crop production and cultivated land areas, Cambodia.................................................. 20 Figure 10: Regional distribution of aromatic paddy production (% of farmers), 2013 ................ 21 Figure 11: Contribution of land and yield to average ag. production growth, 2003-2012 ........... 22 Figure 12: Projections for land expansion and yield changes, Cambodia, 2013 .......................... 25 Figure 13: Private land markets, Cambodia, 2013 ........................................................................ 27 Figure 14: Change in the perception about women’s role in agriculture, Cambodia, 2013 ......... 30 Figure 15: Impact of youth migration on agricultural households, Cambodia, 2013 ................... 31 Figure 16: Access to inputs, Cambodia, 2013 .............................................................................. 33 Figure 17: Access to improved seeds, Cambodia, 2013 ............................................................... 34 Figure 18: Use of manure, Cambodia, 2013 ................................................................................. 36 Figure 19: Use of fertilizers, Cambodia, 2013 .............................................................................. 37 Figure 20: Trends in chemical fertilizer use (kg nutrient/ha), select countries, 2005-2011 ......... 38 Figure 21: Use of pesticides, Cambodia, 2013 ............................................................................. 39 Figure 22: Use of agricultural equipment, Cambodia, 2013 ......................................................... 41 Figure 23: Access to extension services, Cambodia, 2013 ........................................................... 44 Figure 24: Access to financial services, Cambodia, 2013 ............................................................ 47 v Figure 25: Access to information, Cambodia, 2013 ..................................................................... 49 Figure 26: Reasons for creation of FOs and their perceived benefits, Cambodia, 2013 .............. 51 Figure 27: Irrigation, Cambodia, 2013 ......................................................................................... 53 Figure 28: Agricultural facilities, Cambodia, 2013 ...................................................................... 54 Figure 29: Average land by farm size and crop, Cambodia, 2013 ............................................... 56 Figure 30: Gross revenues ($/ha) by crop and farm practice, Cambodia, 2013 ........................... 59 Figure 31: Total variable costs ($/ha) by crop and farm practice, Cambodia, 2013..................... 60 Figure 32: Breakdown of cost structure by crop, Cambodia, 2013 .............................................. 60 Figure 33: Comparison of nominal gross margins ($/ha), Cambodia, 2005 and 2013 ................. 65 Figure 34: Farm wage, off-farm wage, and returns to agricultural labor ($/day), Cambodia ...... 67 Figure 35: CPI and food price index in Cambodia, 2002-2012 .................................................... 75 Figure 36: Transportation costs in Cambodia and neighboring countries, $/100 km/ton ............. 77 Figure 37: Value of the projected food demand in East Asia, $ billion, 2030.............................. 82 Figure 38: Perspectives on the possibility of change in cultivated areas by crop, Cambodia ...... 94 Figure 39: Labor use in production (days/ha), Cambodia .......................................................... 100 Figure 40: Return to labor ($/day), Cambodia ............................................................................ 100 Figure 41: Shift in land used by traditional and modern farmers, Cambodia ............................. 101 Figure 42: MOWRAM’s projected investments, $ million, 2014-2018 ..................................... 121 Figure 43: Structure of MOWRAM’s projected investments by source .................................... 121 Tables Table 1: Impact of lower agric. growth on selected agriculture sector indicators, Cambodia .... xvi Table 2: Summary of the report’s key policy recommendations based on the time of impacts’ realization .................................................................................................................................... xvii Table 3: Gross agricultural production, annual growth, selected countries, 2002-2012 ................ 5 Table 4: Production, area, yields, and growth of major crops, Cambodia, 2002-2012 .................. 7 Table 5: Irrigated areas, selected countries ..................................................................................... 7 Table 6: Production, area, yield, and export of rubber, Cambodia, 2002-2011.............................. 8 Table 7: Evolution of formal trade of select commodities, Cambodia, 1996-2011 ........................ 8 Table 8: Agribusiness industry in Cambodia, 2002-2012 .............................................................. 9 Table 9: Rice commodity balance (‘000 tons), Cambodia, 2008-2013 ........................................ 10 Table 10: Growth elasticity of poverty, selected countries in Asia .............................................. 11 Table 11: Indicators of structural transformation, Cambodia, 2004 and 2012 ............................. 13 Table 12: Country groups based on the role of agriculture in the economy ................................. 14 Table 13: Selected economic indicators, Cambodia, 2004 and 2012 ........................................... 15 Table 14: Labor use for rice production in Central Luzon, Philippines, 1966/67-2011/12 .......... 16 Table 15: Diversification in cultivated area, Cambodia, 2002-2011 ............................................ 19 Table 16: Change in farm size, Cambodia, 2008-2012 ................................................................ 23 Table 17: Average crop yields in selected countries (tons/ha), 2010-2013 .................................. 23 Table 18: Wages of labor hired for rice production, selected countries, 2013 ............................. 28 Table 19: Changes in land prices between 2005 and 2013, Cambodia ........................................ 28 vi Table 20: Use of labor in farm production by crop, Cambodia, 2005 and 2013 .......................... 29 Table 21: Use of labor in dry season paddy production, selected countries, 2013 ....................... 30 Table 22: Illustrative use of fertilizers, Cambodia, 2005 and 2013 .............................................. 38 Table 23: Number of agricultural equipment units, Cambodia, 2006-2010 ................................. 40 Table 24: Average costs of and difficulty finding agricultural machinery, Cambodia, 2013 ...... 42 Table 25: Categorization of farm size by cultivated crop areas, Cambodia ................................. 55 Table 26: Gross margin ($/ha) by crop, technology use, and farm size, Cambodia, 2013 ........... 61 Table 27: Returns to labor ($/day) by crop, technology use, and farm size, Cambodia, 2013 ..... 62 Table 28: Comparison of key farm budget indicators, $/ha, Cambodia, 2005 and 2013 ............. 64 Table 29: Comparison of total crop budgets, Cambodia, 2005 and 2013..................................... 65 Table 30: Changes in crop margins in nominal and real terms, Cambodia, 2005 and 2013 ........ 66 Table 31: Annual changes in crop margins in real terms, Cambodia, 2005 and 2013 ................. 66 Table 32: Changes in returns to labor in nominal and real terms, Cambodia, 2005 and 2013 ..... 67 Table 33: DRCs for selected agricultural crops, Cambodia, 2013 ............................................... 70 Table 34: Sensitivity analysis of impacts of changes of production costs on DRC for rice ......... 71 Table 35: Sensitivity analysis of impacts of changes in prod costs on cassava and maize DRCs 72 Table 36: Sensitivity analysis of impacts of changes in input and output prices on DRCs .......... 73 Table 37: Drivers of agricultural growth during 2002-2012 ........................................................ 74 Table 38: Global agricultural prices, selected commodities, $/ton (in real 2010 prices) ............. 75 Table 39: Selected economic projections, Cambodia, 2012-2030 ................................................ 80 Table 40: Caloric food demand projections, world and selected regions/countries ..................... 81 Table 41: Global rice demand projections (kg per capita) per year .............................................. 82 Table 42: Global agricultural terms of trade, index projections (in real 2010$) .......................... 83 Table 43: Indicators of agricultural transformation at 5 percent agricultural growth................... 87 Table 44: Changes of indicators of ag transformation at 3 percent comp. with 5 percent growth 88 Table 45: Proposed indicators and targets for a Long-Term Vision for Cambodian Agriculture 91 Table 46: Assumptions about land expansion for Env. Sustainability scenarios L1-L3 .............. 95 Table 47: Simulation results for Weak Environmental Sustainability scenarios L1-L3 .............. 96 Table 48: Simulation results for Strong Environmental Sustainability scenarios L4-L6 ............. 98 Table 49: Assumptions about Productivity Growth scenarios P1-P3 ......................................... 102 Table 50: Simulation results for Productivity Growth scenarios P1-P3 ..................................... 103 Table 51: Simulation results for Productivity Growth scenario P4 ............................................ 104 Table 52: Assumptions about Competitiveness scenarios C1-C6 .............................................. 105 Table 53: Simulation results for the Competitiveness scenarios C1-C6..................................... 107 Table 54: Assumptions about rice variety changes under Income Growth scenarios I1-I2 ....... 109 Table 55: Simulation results for Income Growth scenarios I1-I2............................................... 110 Table 56: Assumption about cassava processing under Income Growth scenarios I3-I4........... 110 Table 57: Assumptions about Income Growth under scenarios I5-I6 ........................................ 112 Table 58: Summary of the simulations’ results .......................................................................... 114 Table 59: Key issues and options to reform the seed sector in Cambodia ................................. 117 vii Table 60: Agricultural research intensity.................................................................................... 123 Table 61: Yield response for supplementing nitrogen to soil, Cambodia................................... 125 Table 62: Prices and logistics costs of rice in Cambodia, Vietnam, and Thailand ..................... 130 Table 63: Time and cost in rice export procedures, Cambodia, 2012-2013 ............................... 130 Table 64: Summary of the report’s key policy recommendations by implementing agency ..... 138 Table 65: Product selection for the 2013 field survey ................................................................ 144 Table 66: Selected locations for 2013 field work ....................................................................... 146 Table 67: Matrix of size x technology ........................................................................................ 148 Table 68: List of respondents to 2013 survey ............................................................................. 150 Table 69: Comparison of key indicators across type of commercial cassava............................. 157 Table 70: Wet season rice farm budget by technology and farm size ........................................ 161 Table 71: Wet season rice farm budget by farm size .................................................................. 162 Table 72: Wet season rice farm budget by technology use ........................................................ 163 Table 73: Wet season rice farm budget by province................................................................... 164 Table 74: Dry season rice farm budget by technology and farm size ......................................... 165 Table 75: Dry season rice farm budget by farm size .................................................................. 166 Table 76: Dry season rice farm budget by technology use ......................................................... 167 Table 77: Dry season rice farm budget by province ................................................................... 168 Table 78: Cassava farm budget by technology and farm size .................................................... 169 Table 79: Cassava farm budget by farm size .............................................................................. 170 Table 80: Cassava farm budget by technology use ..................................................................... 171 Table 81: Cassava farm budget by province ............................................................................... 172 Table 82: Maize farm budget by technology and farm size........................................................ 173 Table 83: Maize farm budget by farm size ................................................................................. 174 Table 84: Maize farm budget by technology use ........................................................................ 175 Table 85: Maize farm budget by province .................................................................................. 176 Table 86: Vegetable (mix) farm budget by technology and farm size ....................................... 177 Table 87: Vegetable (mix) farm budget by farm size ................................................................. 178 Table 88: Vegetable (mix) farm budget by technology use ........................................................ 179 Table 89: Vegetable (mix) farm budget by province .................................................................. 180 Table 90: Changes in prices of commodities, Thailand and Cambodia ..................................... 181 Table 91: Change in the prices of inputs, services, and labor, Cambodia .................................. 182 Table 92: Average use of inputs, Cambodia, 2013 ..................................................................... 182 Table 93: Farm budget indicators by crop, farm size, and technology use, Cambodia, 2013 .... 183 Table 94: Results of simulations on land use (baseline) ............................................................. 186 Table 95: Results of simulations on land use (scenario L1) ....................................................... 187 Table 96: Results of simulations on land use (scenario L2) ....................................................... 188 Table 97: Results of simulations on land use (scenario L3) ....................................................... 189 Table 98: Results of simulations on land use (scenario L4) ....................................................... 190 Table 99: Results of simulations on land use (scenario L5) ....................................................... 191 viii Table 100: Results of simulations on land use (scenario L6) ..................................................... 192 Table 101: Demand for labor for environmental sustainability scenarios .................................. 193 Table 102: Demand for fertilizers for environmental sustainability scenarios ........................... 194 Table 103: Demand and average return to labor ($/day) for scenarios with env sustainability . 195 Table 104: Farm budget data comparing traditional and modern technology users (revenue and modern inputs) ............................................................................................................................ 196 Table 105: Farm budget data comparing traditional and modern technology users (labor, services, and gross margins) ....................................................................................................... 197 Table 106: Gross margins and NPV (million $) for TFP scenario (baseline) ............................ 198 Table 107: Gross margins and NPV (million $) for TFP scenario (scenario P1) ....................... 199 Table 108: Gross margins and NPV (million $) for TFP scenario (scenario P2) ....................... 200 Table 109: Gross margins and NPV (million $) for TFP scenario (scenario P3) ....................... 201 Table 110: Summary of gross margins for TFP scenarios .......................................................... 202 Table 111: Examples of costs of irrigation investment (per unit and total investment) ............. 202 Table 112: Demand for labor and return to labor for TFP scenarios .......................................... 203 Table 113: Demand for fertilizers for TFP scenarios ................................................................. 204 Table 114: Cultivated land and gross margins for scenario P4 (TFP through irrigation)........... 205 Table 115: Gross margins on competitiveness (scenarios C1-C6) ............................................. 206 Table 116: Summary of gross margins and NPV for Competitiveness scenarios (C1-C6) ........ 207 Table 117: Return to labor for Competitiveness scenarios (C1-C6)........................................... 208 Table 118: Demand for labor and demand for fertilizers with Competitiveness scenarios ........ 210 Table 119: Characteristics of rice production: fragrant vs. IRRI................................................ 211 Table 120: Cultivated land areas for fragrant rice Income Growth scenarios ............................ 212 Table 121: Gross margins for fragrant rice Income Growth scenarios ....................................... 213 Table 122: Summary of gross margins for fragrant rice Income Growth scenarios ................... 214 Table 123: Demand for labor, return to labor and demand for fertilizers for fragrant rice Income Growth scenarios ........................................................................................................................ 214 Table 124: Cultivated land areas for cassava processing Income Growth scenarios ................. 215 Table 125: Gross margins for cassava processing Income Growth scenarios (million $) .......... 216 Table 126: Comparison between fresh cassava and dry chips production ................................. 217 Table 127: Demand for and return to labor for cassava processing Income Growth scenarios . 218 Table 128: Distribution and farm budget by farm size ............................................................... 219 Table 129: Gross margins with Farm Size Distribution scenario (baseline) .............................. 221 Table 130: Gross margins with Farm Size Distribution scenario (I5) ........................................ 222 Table 131: Gross margins with Farm Size Distribution scenario (I6) ........................................ 223 Table 132: Summary of gross margins from changes on Farm Size Distribution ($ million).... 224 Table 133: Demand for labor and return to labor for Farm Size Distribution scenarios ............ 225 ix ACRONYMS AND ABBREVIATIONS ADB Asian Development Bank ASEAN Association of Southeast Asian Nations CARDI Cambodian Agricultural Research and Development Institute CCSF Cambodian Community Savings Federation CDRI Cambodia Development Resource Institute CICP Cambodian Institute for Cooperation and Peace CPI Consumer price index DAP Diammonium phosphate DRC Domestic resource cost EBA Everything but Arms FAO Food and Agriculture Organization of the United Nations FAOSTAT Food and Agriculture Organization Corporate Statistical Database FGD Focus group discussion FO Farmers’ organization FWUC Farmer Water User Communities GDP Gross domestic product ha Hectares IFC International Finance Corporation IRR Internal rate of return MAFF Ministry of Agriculture, Forestry, and Fisheries MFI Microfinance institution MOWRAM Ministry of Water Resources and Meteorology NGO Nongovernmental organization NIS National Institute of Statistics, Cambodia NPV Net present value O&M Operation and maintenance PDA Provincial Departments of Agriculture PPP Purchasing power parity RGC Royal Government of Cambodia SOWS-REF Secretariat of One Window Service for Rice Export Formality SNEC Supreme National Economic Council TFP Total factor productivity TOT Terms of trade tph Tons per hour USAID United States Agency for International Development WDI World Development Indicators WRS Warehouse Receipt System x ACKNOWLEDGMENTS This report was prepared by the World Bank’s Agriculture Global Practice, with financial support from Australian Aid. The report preparation was initiated by Paavo Eliste (Lead Rural Development Specialist, GFADR) and completed by Sergiy Zorya (Senior Economist, GFADR). The team included Mudita Chamroeun (Senior Rural Development Specialist, GFADR) and Munichan Kung (Rural Development Officer, GSURR). The team was supported by Narya Ou (Program Assistant, EACSF) in Phnom Penh. Primary data were collected and background reports were prepared by AgriFood Consulting International, with the team composed of Francesco Goletti, Jean Claude Randrianarisoa, and Cheng Vanny. The report benefited from comments from its peer reviewers: Madhur Gautam (Lead Economist, GFADR), Enrique Aldaz-Caroll (Senior Economist, GMFDR), and Hendericus Bastiaan Rozemuller (Operations Officer, CMGSB/IFC). Helpful guidance and supports was also received from Ulrich Zachau (Country Director, EACTF), Alassane Sow (Country Manager, Cambodia, EACSF), Louise Scura (Program Leader, EACTF), Nathan Belete (Practice Manager, GFADR), and Steven Jaffee (Lead Rural Development Specialist, GFADR). The task team is grateful to H.E. Dr. Hang Chuon Naron, Minister of Education, Youth and Sports for initiating this analytical work and H.E. Dr. Ouk Rabun, Minister of Agriculture, Forestry and Fisheries (MAFF) for taking ownership and support; H.E. Ty Sokhun, Secretary of State, MAFF, and Co-Chair of the Technical Working Group for Agriculture and Water, who supported and coordinated the consultation process with government agencies and other stakeholders; H.E. So Khanrithykun, General Director of the MAFF’s General Directorate of Agriculture, and other MAFF staff for the time they devoted to provide information, discussions, and support during preparation of the report. We would also like to acknowledge valuable comments and contributions from government agencies in Phnom Penh and in provinces, from the donor community, including those received through the joint Government-Donors Technical Working Groups, and from the many representatives of the private sector, research institutes and NGOs, who contributed their views and feedback on the initial drafts, background reports, and the 2013 survey. The excellent comments and feedback from the participants of the consultative workshops held in May 2014, January 2015, and May 2015 in Phnom Penh are particularly appreciated. We are grateful to Nicholas Wolf and other Australian Aid staff for support for preparation of this report provided under the Cambodia Food Crisis Capacity Support Partnership Trust Fund and the preparation of many other reports produced under this partnership that fed into this strategic analytical work. The document was edited by Amy Gautam and Tiina Joosu-Palu assisted with the preparation of communication materials and dissemination. xi EXECUTIVE SUMMARY This report seeks to understand the successes, challenges and opportunities of Cambodia’s agricultural transformation over the past decade to derive lessons and insights on how to maintain future agricultural growth, and particularly on the government’s role in facilitating it. It is prepared per the request of the Supreme National Economic Council and the Ministry of Agriculture Forestry and Fisheries and is based on the primary farm data surveys from 2005 and 2013, and the secondary data from various sources. In 2013- 2014, the agricultural growth slowed down to 1 percent from its average of 5.3 percent over 2004- 2012. Is the country in transition to a slower agricultural growth? Cambodia can ill afford it because agricultural growth will be critical to continued poverty reduction in the country, given its large size in the economy. Market and private investment friendly policies and targeted public sector investments in irrigation, extension, and other “public good” agricultural services, as feasible within the government’s total budget, can help secure continued robust agricultural growth. KEY FINDINGS THE PAST DECADE’S AGRICULTURAL GROWTH IS A STORY OF MANY SUCCESSES In the last decade, Cambodia’s agriculture sector has undergone significant structural transformation. Although still playing a large role, the agriculture sector became relatively less “important” in overall GDP and the total labor force, but more “productive”, i.e. agricultural land and labor productivity has increased. This transformation was driven by vibrant and pro-poor agricultural growth. The agricultural growth in Cambodia was high. During 2004-2012, the annual growth in agricultural gross production was 8.7 percent. Agricultural value added1 grew by 5.3 percent during this period. This exceptional growth, among the highest in the world, was driven by crop production, mainly of paddy rice (annual growth of 9 percent), but also maize (20 percent), cassava (51 percent), sugarcane (22 percent), and vegetables (10 percent). The growth in livestock and fisheries was modest. The agricultural growth was also pro-poor. Cambodia’s poverty headcount declined from 50 percent in 2007 to 21 percent in 2011, with the number of poor declining from 7 million to 3 million. Most poverty reduction took place in rural areas. More than 60 percent of the poverty reduction was attributed to the agriculture sector: higher rice prices stimulated the larger rice production that increased farm wages. Further poverty reduction will continue to depend on the success of agriculture for many years to come, due to its large role in the labor force, value added, and exports, as well as the fact that many farmers are among the poor and vulnerable. 1 The agricultural value added is defined as the value of gross production less the value of intermediate farm inputs. xii Agricultural wages have been converging with nonagricultural wages. Agricultural wages grew by 206 percent between the surveys conducted in 2005 and 2013, while nonfarm wages increased by 60 percent. As a result, the ratio between per worker nonagricultural to agricultural valued added in current prices fell to 2.1 in 2012 from 3.2 in 2004. Returns to farm labor have been increasing. The diversification of crop production has started. Although crop production is still mainly focused on paddy, the crop mix has been changing, driven by higher profitability of non-rice crops. In 2013, average farm gross margins (and returns to labor) were $506/ha ($9.4/day) for cassava, $303/ha ($8.8/day) for maize, and $1,393/ha ($7.2/day) for vegetable production, compared to $245/ha ($4.6/day) for wet season rice and $296/ha ($9.6/day) for dry season rice. Dry season rice competes with non-rice crops in terms of returns to labor but its expansion is constrained by limited irrigation. The share of total area planted under paddy declined from 86 percent in 2002 to 74 percent in 2011, while the share of planted area for maize and cassava production increased significantly. Even the paddy sector has started to diversify. Triggered by the demand from modernized rice mills, more farmers grow more profitable aromatic paddy, estimated at 10 percent of rice cultivated area and 30 percent of total production. Further expansion of farmland under aromatic paddy is possible with improvements in quality seed supply, agricultural extension, and irrigation. Yields increased for most crops. With the annual growth in cultivated land areas at 4.7 percent and agricultural gross production at 8.7 percent, the average growth of yields was 4 percent during 2004-2012. This growth in yields was triggered by wide adoption of improved technologies, expanded (yet still limited) irrigation, more use of modern inputs, and better access to mechanized services, pointing to the advances in commercialization. Farmers also have better access to markets. Cambodian farm products remain price competitive at farm gate. A domestic resource cost analysis of competitiveness shows that despite rising labor costs and prices of farm inputs, the value added generated by farmers exceeds the costs of domestic factors of production (land, labor, and capital). This high competitiveness explains the large increase in agricultural exports in the recent decade. However, the competitiveness of ordinary rice produced during the wet season has xiii worsened in recent years and many small farmers producing ordinary rice started to lose their competitive edge. Past agricultural growth was driven by several factors. Among the major ones were open trade and, in general, market-oriented agricultural policy. Cambodia was one of the few developing countries that did not overact to the 2008 global food price spike but actually saw higher food prices as an opportunity to leverage agricultural growth. Other net-exporting countries such as India and Vietnam used export restrictions to limit transmission of the global food price spike into their markets. Higher agricultural prices in Cambodia made farmland expansion profitable. In addition, the agriculture sector benefited from: (i) improved access to overseas markets through the Everything but Arms Agreement with the European Union and open cross- border trade with neighbors; (ii) better availability and wider use of mechanization services triggered by the higher cost of rural labor; (iii) better farm access to finance; and (iv) the private sector investments in rice mills. BUT THERE ARE CHALLENGES A large share of the past agricultural growth was driven by farmland expansion. The average contribution (weighted by crop areas) of land expansion to the change in farm gross margins in real terms between 2005 and 2013 was about 60 percent. Farmland expanded annually by 4.7 percent, with very large increases (128 percent) for cassava. This farmland expansion has contributed to accelerated deforestation, especially in upland areas. The average increase in per hectare gross margin2 was good, at 3.4 percent per year. But it Changes in farm gross margins, real terms, 2005-2013, % varied between 2.1 percent for dry season rice to 44.5 Change in Cultivated Change in Gross Margin Change in Total percent for vegetables. Maize’s per hectare margin Area per Hectare Gross Margin even declined. In many cases, farmers who expanded Wet season rice 2.1 2.4 4.5 their land areas received higher incomes, but farmers Dry season rice 6.7 2.1 8.8 with unchanged land areas were not able to Cassava 128. 1 15.1 143. 2 substantially increase their incomes. The period of Maize 17.2 -10.3 6.9 relatively high food prices was largely used to expand Vegetables 6. 4 44.5 50. 9 land areas rather than to build a strong foundation Weighted average 4. 7 3. 4 8.1 through productivity increases. Source: 2005 and 2013 surveys. While poverty reduced significantly, the number of vulnerable people also increased significantly. Most people who escaped poverty did so by a small margin. The loss of only 1,200 Riels per day (the cost of two small water bottles) would cause Cambodia’s poverty rate to double to 40 percent. This high rate of vulnerability is the sign of still modest agricultural productivity increases. Nutritional security has also lagged behind the reduction in poverty: although the prevalence of stunting among children under five has declined by 7 percent between 2010 and 2014, it remained at the high rate of 33 percent. 2 Gross margin is defined as gross revenue less intermediate inputs, including hired labor. The farm gross margin estimated in this report includes major, but not all crops, i.e. rice, cassava, maize, and vegetables. xiv Poor are those living below $1.15 per day (poverty line) Near poor or vulnerable are those living above $1.15 per day, but below $2.30 per day Vulnerability is greatest among the smallest farms. Small farmers, with land area less than a hectare, reported finding it difficult to expand and integrate into the emerging modern food value chains. The share of such farms in total number of rural households owning land was 48 percent in 2011. In the last decade, there has been the trend in Cambodia for large farms (above 3 ha) to become larger and small farms (less than 1 ha) to become smaller.3 The average size of the interviewed farms4 with less 1 ha declined from 0.99 ha in 2008 to 0.88 ha in 2012, while the average size of medium farms (between 1 and 3 ha) increased from 1.55 ha to 2.38 ha and the larger farms (above 3 ha) from 3.61 ha to 7.03 ha. In spite of the productivity improvements of some small farms in the past, the productivity of most small traditional farms has remained low, as agricultural extension and other public services have not reached them at a large scale. It appears that the income increase for this group of farmers in the last decade largely came from higher production values, driven by high agricultural prices, and the sale of their labor to larger farms at higher wages, rather than higher land and labor productivity. Except for rice, the agroprocessing industry has played a limited role in agricultural growth. Almost all crops were exported to neighboring countries unprocessed. This indicates serious weaknesses in the value chain, particularly in the post-harvest system of supply chain management (collection of raw material, storage, finance, logistics, transportation, and information). WHAT WORKED IN THE PAST WILL NOT BE SUFFICIENT IN THE FUTURE With global food prices declining and the land frontier diminishing, Cambodian agriculture is losing its two major growth drivers. Moreover, agricultural labor is becoming scarcer and more expensive, adding to production costs. Global agricultural prices are projected to continue their decline in the next decade, and selling low-quality ordinary rice on domestic and foreign markets will be less and less profitable. Relying on higher domestic demand driven by the increase in GDP and higher import demand alone, without more efficient use of resources, will not be enough to maintain the high rate of past growth. Cambodia would need to find other drivers of growth to help maintain agricultural growth at or near 5 percent and make it more sustainable. The recent slowdown in agricultural growth in Cambodia may signal a potential longer-term decline. As global food prices gradually declined and the global rice market became more competitive with the reentry of Thailand and Myanmar, rice production in Cambodia stalled. And with it, total crop value added declined, from 4.9 percent in 2012 to 0.6 percent in 2013. The 3 This change in distribution of farmland is occurring independently of economic land concessions. 4 The farm survey was carried out for this report in 2013, see Annex 2 for details. xv growth in agricultural value added slowed down to 1.6 percent in 2013 after average growth of 5.3 percent between 2004 and 2012. Returning to a high agricultural growth rate is important. Agriculture still maintains a large share in GDP, trade, and the labor force, which means that it will remain the key to further reduction in poverty and vulnerability in the upcoming decade. If Cambodia’s structural transformation continues, with agricultural growth averaging 5 percent, by 2030 Cambodia will have relatively smaller but more productive agriculture sector. The agriculture’s share in GDP and total labor force are projected to go down respectively to 17 percent and 31 percent in 2030, from 26 percent and 51 percent in 2012, while land productivity would increase from $1,300/ha to $2,700/ha and labor productivity would rise from $1,200/person to $3,700/person (Table 1). By contrast, the cost of slow agricultural growth would be huge. If average agricultural growth is only 3 percent between 2012 and 2030, agricultural value added would be 29 percent lower by 2030 compared with a 5 percent growth scenario, slicing overall GDP by 18 percent, keeping more people in agriculture, reducing agricultural labor productivity by 34 percent, and slowing any further income convergence between farmers and nonfarmers. Lower agricultural growth would also lead to a much slower reduction in poverty. Table 1: Impact of lower agricultural growth on selected agriculture sector indicators, Cambodia Indicators 2012 2030 3% Ag GDP growth 5% Ag GDP growth Share of agriculture in GDP, % 26 15 17 Share of agriculture in labor force, % 51 34 31 Agricultural labor productivity, $/person 1,200 2,450 3,700 Agricultural land productivity, $/ha 1,300 1,900 2,700 Labor productivity ratio of agricultural to 2.1 2.1 1.6 nonagricultural workers A long-term vision for Cambodian agriculture would need to encompass elements of sustainability, productivity, competitiveness, and income growth, shifting from still strong focus on production. Simulation of different scenarios for future agricultural development reveals that a rise in farm productivity would have the largest positive effect on farm incomes, especially if the shift from traditional to modern technologies is accompanied by higher efficiency of modern input use and irrigation. Farmers producing fragrant rice, processing cassava into dry chips, and undertaking other value addition activities can further increase their incomes, illustrating the importance of the agroprocessing industry. Lowering production costs through better use of existing resources (e.g., productivity increase) and minimizing drops in farm output prices through lower logistics costs are the keys to maintaining farm competitiveness. Continued land expansion also provides additional income but sustainability considerations will limit large expansions in the future. This constraint makes agricultural productivity, commercialization, and diversification even more critical for ensuring future agricultural growth that reduces poverty and boosts shared prosperity in Cambodia. RECOMMENDATIONS Continued rapid agricultural growth and further structural transformation in the agriculture sector are possible in Cambodia. Four sets of policies together can help support such continued agricultural growth during the next five years (short-to medium run). The first is xvi maintaining a private sector friendly agricultural policy environment, with added attention to lower the regulatory burden in farm input sectors. The second is strengthening the environmental sustainability of agricultural production. The third is improving the quality of agricultural public programs and, as feasible within Cambodia’s total government budget, increasing allocations to more effective programs. And the fourth is helping develop the agribusiness and agroprocessing industry. Investments and policy improvements are needed immediately but not all will show immediate results. For some of them, more time and effort are required to have an impact on the ground. Table 2 summarizes key policy recommendations that need to be set in motion immediately to produce impacts in the short to medium run that would help reverse the recent slowdown in agricultural growth and help fully utilize the significant potential of Cambodia’s agriculture. Table 2: Summary of the report’s key policy recommendations based on the time of impacts’ realization Policy recommendations Impacts Impacts realizable realizable in in Short Medium-to- Run Long Run Maintaining a private sector friendly agricultural policy environment Continue the open trade policy and non-distortive agricultural policy X Reduce the regulatory costs of importation of fertilizers X Open up seeds sector for private investments through deregulation and X institutional strengthening Strengthening the environmental sustainability Improve land use planning and better enforce land laws/regulations X Strengthen land tenure security X Promote sustainable land management practices X Promote the safe use of agricultural chemicals X Improving agricultural “public good” investment programs As feasible within Cambodia’s total government budget, increase the budget for core agricultural public goods, especially irrigation, applied X research, extension, soil nutrient management, input quality control, food safety, vocational training, and rural roads In irrigation, pay more attention to upgrades and rehabilitation of existing systems and participatory management of irrigation X infrastructure Improve the quality of agricultural public programs X Facilitate integration of small farms into food value chains X Helping develop agribusiness and agroprocessing industry Improve access to and reduce electricity costs X Invest in food safety public infrastructure X Improve rice trade logistics, including cost reduction X Remove barriers for introduction of warehouse receipts X xvii សេចក្ដីេសខេបរួម របាយការណ៍ស េះ បា ព្យាយាមស្េខ ែ យល់ព្យសី ោគជ័យ បញ្ហាប្បឈម ខ ុ វិេយ ិ កាលា ុវត្ដភាព្យន បម្លាេ់បរដូ ក្នខ ័ ក្េិក្មម សៅក្មពុោ សៅក្នខុ ទេវត្សរ៍ក្ ខា សៅស េះ ស ម ើ បីទាញ បា ោសមសរៀ ិ គំ ត្ ខ ់ េប្ម្លប់រក្ាក្ំសណើ ក្េិក្មម នាសព្យលអនាគត្ ោព្យិសេេ ិ េំខា ៗ ត្ួនាទីរបេ់រដ្ឋាភិបាល សៅក្នខ ុ ការបខកលក្េណៈ ឱ្យម្ល ក្ំសណើ ស្បបស េះ។ របាយការណ៍ស េះប្ត្ូវបា សរៀបចំស ើខ សៅតាម ការសេនើេុំរបេ់ ឧត្ដមប្ក្ុមប្បឹក្ាសេ ាក្ិចចោត្ិ ិខប្ក្េួខក្េិក្មម រុកាេប្បម្លញ់ ិខស សាទ សដ្ឋយស្អែក្សៅតាមការអសខកត្ប្បមូលទិ ន ័យផ្ទាល់ព្យីក្េិដ្ឋា ព្យីឆ្ន២ ំ ០០៥ ល់២០១៣ ិខទិ ន ័យស្ លបា ព្យីប្បភព្យសអសខៗ។ សៅក្ំ ុខឆ្នំ ២០១៣-២០១៤ ក្ំសណើ ក្េិក្មមម្ល ការថមថយមក្សៅប្ត្ឹម ១ភាគរយ ព្យីក្ំសណើ ោមធ្យម ៥,៣ភាគរយ សៅក្ំ ុខឆ្នំ ២០០៤-២០១២។ សត្ើប្បសទេក្មពុោ េថិត្សៅក្នុខអ ដរកាលសៅរក្ក្ំសណើ សេ ាក្ិចចយឺត្ោខមុ ឬ? ប្បសទេក្មពុោមិ អាចបសដ ដ យឱ្យម្ល បញ្ហាស្បបស េះ ប ដសក្ើត្ម្ល ស ើខស ើយ សដ្ឋយសេត្ុថាក្ំសណើ ក្នុខវិេ័យក្េិក្មមម្ល សារៈេំខា ់ េប្ម្លប់ប ដការកាត្់ ប ថយភាព្យប្ក្ីប្ក្សៅក្នុខប្បសទេស េះ ិខសដ្ឋយសារស្ត្ក្េិក្មមម្ល េម្លេភាគធ្ំ សៅក្នុខសេ ាក្ិចចរបេ់ប្បសទេក្មពុោ។ សោល សយាបាយ ស្ លបខកភាព្យងាយប្េួល ល់ទីអារ ិខការវិ ិសយាគឯក្ជ ិខការវិ ិសយាគក្នុខវិេ័យសាធារណៈ ស្ លសផ្ទដត្ចំសោលសៅ ូចោ វិេ័យធារាសាស្ដេដ អសព្យែអាយបសចចក្សទេក្េិក្មម ិខ សេវាក្េិក្មមោ “ប្បសយាជ ៍សាធារណៈ” នទសទៀត្ សៅតាមលទធភាព្យស្ លអាចសធ្ែើបា ក្នុខរខែខ់លទធភាព្យថវិកាេរុបរបេ់រដ្ឋាភិបាល អាច ឹខជួយរក្ាឱ្យវិេ័យក្េិក្មមសៅស្ត្ប ដក្ំសណើ រឹខម្លំបា ។ ើ េំខា ៗ លទធអលរក្ស ញ ់ ័ ក្េិក្មមសៅក្នខ ក្ំសណើ ន វិេយ ា សៅ បា បងាាញព្យីសោគជ័យោសប្ចើ ុ ទេវត្សរ៍ក្ ខ ុ ទេវត្សរ៍ចខ សៅក្នខ ័ ក្េិក្មមរបេ់ប្បសទេក្មពោ ុ សប្កាយស េះ វិេយ ធ ំខា ៗ ុ បា ឆ្ាខកាត្់បម្លាេ់បដូររចនាេមព័ េ ់ គួរឱ្យក្ត្់េម្លាល់។ សបើសទាេះបីោសៅម្ល ត្ួនាទី ៏ធ្ំយា៉ាខដក្៏សដ្ឋយ វិេ័យក្េិក្មមមិ េូវម្ល “សារៈេំខា ់” សៅក្នុខ អេេ ិខក្ម្លាខ ំ ព្យលក្មម ម្ល ការសក្ើ ស ើខ។ បម្លាេ់បដូរស េះប្ត្ូវបា ជប្មុញ សដ្ឋយសារស្ត្ក្ំសណើ ក្នុខវិេ័យក្េិក្មម ស្ លោំប្ទ ល់ជ ប្ក្ីប្ក្។ ំ ព្យលក្មមេរុបស ើយ ប៉ាុស្ ដម្ល “អលិត្ភាព្យ” ោខមុ សោលគឺអលិត្ភាព្យរបេ់ ីក្េិក្មម ិខក្ម្លាខ ក្ំសណើ វិេ័យក្េិក្មម សៅក្នខ ុ ប្បសទេក្មពុោម្ល ក្ប្មត្ ិ ខ្ពេ។ ំ ០០៤ ល់២០១២ ក្ំសណើ ប្បចំឆ្ននំ អលិត្ក្មមក្េិក្មមេរុបម្ល ៨,៧ភាគរយ។ ត្នមាបស្ ថមរបេ់វិេ័យក្េិក្មម5 ម្ល ការសក្ើ ៥,៣ភាគរយ សៅក្នុខរយៈសព្យលស េះ។ ់ ចសនាាេះព្យីឆ្ន២ ក្ំសណើ ព្យិសេេ ស្ លេថិត្សៅក្នុខចំសដមក្ំសណើ ខ្ពេ់ោខសគ សៅក្នុខព្យិភព្យសលាក្ស េះ ប្ត្ូវបា ជប្មុញសដ្ឋយអលិត្ក្មម ំដំ ោព្យិសេេប្េូវ (ក្ំសណើ ប្បចំឆ្នំ ៩ ភាគរយ) ិខសោត្ (២០ ភាគរយ) ំ ូខមី (៥១ ភាគរយ) អំសៅ (២២ ភាគរយ) ិខបស្ ា (១០ ភាគរយ)។ ក្ំសណើ សៅក្នុខវិេ័យបេុេត្ែ ិខស សាទ ម្ល ក្ប្មិត្មធ្យម។ ុ វិេយ ក្ំសណើ ក្នខ ំ ០០៧ មក្សៅ ២១ភាគរយ សៅក្នុខឆ្ន២ ័ ក្េិក្មមស េះ ក្៏ោំប្ទ ល់ប្បោព្យលរ ាប្ក្ប្ី ក្អខស្ រ។ អប្តាន ភាព្យប្ក្ីប្ក្របេ់ប្បសទេ ក្មពុោ ម្ល ការថយចុេះព្យី ៥០ភាគរយ សៅក្នុខឆ្ន២ ំ ០១១ សដ្ឋយចំ ួ ប្បោជ ប្ក្ីប្ក្ថយចុេះព្យី ៧លា នាក្់ មក្សៅ ៣លា នាក្់។ ការកាត្់ប ថយភាព្យប្ក្ីប្ក្ភាគសប្ចើ សក្ើត្ម្ល ស ើខសៅតាមត្ំប ់ជ បទ។ ការកាត្់ប ថយ ភាព្យប្ក្ីប្ក្ោខ ៦០ភាគរយ សក្ើត្ស ើខសដ្ឋយសារវិេ័យក្េិក្មម៖ ត្នមាអខករខ្ពេ់ោខមុ បា ជប្មុញឱ្យអលិត្ក្មមប្េូវ សក្ើ ស ើខ ស្ លជួយបសខកើ ប្បាក្់ចំណូល ល់អនក្ស្ លសធ្ែើស្ប្េ ិខក្េិដ្ឋា ប្េូវនានាស្ រ។ ការកាត្់ប ថយភាព្យប្ក្ីប្ក្បស្ ថមសទៀត្ សៅប ដព្យឹខអាប្េ័យសលើសោគជ័យន វិេ័យក្េិក្មមសប្ចើ ឆ្នំសៅមុខ្សទៀត្ សដ្ឋយសារស្ត្ត្ួនាទី ៏ធ្ំរបេ់វិេ័យស េះ ចំសោេះ ក្ម្លាខ ំ ព្យលក្មម ត្នមាបស្ ថម ិខការនាំសចញ ក្៏ ូចោសដ្ឋយសារស្ត្ក្េិក្រោសប្ចើ េថិត្សៅក្នុខចំសដមអនក្ស្ លប្ក្ីប្ក្ ិខ ងាយរខសប្ោេះោខសគ។ នួ ក្នខ ប្បាក្់ឈល ័ ក្េិក្មម សក្ើ ស ខ ុ វិេយ ើ សៅសក្ៀក្ ខ ឹ ប្បាក្់ឈល នួ ក្នខ ុ វិេ័យមិ ស្ម ក្េិក្មម។ ប្បាក្់ឈនួលក្នុខ វិេ័យក្េិក្មមម្ល ការសក្ើ ស ើខ ២០៦ភាគរយ សៅក្នុខការអសខកត្ស្ លសធ្ែើស ើខចសនាាេះ ឆ្ន២ ំ ០០៥ ខ ិ ២០១៣ ខ្ណៈសព្យល ុ ឆ្ន២ ស្ លប្បាក្់ឈនួលក្នុខវិេ័យមិ ស្ម ក្េិក្មម ម្ល ការសក្ើ ស ើខ ៦០ភាគរយ។ ូសចនេះ អលសធ្ៀបរវាខត្នមាបស្ ថម េប្ម្លប់អនក្ សធ្ែើការក្នុខវិេ័យមិ ស្ម ក្េិក្មមម្លនក្់ សធ្ៀបោមួយ ឹខអនក្សធ្ែើរបរក្េិក្មម គិត្ោត្នមាបចចុបប ន ម្ល ការថយចុេះព្យី ៣,២ សៅក្នខ ំ ០០៤ មក្ ២,១ ំ ០១២។ ចំណូលស្ លបា មក្ព្យីការសធ្ែើការងារសៅក្នុខក្េិដ្ឋា ម្ល ការសក្ើ ស ើខ។ សៅក្នុខឆ្ន២ 5 ត្នមាបស្ ថមក្នុខវិេ័យក្េិក្មម ប្ត្ូវបា សគក្ំណត្់ ិយម ័យឱ្យថាោត្នមាេរុបរបេ់អលិត្ក្មម ក្ត្នមាធាត្ុចូលក្េិក្មមសចញ។ xviii ិ ភាព្យអលិត្ក្មម ំដបា ព្យិព្យធ្ ើ ខ ំ ចប់សអដម ុប។ ិ ក្ំព្យខ ត សបើសទាេះបីោអលិត្ក្មម ំដំ សផ្ទដត្ោចមបខសលើ ំដប្ំ េូវក្តី មុខ្ ំ ំព្យុខម្ល ការស្ប្បប្បួល សដ្ឋយសារស្ត្លទធភាព្យចំសណញ ខ្ពេ់ន ប្បសភទ ំដសំ ប្ៅព្យីប្េូវ។ សៅឆ្ន២ ំដក្ ំ ០១៣ អលចំណូល េរុបគិត្ោមធ្យម (រួមទាំខក្ំនរព្យលក្មម) របេ់ក្េិដ្ឋា រួមម្ល ៥០៦ ុលាារ/េិចតា (៩,៤ ុលាារ/នថៃ) េប្ម្លប់ ំ ូខមី, ៣០៣ ុលាារ/េិចតា (៨,៨ ុលាារ/នថៃ) េប្ម្លប់សោត្ ិខ ១៣៩៣ ុលាារ/េិចតា (៧,២ ុលាារ/នថៃ) េប្ម្លប់ ំដប ំ ស្ ា សបើសធ្ៀប ោមួយ ឹខ ២៤៥ ុលាារ /េិចតា (៤,៦ ុលាារ/នថៃ) េប្ម្លប់ប្េូវ វេា ិខ ២៩៥ ុលាារ/េិចតា (៩,៦ ុលាារ/នថៃ) េប្ម្លប់ប្េូវប្បាខ ំ ។ ប្េូវរ ូវប្បាខ ំ ប្ត្ូវប្បស្ជខ ោមួយ ឹខ ំដម ំ ិ ស្ម ប្េូវ នទសទៀត្ សបើគិត្ព្យីចំណូលរបេ់ក្ម្លាខ ំ ប្ត្ូវជួបឧបេគា ំ ព្យលក្មម ប៉ាុស្ ដការព្យប្ខីក្អលិត្ក្មមប្េូវរ ូវប្បាខ សដ្ឋយសារស្ត្ ប្បព្យ័ ធធារាសាស្ដេដសៅម្ល ក្ប្មិត្សៅស ើយ។ ភាគរយន នអា ីេរុបស្ លសប្បើប្បាេ់ េប្ម្លប់បខកបសខកើ អលប្េូវ ម្ល ការ ថយចុេះព្យី ៨៦ភាគរយ សៅក្នុខឆ្ន២ ំ ០០២ រេូត្ ល់ ៧៤ភាគរយ សៅក្នុខឆ្នំ២០១១ ខ្ណៈសព្យលស្ លភាគរយន នអា ីេប្ម្លប់ដ្ឋសំ ោត្ ិខ ំ ូខមី ម្ល ការសក្ើ ស ើខគួរឱ្យក្ត្់េម្លាល់។ ើ សធ្ែព្យ េូមបីស្ត្វិេ័យប្េូវក្៏ចប់សអដម ិ ធ្ ើ ព្យ ិ ក្មមអខស្ រ។ សដ្ឋយសារស្ត្ម្ល ត្ប្មូវការព្យីសរាខម្ល៉ាេុី ក្ិ ប្េូវទំស ើប ក្េិក្រ កា ់ស្ត្សប្ចើ ោខមុ នាំោនបសខកើ ការអលិត្ប្េូវប្ក្អូបស្ លអដល់ប្បាក្់ចំសណញខ្ពេ់ោខមុ ប្បម្លណោ១០ភាគរយ ន នអា ី ដ្ឋំប្េូវេរុប ិខប្បម្លណោ ៣០ភាគរយន អលិត្ក្មមេរុប។ ការព្យប្ខីក្ ីស្ប្េបស្ ថមសទៀត្ េប្ម្លប់ប្េូវប្ក្អូបស េះ គឺអាច សធ្ែើសៅបា ោមួយ ឹខការស្ក្លំអការអាត្់អាខ់ប្ោប់ព្យូជប្បក្បសដ្ឋយគុណភាព្យ ការអសព្យែអាយក្េិក្មម ិខប្បព្យ័ ធធារាសាស្ដេដ។ ន លម្ល ការសក្ើ ស ខ ទិ អ ំ ភា ើ េប្ម្លប់ ដ ំ គសប្ចើ ។ ោមួយ ឹខក្ំសណើ ប្បចំឆ្ននំ នអា ីដ្ឋំ ុេះ ៤,៧ភាគរយ ំ ០០៤ ិខអលិត្ក្មមក្េិក្មមេរុប ៨,៧ភាគរយ ទិ ន ័យម្ល ក្ំសណើ ោមធ្យមប្បចំឆ្នំ ៤ភាគរយ ចសនាាេះព្យីឆ្ន២ ល់២០១២។ ុ ក្ំសណើ ទិ នអលទាំខស េះប្ត្ូវបា ជប្មញ សដ្ឋយការសប្បើប្បាេ់បសចចក្វិទាថមីៗសប្ចើ ោខមុ ការព្យប្ខីក្ប្បព្យ័ ធធារាសាស្ដេដ (សបើសទាេះបីោសៅម្ល ក្ប្មិត្សៅស ើយក្៏សដ្ឋយ) ការសប្បើប្បាេ់ព្យូជ ិខជីរក្េិក្មមលែៗកា ់ស្ត្សប្ចើ ោខមុ ិខលទធភាព្យ ទទួលបា សេវាសប្គឿខយ ដលែប្បសេើរោខមុ ស្ លទាំខស េះេុទធស្ត្បងាាញព្យីភាព្យសជឿ សលឿ ន ោណិជផូ ីយក្មម។ ក្េិក្រ ក្៏ម្ល លទធភាព្យទទួលបា ទីអារកា ់ស្ត្ប្បសេើរោខមុ អខស្ រ។ ត្នមាអលិត្អលក្េិក្មមរបេ់ក្មពោ ុ សៅម្ល ភាព្យប្បក្ួត្ប្បស្ជខសៅ ខ ឹ ក្េិដ្ឋា ផ្ទាល់។ ការវិភាគសលើភាព្យប្បក្ួត្ ប្បស្ជខន ត្នមាអលិត្អលក្នុខប្េុក្ ំ ព្យលក្មម បងាាញថាសបើសទាេះបីោត្នមាក្ម្លាខ ិខត្នមាធាត្ុចូលស្ លសប្បើប្បាេ់េប្ម្លប់ដ្ឋំ ុេះ ម្ល ការសក្ើ ស ើខយា៉ា ខដក្តី ត្នមាបស្ ថមស្ លគិត្សដ្ឋយក្េិក្រសៅខ្ពេ់ោខត្នមា ន ក្តាដអលិត្ក្មមក្នុខប្េុក្ ( ី ព្យលក្មម ិខស ើមទុ )។ ភាព្យប្បក្ួត្ប្បស្ជខខ្ពេ់ស េះ អាចោមូលសេត្ុនាឱ្ ំ យម្ល ការសក្ើ ស ើខការនាំសចញក្េិអល ៏សប្ចើ សៅក្នុខ ទេវត្សរ៍ក្ ាខសៅស េះ។ ប៉ាុស្ ដភាព្យ ប្បក្ួត្ប្បស្ជខន ប្េូវធ្មមតាស្ លអលិត្សៅរ ូវវេា ម្ល ការថយចុេះក្នុខរយៈប៉ាុនាម ឆ្នំ ចុខសប្កាយស េះ សេើយ ក្េិក្រខានត្ត្ូចោសប្ចើ ស្ លអលិត្ប្េូវធ្មមតា បា ចប់សអដើមបាត្់បខ់ភាព្យប្បក្ួត្ប្បស្ជខរបេ់ខ្ាួ ។ ក្ំសណើ សៅក្នខ ុ វិេយ័ ក្េិក្មមក្ ខ ា មក្ ប្ត្វូ បា ជប្មញ ុ សដ្ឋយក្តាដោសប្ចើ ។ ក្នខ ុ ចំសដមក្តាដធ្ំៗ រួមម្ល ការ សបើក្ចំេរការសធ្ែើោណិជផក្មម ិខសោល សយាបាយក្េិក្មមសផ្ទដត្សលើទីអារោទូសៅ។ ប្បសទេក្មពុោ ោប្បសទេមួយ ក្នុខ ចំសដមប្បសទេក្ំព្យុខអភិវឌ្ឍនានា ស្ លមិ បា ប្បត្ិក្មមប្ជុលខាាខ ំ សព្យក្សៅ ឹខការសលាត្ ស ើខ ូវត្នមាសេបៀខអាហារ ោេក្លសៅក្នុខឆ្នំ២០០៨ ប៉ាុស្ ដចត្់ទុក្ការស ើខនថាសេបៀខអាហារស េះ ថាោឱ្កាេមួយស ើមបីទាញប្បសយាជ ៍ េប្ម្លប់ ក្ំសណើ វិេ័យក្េិក្មម។ ប្បសទេនានាស្ លម្ល ការនាំសចញសប្ចើ ោខការនាំចូល ូចោឥដ ា ិខសវៀត្ដម ោស ើម បា ដ្ឋក្់ក្ំេិត្សលើការនាំសចញ ស ើមបីទប់ឥទធិព្យលន ការស ើខនថាសេបៀខអាហារោេក្ល មក្សលើទីអាររបេ់ព្យួក្សគ។ ត្នមា អលក្េិក្មមខ្ពេ់សៅក្នុខប្បសទេក្មពុោ បា សធ្ែើឱ្យការព្យប្ខីក្ ីក្េិក្មម អដលអ ់ លចំណូលមក្វិញខ្ពេ់។ សលើេព្យីស េះ វិេ័យក្េិក្មមបា ទទួលប្បសយាជ ៍ព្យី (i) ការទទួលបា ទីអារសៅបរសទេសប្ចើ ោខមុ តាមរយៈក្ិចចប្ព្យមសប្ព្យៀខទំ ិញប្គប់ស្បបយា៉ា ខសលើក្ស្លខស្ត្អាវុធ្សចញ ោមួយ ឹខេេភាព្យអឺរុប ិខការសបើក្ចំេោណិជផក្មមឆ្ាខស្ ោមួយ ឹខប្បសទេ ជិត្ខាខ (ii)ការទទួលបា ិខការសប្បើប្បាេ់យា៉ាខទូលាយ ូវសេវាសប្គឿខយ ដ ជំ ួេសអាយការចំដយកា ់ស្ត្ខ្ពេ់សលើក្ម្លាខ ំ ព្យលក្មមសៅជ បទ (iii)ការទទួលបា ឥណទា ប្បសេើរោខមុ ិខ (iv)ការ វិ ិសយាគរបេ់វិេ័យឯក្ជ សលើម្ល៉ាេុី ក្ិ ប្េូវខានត្ធ្ំ។ បញ្ហាប្បឈមនានា ភាគរយ ខ្ ៏េ ព ់ន ក្ំសណើ ក្នខ ុ វិេយ ័ ក្េិក្មមក្ ខ ុ សដ្ឋយសារស្ត្ការព្យប្ខក្ ា សៅស េះ ប្ត្វូ បា ជប្មញ ុ ។ ការរួមចំស្ណក្ោមធ្យម ន ការព្យប្ខីក្នអា ីដ្ឋំ ុេះ ចំសោេះការស្ប្បប្បួលអប្តាចំណូលេរុបរបេ់ក្េិក្មម គិត្ោត្នមាព្យិត្ ចសនាាេះព្យីឆ្ន២ ី នអា ីដ្ឋំ េះ ំ ០០៥ ល់២០១៣ គឺប្បស្េលោ ៦០ភាគរយ។ ីក្េិក្មម ប្ត្ូវបា ព្យប្ខីក្ប្បម្លណោ ៤,៧ភាគរយ ក្នុខមួយឆ្នំ សដ្ឋយម្ល ការសក្ើ ស ើខយា៉ា ខសប្ចើ (១២៨ភាគរយ) េប្ម្លប់ ំ ូខមី។ ការព្យប្ខីក្ ីក្េិក្មមស្បបស េះ បា រួមចំស្ណក្ សធ្ែើឱ្យម្ល ការបាត្់បខ់នប្ព្យសឈើកា ់ស្ត្សលឿ ោព្យិសេេ សៅត្ំប ់ខ្ពខ់រាប។ ក្ំសណើ ោមធ្យមន អប្តាចំសណញក្នខ ិ លែប្បសេើរ គឺ៣,៤ភាគរយក្នខ ុ មួយេិចតា6 ម្ល ក្ប្មត្ ំ ប៉ាុស្ ដ វាម្ល ការស្ប្បប្បួលចសនាាេះព្យី ២,១ភាគរយ េប្ម្លប់ប្េូវរ ូវប្បាខ ុ មួយឆ្ន។ ំ សៅ៤៤,៥ភាគរយ េប្ម្លប់បស្ ា។ អប្តា ចំសណញេប្ម្លប់ សោត្ក្នុខមួយេិចតាម្ល ការថយចុេះ។ សៅក្នុខក្រណីោ សប្ចើ ក្េិក្រស្ លបា ព្យប្ខីក្នអា ីរបេ់ខ្ាួ បា ទទួលប្បាក្់ចំណូលខ្ពេ់ ោខមុ ប៉ាុស្ ដក្េិក្រស្ លរក្ានអា ីដ្ឋំ ុេះ ស្ លមិ អាចបសខកើ ប្បាក្់ចំណូលរបេ់ខ្ាួ ឱ្យបា សប្ចើ ស ើយ។ ក្ំ ុខសព្យលស្ ល ត្នមាសេបៀខស ើខខ្ពេ់ ប្ត្វូ បា ជំរុញសលើការព្យប្ខីក្នអា ី ោោខការក្សាខមូលដ្ឋា រឹខម្លំ Changes in farm gross margins, real terms, 2005-2013, % តាមរយៈការបសខកើ អលិត្ភាព្យ។ Change in Cultivated Change in Gross Margin Change in Total សព្យលស្ លភាព្យប្ក្ប្ី ក្ម្ល ការថយចុេះគួរឱ្យក្ត្់េម្លាល់ ចំ ួ ប្បោជ ស្ លងាយរខសប្ោេះក្៏ម្ល ការសក្ើ ស ខ ើ សប្ចើ អខស្ រ។ Area per Hectare Gross Margin Wet season rice 2.1 2.4 4.5 ភាគសប្ចើ ន ប្បោជ ស្ លសខើបសចញព្យីភាព្យប្ក្ីប្ក្ សៅមិ ទា ់សៅអុត្ឆ្ៃយព្យីវាសៅស ើយ។ ការបាត្់បខ់ ចំណូលប្ត្ឹមស្ត្១២០០សរៀល ក្នុខមួយនថៃ (ត្នមាទឹក្េុទធព្យីរ បត្ូច) Dry season rice 6.7 2.1 8.8 ឹខសធ្ែើឱ្យអប្តា ន ភាព្យប្ក្ីប្ក្របេ់ប្បសទេក្មពុោ សក្ើ ស ើខ សទែ ខ ល់៤០ភាគរយ។ អប្តាន ភាព្យងាយរខសប្ោេះ ៏ខ្ពេ់ស្បបស េះ គឺោេញ្ហាស្ លបងាាញថា Cassava 128.1 15.1 143.2 ក្ំសណើ អលិត្ភាព្យក្េិក្មម សៅម្ល ក្ប្មិត្ត្ិចត្ួចសៅស ើយ។ េ ដិេុខ្អាហារូបត្ថមភ ក្៏សៅម្ល សលបឿ យឺត្ោខការកាត្់ប ថយភាព្យប្ក្ីប្ក្អខស្ រ៖ សទាេះបី Maize 17.2 -10.3 6.9 Vegetables 6.4 44.5 50.9 ោសប្បវា ខ់ន ភាព្យប្ក្ិ របេ់ក្ុម្លរអាយុសប្កាម៥ឆ្នំ បា ថយចុេះ៧ភាគរយ សៅចសនាាេះព្យីឆ្ន២ ំ ០១០ ិខ២០១៤ក្តី អប្តា ស េះសៅស្ត្ម្ល ក្ប្មិត្ខ្ពេ់ (៣៣ ភាគរយ) Weighted average 4.7 3.4 8.1 សៅស ើយ។ Source: 2005 and 2013 surveys. 6 អប្តាចំណូលេរុប ោចំណូលេរុប ក្ធាត្ុចូលប្បសយាលសចញ រាប់ទាំខព្យលក្មមស្ លប្ត្ូវបា ជួលអខ។ អប្តាចំសណញរបេ់ក្េិដ្ឋា ស្ លប្ត្ូវបា បា៉ា ់ប្បម្លណសៅក្នុខរបាយការណ៍ស េះ រាប់បញ្ចូលស្ត្ ំដំេំខា ់ៗ ូចោ ប្េូវ ំ ូខមី សោត្ ិខបស្ ា ោស ើម។ xix ប្បោជ ប្ក្ីប្ក្ ស្ លក្ំព្យុខរេ់សៅសប្កាម ១,១៥ ុលាារ ក្នុខមួយនថៃ (បនាាត្់ន ភាព្យប្ក្ីប្ក្) ប្បោជ ស្ លសក្ៀក្ប្ក្ ឬងាយរខសប្ោេះ គឺោអនក្ស្ លរេ់ សៅសលើ ១,១៥ ុលាារ ក្នុខមួយនថៃ ប៉ាុស្ ដសប្កាម ២,៣០ ុលាារ ក្នុខមួយនថៃ ិ ខ្ពេោ ភាព្យងាយរខសប្ោេះម្ល ក្ប្មត្ ុ ចំសដមក្េិដ្ឋា ស្ លម្ល ទំេត្ ់ ខ ក្នខ ំច ូ ៗ។ ក្េិដ្ឋា ត្ូចៗ ស្ លម្ល ទំេំ ី ត្ិចោខ១េិចតា ប្ត្ូវបា រក្ស ើញថាជួប ឹខការលំបាក្ ស ើមបីព្យប្ខីក្ខ្ាួ ិខសធ្ែើេម្លេរណក្មម ចូលសៅក្នុខស្ខ្សេងាែក្់ ំ ០១១។ សៅក្នុខទេវត្សរ៍ចុខសប្កាយស េះ សៅក្នុខប្បសទេក្មពុោ ម្ល ិនានការស្ លក្េិដ្ឋា ធ្ំៗ (សលើេ៣េិចតា)កា ់ស្ត្រីក្ធ្ំសៅៗ អាត្់អាខ់អាហារស្ លម្ល លក្េណៈទំស ើប។ ន ក្េិដ្ឋា ទាំខស េះ គិត្ោចំ ួ ប្គួសារេរុបស្ លម្ល ីោក្មមេិទធិផ្ទាល់ខ្ាួ គឺ ៤៨ភាគរយ សៅក្នុខឆ្ន២ សេើយក្េិដ្ឋា ត្ូចៗ (ត្ិចោខ១េិចតា)កា ់ស្ត្រួញត្ូចសៅៗ។7 ទំេំមធ្យមរបេ់ក្េិដ្ឋា ខានត្ត្ូច8 បា ថយចុេះព្យី០,៩៩េិចតាក្នុខឆ្ន២ ំ ០១២ សេើយក្នុខសព្យលោមួយោន សនាេះ ទំេំ ោមធ្យមរបេ់ក្េិដ្ឋា ខានត្មធ្យម (ចសនាាេះព្យី១ ំ ០០៨ មក្សៅ០,៨៨េិចតាក្នុខឆ្ន២ សៅ៣េិចតា) បា សក្ើ ស ើខព្យី១,៥៥េិចតា ល់២,៣៨េិចតា សេើយក្េិដ្ឋា ធ្ំៗ (សលើេព្យី ៣េិចតា) បា សក្ើ ស ើខព្យី ៣,៦១េិចតា ល់៧,០៣េិចតា។ សបើសទាេះបីោអលិត្ភាព្យរបេ់ក្េិដ្ឋា ត្ូចៗម្ល ការសក្ើ ស ើខខ្ាេះក្តី ក្៏សៅម្ល ក្ំរិត្ទាប សដ្ឋយភាគសប្ចើ ន ក្េិដ្ឋា ត្ូចសៅមិ ទា ់បា ទទួល សេវាអសព្យែអាយ ក្េិក្មម ិខសេវាសាធារណៈ នទសទៀត្ ក្នុខប្ទខ់ប្ទាយធ្ំស ើយ។ ការសក្ើ ស ើខ ូវប្បាក្់ចំណូលរបេ់ក្េិក្រក្នុខប្ក្ុមស េះ សៅក្នុខទេវត្សរ៍ចុខសប្កាយស េះ ហាក្់បី ូចោសក្ើត្ស ើខសដ្ឋយសារស្ត្ត្នមាអលិត្ក្មមខ្ពេ់ោខមុ ិខការលក្់ព្យលក្មមសអាយ ក្េិដ្ឋា ធ្ំៗបា ប្បាក្់ឈនួលខ្ពេ់ោខមុ ោោខសដ្ឋយសារការសក្ើ ស ើខន អលិត្ភាព្យ ី ិខគុណភាព្យព្យលក្មម។ ិ ត្ួច សៅក្នខ សលើក្ស្លខស្ត្អខករសចញ ឧេាេក្មមស្ក្នចនក្េិអលបា បំសព្យញត្ួនាទីត្ច ុ ក្ំសណើ វិេ័យក្េិក្មម។ ក្េិអលសេាើរស្ត្ទាំខអេ់ប្ត្ូវបា នាំសចញសៅប្បសទេជិត្ខាខ សដ្ឋយព្យុំម្ល ការស្ក្នចន។ ស េះបងាាញព្យីចំណុចសខ្ាយធ្ៃ ់ធ្ៃរ សៅក្នុខស្ខ្សេងាែក្់ត្ំនល ោព្យិសេេសៅក្នុខប្បព្យ័ ធប្គប់ប្គខស្ខ្សេងាែក្់អាត្់អាខ់សប្កាយប្បមូលអល (ការប្បមូលក្េិអល មិ ទា ់ស្ក្នចន ការរក្ាទុក្ ការអដល់េិរញ្ាបបទា ភេដុភារ ការ ឹក្ជញ្ផូ ិខព្យ័ត្៌ម្ល )។ អែស្ី លម្ល ប្បេិទភា ា មក្ ខ ធ ព្យក្ ខ ើ ឹ មិ ប្គប់ប្ោ ់ េប្ម្លប់អនាគត្ស យ ោមួយ ខ ឹ ត្នមាសេបៀខអាហារម្ល ការថយចុេះ សេើយនអា ីដ្ឋំ េះុ ស្លខសក្ើ ស ខ ើ វិេ័យក្េិក្មមសៅក្មពោ ុ ក្ំព្យខ ុ ក្ំសណើ ធ្ំៗរបេ់ខ្ាួ ចំ ួ ព្យីរ។ សលើេព្យីស េះ ក្ម្លាខ ុ បាត្់បខ់ ក្តាដជប្មញ ំ ព្យលក្មមក្នុខវិេ័យក្េិក្មមកា ់ស្ត្ខ្សត្់សៅៗ ិខម្ល ត្នមាកា ់ស្ត្ខ្ពេ់ ស្ លនាំឱ្យចំដយនថាស ើមអលិត្ក្មមកា ់ស្ត្សក្ើ ស ើខ។ ត្នមាក្េិអលោេក្ល ប្ត្ូវបា សគព្យាក្រថា ឹខ ប ដថយចុេះសៅក្នុខទេវត្សរ៍ខាខមុខ្ស េះ សេើយការលក្់អខករធ្មមតាស្ លម្ល ត្នមាទាប សៅក្នុខទីអារក្នុខប្បសទេ ិខអ ដរោត្ិ ឹខទទួលបា ចំសណញត្ិចោខមុ ។ ការព្យឹខអាប្េ័យសលើត្ប្មូវការក្នុខប្េុក្ស្ លសក្ើ ស ើខ សដ្ឋយសារស្ត្ក្ំសណើ ន អេេ ិខក្ំសណើ ត្ប្មូវការន ការនាំចូល សដ្ឋយមិ ម្ល ការសប្បើប្បាេ់ធ្ ធា ឱ្យកា ់ស្ត្ម្ល ប្បេិទធភាព្យសទសនាេះ ឹខមិ ប្គប់ប្ោ ់ ស ើមបីរក្ាក្ំសណើ ខ្ពេ់ ូចសព្យលក្ ាខស ើយ។ ំ ចលក្រក្ំសណើ សអសខសទៀត្ ស ើមបី ជួយរក្ាក្ំសណើ ក្នុខវិេ័យក្េិក្មមឱ្យបា សេមើ ឬសក្ៀក្ ឹខ៥ភាគរយ ិខសធ្ែើឱ្យក្ំសណើ ស េះម្ល ិរ ដរភាព្យបស្ ថមសទៀត្។ ប្បសទេក្មពុោចំបាច់ប្ត្ូវស្ត្ស្េែខរក្ ក្ម្លាខ ុ វិេយ ការថមថយ ូវក្ំសណើ ក្នខ ុ កាលព្យីសព្យលថមៗ ័ ក្េិក្មមរបេ់ក្មពោ ី ស េះ អាចោេញ្ហាបងាាញព្យីការថយចុេះ ស្ ល អាចសក្ើត្ស ខ ុ រយៈសព្យលស្វខ។ ខ្ណៈសព្យលស្ លត្នមាសេបៀខអាហារម្ល ការថយចុេះោប ប ើ ក្នខ ដ នាាប់ សេើយទីអារអខករ េក្សលាក្ កា ់ស្ត្ម្ល ការប្បក្ួត្ប្បស្ជខោមួយ ឹខការចូលមក្ក្នុខទីអារសារោថមីរបេ់ប្បសទេនថ ិខមីយា៉ា ់ម្ល៉ា អលិត្ក្មម ប្េូវសៅក្នុខប្បសទេក្មពុោសៅប្ទឹខ។ ោមួយោន ស េះ ត្នមាបស្ ថមេរុបរបេ់ ំដម្ល ំ ការថយចុេះព្យី ៤,៩ភាគរយ សៅក្នុខឆ្នំ ២០១២ មក្០,៦ភាគរយ សៅក្នុខឆ្ន២ំ ០១៣។ ក្ំសណើ ន ត្នមាបស្ ថមរបេ់វិេ័យក្េិក្មមបា ថយមក្សៅ ១,៦ភាគរយ សៅក្នុខឆ្នំ ២០១៣ សប្កាយសព្យលស្ លម្ល ក្ំសណើ ោមធ្យម ៥,៣ ភាគរយ ចសនាាេះព្យីឆ្នំ ២០០៤ ល់ ២០១២។ ិ ខ្ពេ់ ម្ល សារៈេំខា យា ការវិលប្ត្លប់សៅរក្អប្តាក្ំសណើ ក្េិក្មមក្ប្មត្ ំ ។ វិេ័យក្េិក្មម សៅស្ត្រក្ា ចំស្ណក្មួយ ៏ធ្ំន អេេ ោណិជផក្មម ិខក្ម្លាខ ់ ៉ា ខខាាខ ំ ព្យលក្មម ស្ លម្ល ័យថា វិេ័យស េះសៅស្ត្ម្ល សារៈេំខា ់ចំសោេះ ការកាត្់ប ថយភាព្យប្ក្ីប្ក្ ិខភាព្យងាយរខសប្ោេះ ក្នុខទេវត្សរ៍ខាខមុខ្ស េះ។ ប្បេិ សបើបស្ប្មបប្មួលក្នុខប្បសទេក្មពុោ សៅស្ត្ប ដម្ល សទៀត្ ោមួយ ឹខក្ំសណើ ោមធ្យមក្នុខវិេ័យក្េិក្មម៥ភាគរយ ប្ត្ឹមឆ្ន២ ំ ០៣០ ប្បសទេក្មពុោ ឹខម្ល វិេ័យក្េិក្មម ស្ លម្ល ទំេំត្ូចោខស េះ ប៉ាុស្ ដម្ល អលិត្ភាព្យ ខ្ពេ់ោខមុ ។ ចំស្ណក្ន វិេ័យក្េិក្មម សៅក្នុខ អេេ ិខ ក្ម្លាខ ំ ព្យលក្មមេរុប ប្ត្ូវបា ព្យាក្រថាថយចុេះមក្សៅប្ត្ឹម ១៧ភាគរយ ិខ ៣១ភាគរយ សៅក្នុខឆ្នំ ២០៣០ ព្យី ២៦ភាគរយ ិខ ៥១ភាគរយ សៅក្នុខឆ្ន២ ំ ០១២ ខ្ណៈសព្យលស្ លអលិត្ភាព្យ ីធ្ាី ឹខម្ល ការសក្ើ ស ើខព្យី ១៣០០ ុលាារ/េិចតា ល់ ២៧០០ ុលាារ/េិចតា សេើយអលិត្ភាព្យព្យលក្មម ឹខម្ល ការសក្ើ ស ើខព្យី ១២០ ុលាារ/នាក្់ ល់ ៣៧០០ ុលាារ/នាក្់ (តារាខ ទី ១)។ ុ សៅវិញ ការខាត្បខ់សដ្ឋយសារស្ត្ក្ំសណើ ក្េិក្មមម្ល សលបឿ យឺត្ ខ អាយ ំ ំសធ្ខ។ ប្បេិ សបើ ក្ំសណើ ោមធ្យមន វិេ័យក្េិក្មមម្ល ប្ត្ឹមស្ត្ ៣ភាគរយ ចសនាាេះព្យីឆ្ន២ ឹ ម្ល ទំេធ្ ំ ០១២ ល់២០៣០ សនាេះត្នមាបស្ ថម ន វិេ័យក្េិក្មម ខ ឹ ម្ល ការថយចុេះោខមុ ២៩ភាគរយ សៅឆ្ន២ ំ ០៣០ សបើសធ្ៀបោមួយ ឹខសេដរីយ៉ាូ ក្ំសណើ ៥ភាគរយ ស្ លសធ្ែើឱ្យ អេេ ោរួមថយចុេះប្ត្ឹម១៨ភាគរយ សេើយទ ាឹមោន សៅស្ត្រក្ាប្បោោជ សប្ចើ នាក្់សៅក្នុខវិេ័យក្េិក្មម អលិត្ភាព្យព្យលក្មម ំ។ ក្នុខវិេ័យក្េិក្មមថយចុេះ៣៤ភាគរយ ិខសធ្ែើឱ្យការខ្ិត្មក្ជិត្ោន ន ចំណូលរវាខក្េិក្រ ិខអនក្មិ ប្បក្បរបរក្េិក្មម កា ់ស្ត្ម្ល ភាព្យយឺត្យា៉ា វ។ ក្ំសណើ ក្នុខវិេ័យក្េិក្មមទាបោខមុ ឹខនាំឱ្យការកាត្់ប ថយភាព្យប្ក្ីប្ក្ម្ល សលបឿ យឺត្ខាាខ ូ នាក្រវិេយ តារាខទី ១ ៖ អលប៉ាេះោល់ន ក្ំសណើ ក្េិក្មមទាបមក្សលើេច ួ ចំ ួ ប្បសទេក្មពុោ ័ ក្េិក្មមមយ េូចនាក្រ ២០១២ ២០៣០ ៣% ន ក្ំសណើ អេេ ក្នខ ុ វិេ័យក្េិក្មម ៥% ន ក្ំសណើ អេេ ក្នខ ័ ក្េិក្មម ុ វិេយ េម្លេភាគន ក្េិក្មមសៅក្នុខ អេេ គិត្ោ % ២៦ ១៥ ១៧ េម្លេភាគន ក្េិក្មមសៅក្នុខក្ម្លាំខព្យលក្មម គិត្ោ % ៥១ ៣៤ ៣១ អលិត្ភាព្យក្ម្លាំខព្យលក្មមក្នុខវិេ័យក្េិក្មម គិត្ោ ុលាារ/នាក្់ ១២០០ ២៤៥០ ៣៧០០ ី េិក្មម គិត្ោ ុលាារ/េិចតា អលិត្ភាព្យ ក្ ១៣០០ ១៩០០ ២៧០០ អលសធ្ៀបន អលិត្ភាព្យក្ម្លាំខព្យលក្មមរវាខ អនក្សធ្ែើការងារ ក្នុខ វិេ័យក្េិក្មម ោមួយ ឹខអនក្សធ្ែើការងារក្នុខវិេ័យសអសខសទៀត្ ២,១ ២,១ ១,៦ 7 ការស្ប្បប្បួលរបាយ ីក្េិក្មមស េះ សក្ើត្ស ើខសដ្ឋយមិ ោក្់ព្យ័ ធោមួយ ឹខេមបទា ីសេ ាក្ិចចស ើយ។ 8 ការអសខកត្សលើក្េិដ្ឋា ប្ត្ូវបា សធ្ែើស ើខ សៅក្នុខឆ្នំ ២០១៣ េប្ម្លប់របាយការណ៍ស េះ។ េូមសមើលឧបេមព័ ធទី ២ េប្ម្លប់ព្យ័ត្៌ម្ល លំអិត្។ xx ័ រយៈសព្យលស្វខេប្ម្លប់វិេយ ចក្េវុ ិេយ ុ ចំបាច់ប្ត្វូ សកាដបសលើេម្លេធាត្ុន ិរ រដ ភាព្យ ទិ អ ័ ក្េិក្មមសៅក្មពោ ិ ក្ំសណើ ប្បាក្់ចណ ន ល ភាព្យប្បក្ួត្ប្បស្ជខ ខ ំ សព្យក្ស្ត្សលើអលិត្ក្មម។ ការគណនាបា៉ា ់សាម តាម សេដរីយ៉ាូសអសខៗ ំ ូ ល សដ្ឋយេំរាលព្យីការសផ្ទដត្ខាាខ េប្ម្លប់ការអភិវឌ្ឍវិេ័យក្េិក្មមនាសព្យលអនាគត្ បងាាញថាការសក្ើ ស ើខ ូវអលិត្ភាព្យក្េិក្មម នាសព្យល អនាគត្ ឹខម្ល អលោវិជផម្ល ខាាខ ំ មក្សលើចំណូលរបេ់ក្េិក្រ ោព្យិសេេ ប្បេិ សបើម្ល ការផ្ទាេ់បដូរព្យីការដ្ឋំ ុេះតាមស្បប ប្បនព្យណី សៅោការសប្បើប្បាេ់បសចចក្វិទាទំស ើប សដ្ឋយអមោមួយ ឹខការសក្ើ ស ើខ ូវប្បព្យ័ ធធារាសាស្ដេដ ិខការសប្បើប្បាេ់ េំភារៈក្េិក្មមថមីៗស្ លម្ល ប្បេិទធភាព្យខ្បេ់។ ក្េិក្រស្ លដ្ឋំប្េូវប្ក្អូប ស្ក្នចន ំ ូខមីោចំណិត្ ំ ូខមីសប្ក្ៀម ិខសធ្ែើ ការបស្ ថមត្នមាសអសខសទៀត្សលើក្េិអល ឹខអាចទទួលបា ចំណូលខ្ពេ់ោខមុ ិខបងាាញចាេ់ព្យីសារៈេំខា ់ន ការស្ក្នចន ក្េិអល។ ការកាត្់ប ថយចំដយអលិត្ក្មម តាមរយៈការសប្បើប្បាេ់ឱ្យបា ប្បសេើរោខមុ ូវធ្ ធា ស្ លម្ល ប្សាប់ (តាមរយៈការបសខកើ ទិ នអល) ិខការកាត្់ប ថយការធាាក្់ចុេះត្នមាក្េិអល តាមរយៈការកាត្់ប ថយត្នមា ឹក្ជញ្ជូ គឺោគ ាឹេះ េប្ម្លប់រក្ាភាព្យប្បក្ួត្ប្បស្ជខន ក្េិអលនានា។ ការប ដព្យប្ខីក្នអា ី ក្៏អដល់ ូវចំណូលបស្ ថមអខស្ រ ប៉ាុស្ ដភាព្យក្ំណត្់ន ធ្ ធា ី ឹខកាត្់ប ថយការព្យប្ខីក្ ីក្នុខប្ទខ់ប្ទាយធ្ំនាសព្យលអនាគត្។ បញ្ហាប្បឈមស េះ សធ្ែើឱ្យការបសខកើ ទិ នអលក្េិក្មម ោណិជផូប ីយក្មម ិខ ំ ស ើខ ក្នុខការធានាឱ្យម្ល ក្ំសណើ ក្េិក្មម កាត្់ប ថយភាព្យ ប្ក្ីប្ក្ ិខសលើក្ក្មពេ់វិបុលភាព្យរួម សៅក្នុខប្បសទេក្មពុោ នាសព្យលអនាគត្។ ការសធ្ែើព្យិព្យិធ្ក្មម រឹត្ស្ត្ម្ល សារៈេំខា ់ខាាខ អនុសាសន៍ ដ សំ ណើ ក្េិក្មមយា៉ាខសលឿ ការប ក្ ិ ការផ្ទាេ់បដូររចនាេមព័ នធ វិេ័យក្េិក្មមបស្ ម ខ ថ សទៀត្ គឺអាចសធ្ែសើ ៅបា ុ ។ េំរាប់ប្បសទេក្មពោ េំណុំសោល សយាបាយរួមោន ចំ ួ ៤ អាចជួយអដល់ការោំប្ទ ល់ការប ដក្ំសណើ ក្េិក្មមស្បបស េះ សៅក្នុខរយៈសព្យល៥ឆ្នខា ំ ខមុខ្ (រយៈសព្យលខ្ាី ិខមធ្យម)។ ទី១ គឺប្ត្ូវរក្ាបរិយាកាេសោល សយាបាយក្េិក្មម ស្ លបខកភាព្យ ងាយប្េួល ល់វិេ័យឯក្ជ សដ្ឋយសផ្ទដត្ការយក្ចិត្ដទុក្ដ្ឋក្់បស្ ថមសលើការកាត្់ប ថយប ាុក្ស្អនក្ ិត្ិក្មម សៅក្នុខវិេ័យ េំភារៈក្េិក្មម។ ទី២ គឺការព្យប្ខឹខ ិរ ដរភាព្យបរិសាថ េប្ម្លប់អលិត្ក្មមក្េិក្មម។ ទី៣ គឺការបសខកើ គុណភាព្យ ក្មមវិធ្ី អសព្យែអាយក្េិក្មមសាធារណៈ ិខវិភាជ ៍ថវិកាបស្ ថម ល់ក្មមវិធ្ីទាខ ំ ឡាយស្ លម្ល ប្បេិទធភាព្យខ្ពេ់ សៅតាមលទធភាព្យ ន ថវិកាេរុបរបេ់រដ្ឋាភិបាលក្មពុោ។ ទី៤ គឺការជួយអភិវឌ្ឍវិេ័យក្េិោណិជផក្មម ិខឧេាេក្មមស្ក្នចនក្េិអល។ ការបសខកើ ទុ វិ សិ យាគ ខ ើ បនាា ់ ប៉ាស្ុ ដ ខ ិ ស្ក្លំអសោល សយាបាយ ចំបាច់ប្ត្វូ សធ្ែោ ើ លទធអល ភាាមៗស យ ឹ មិ បា ស ញ ើ ។ េប្ម្លប់ការស្ក្លំអមួយចំ ួ ចំបាច់ប្ត្ូវម្ល ការចំដយសព្យល ិខក្ិចចប្បឹខស្ប្បខបស្ ថមសទៀត្ ស ើមបីឱ្យស ើញ អលោក្់ស្េដខសៅ ឹខមូលដ្ឋា ។ តារាខទី២ េសខេបព្យីអ ុសាេ ៍សោល សយាបាយេំខា ់ៗ ស្ លចំបាច់ប្ត្ូវអ ុវត្ដភាាមៗ ស ើមបីបា ទទួលលទធអលក្នុខរយៈសព្យលខ្ាីសៅមធ្យម សេើយស្ ល ឹខជួយទប់ ិនានការថយចុេះ ន ក្ំសណើ វិេ័យក្េិក្មម បចចុបប នស េះ ប្ព្យមទាំខជួយ ័ ក្េិក្មមរបេ់ប្បសទេក្មពុោឱ្យបា សព្យញសលញ។ សប្បើប្បាេ់េកាដ ុព្យលន វិេយ ុ េ ៍េខា តារាខទី២៖ េសខេបអ សា ំ ៗ ែ សៅតាមសព្យលសវលា ស ម ់ ន របាយការណ៍ សដ្ឋយស្អក្ ើ បីេសប្មចលទធអល លទធអលស្ ល េសប្មចបា លទធអលស្ ល េសប្មចបា ក្នខុ រយៈសព្យលព្យី អ ុសាេ ស៍ ោល សយាបាយ ុ រយៈសព្យល ខ្ាី ក្នខ មធ្យម សៅស្វខ រក្ា វូ បរិយាកាេសោល សយាបាយក្េិក្មម ស្ លបខកភាព្យងាយប្េល ួ ល់វេ ិ យ ័ ឯក្ជ ប ដសោល សយាបាយោណិជផក្មមសបើក្ចំេរ ិខសោល សយាបាយក្េិក្មមស្ លមិ នាំឱ្យខ្ូចប្ទខ់ប្ទាយរបេ់វិេ័យ X កាត្់ប ថយចំដយសលើស្បបបទការនាំចូលជីក្េិក្មម X សបើក្ចំេរវិេ័យប្ោប់ព្យូជ េប្ម្លប់ការវិ ិសយាគព្យីវិេ័យឯក្ជ តាមរយៈការសធ្ែើ ប ិបបញ្ាត្ដិក្មម ិខការព្យប្ខឹខសាថ ប័ X ព្យប្ខឹខ រិ រដ ភាព្យបរិសាថ ស្ក្លំអការសរៀបចំស្អ ការសប្បើប្បាេ់ ីធ្ាី បសខកើ ប្បេិទធភាព្យការុេះ ុវត្តចាប់ ីធ្ាី ិខបទដ្ឋា គត្ិយុត្ដិោក្់ព្យ័ ធនានា X ព្យប្ខឹខេិទធិកា ់កាប់ ីធ្ាី X សលើក្ក្មពេ់ការប្គប់ប្គខ ីធ្ាីប្បក្បសដ្ឋយ ិរ ដរភាព្យ X សលើក្ក្មពេ់ការសប្បើប្បាេ់ជី ិខថានំគីមីក្េិក្មមប្បក្បសដ្ឋយេុវត្ថិភាព្យ X ស្ក្លំអក្មមវធ្ ិ វី ិ សិ យាគសលើ “ប្ទព្យយសាធារណៈ” ន វិេយ ័ ក្េិក្មម បសខកើ ថវិកាេប្ម្លប់ប្ទព្យយសាធារណៈក្េិក្មមេំខា ់ៗ សៅតាមលទធភាព្យថវិកា របេ់រដ្ឋាភិបាល ោព្យិសេេ សលើប្បព្យ័ ធធារាសាស្ដេដ ការេិក្ាប្សាវប្ោវ ការអសព្យែអាយ ការប្គប់ប្គខជីោត្ិ ី X ការប្ត្ួត្ព្យិ ិត្យគុណភាព្យេំភារៈក្េិក្មម េុវត្ថិភាព្យចំណីអាហារ ការបណដុ េះបដ ដ លវិោផជីវៈ ិខអាូវជ បទ។ េប្ម្លប់វិេ័យធារាសាស្ដេដ សផ្ទដត្ការយក្ចិត្ដទុក្ដ្ឋក្់ឱ្យបា កា ់ស្ត្សប្ចើ ោខមុ សលើការស្ក្លំអ ិខការសាដ រប្បព្យ័ ធស្ លម្ល ប្សាប់ ិខការប្គប់ប្គខសេដ្ឋារចនាេមព័ ធធារាសាស្ដេដ សដ្ឋយម្ល ការចូលរួម។ X បសខកើ គុណភាព្យន ក្មមវិធ្ីក្េិក្មមសាធារណៈ X េប្មបេប្មួលការសធ្ែើេម្លេរណក្មមក្េិដ្ឋា ត្ូចៗ ចូលសៅក្នុខស្ខ្សេងាែក្់ត្ំនលន សេបៀខ X ជួយអភិវឌ្ឍក្េិោណិជក្ ផ មម ិខឧេាេក្មមស្ក្នចនក្េិអល បសខកើ លទធភាព្យទទួលបា អគាិេ ីសប្បើប្បាេ់ ិខកាត្់ប ថយចំដយសលើការសប្បើប្បាេ់អគាិេ ី X វិ ិសយាគសលើសេដ្ឋារចនាេមព័ ធសាធារណៈ ស្ លោក្់ព្យ័ ធោមួយ ឹខេុវត្ថិភាព្យចំណីអាហារ X សធ្ែើឱ្យកា ់ស្ត្ប្បសេើរ ូវសេវា ឹក្ជញ្ជូ ន ជំ ួញប្េូវអខករ រួមទាំខកាត្់ប ថយចំដយ X ុ ញ្ហចំ (warehouse receipts) លុបសចលរបាំខនានា ស ើមបីដ្ឋក្់ឱ្យសប្បើប្បាេ់ ូវប្បព្យ័ ធទទួលយក្ទំ ិញក្នុខឃ្ា ំខោវត្ថប X xxi 1. INTRODUCTION 1. Cambodian agriculture is in the midst of a rapid transformation. Its annual production growth during 2002-2012 averaged 9.6 percent, far exceeding agricultural production growth in Asia and around the world. This high agricultural growth has greatly contributed to poverty reduction, exports, agribusiness development, and farm commercialization in the last decade. While the Royal Government of Cambodia (RGC)’s agricultural policy focus has mainly been on rice production and exports, it is increasingly recognized that the current rice-based traditional agriculture would need to transform into modern, diversified agriculture to generate the returns necessary to keep farming an attractive source of income and a contributor to economic development. Cambodian agriculture is still a long way from being an economic activity indistinguishable from other sectors, at least in terms of productivity of labor and capital or in the location of poverty, which would be the final outcome of a structural transformation. The agricultural growth has recently slowed down to about 1 percent, from 5.3 percent during 2004- 2012. Still, the last decade has seen significant positive changes. 2. How can Cambodia build on this positive development? While there is a wealth of literature on smallholder farming systems in Cambodia and sectoral information from the national accounts, the existing analytical work lacks a systematic analysis of changes in farming systems. Combined with the lack of information on related changes in farm versus nonfarm incomes, this has limited Cambodia’s ability to have an informed agricultural policymaking process and has hampered efforts to compare and contrast the impacts and effectiveness of alternative policies and programs for long-term agricultural development. 3. Further, little empirical knowledge has been generated about past and future sources of agricultural growth and the government’s role in facilitating growth. Why did agriculture grow so rapidly and will this growth last? What made that growth pro-poor and how can Cambodia keep its strong pro-poor impact? How would farming systems need to look to ensure continuation of the 5 percent growth in agricultural value added? What forces will shape an enabling environment for agriculture in the future and how different will they be from those that drove development in the past? And how can the government facilitate future growth, helping farmers overcome challenges and capitalize on opportunities? 4. This report addresses this knowledge gap. It was requested by the Supreme National Economic Council (SNEC) and the Ministry of Agriculture Forestry and Fisheries (MAFF). The report uses primary data on farming systems collected in 2013, and compares them to comparable data from 2005. It also uses evidence from interviews with villages and experts, discussions with government staff, and analytical work carried out by the World Bank Group and other development partners in the recent decade, including the analytical work resulting from the five-year collaboration between the World Bank and Australian Aid under the Cambodia Food Crisis Capacity Support Partnership Trust Fund (Annex 1). 5. The report covers the main crops with the largest changes in the past decade and focuses on smallholders, including small farms (up to 1 hectare), medium farms (from 1 to 3 hectares), and large farms (above 3 hectares). It does not cover the livestock and fisheries sector, nor does it cover in detail economic land concessions. 6. The remaining report is organized as follows. Chapter 2 presents key facts about Cambodia’s recent agricultural development using data from national accounts and various 1 reports. Chapter 3 provides evidence from the field that explains the changes observed in the national accounts. Chapter 4 illustrates developments in farming systems, farm budgets, and farm incomes compared to nonfarm incomes. Chapter 5 presents a farm competitiveness analysis. Chapter 6 discusses the sources of past growth and their limitations and presents an analysis of (likely) future sources of agricultural growth. Chapter 7 presents a long-term vision for the sector, while Chapter 8 simulates policies and the changes in farm incomes needed to realize this vision. Chapter 9 discusses the policy agenda, with implementation details based on national and global experiences. Chapter 10 concludes with a summary of the report and policy recommendations. Annexes present the methodology of the 2013 farm survey, detailed farm budgets by crop, projections of selected indicators, and results of the policy simulations. 2 2. A DECADE OF AGRICULTURAL TRANSFORMATION IN CAMBODIA 7. The story of agriculture in Cambodia over the past decade is one of rapid growth and of the profound transformation of a rural society. The sector has been growing in absolute terms but its relative share in the economy is on the decline, giving room to manufacturing and services. Agricultural exports have increased. Agricultural labor has been migrating to other sectors and outside of the country and is becoming increasingly scarce. Farmers use more inputs and more modern machinery to substitute for scarcer labor and respond to market incentives. 8. Agricultural policy has in general been conducive for non-distortive growth, i.e., there are no output and input subsidies and no significant market distortions. However, in addition to the high costs of doing business in the country overall, regulatory weaknesses exist, especially in the seed industry, which deters private investments. The weak land tenure security and poor land administration and governance deter private investments further and lead to unsustainable land use and farmland expansion. Yet, in spite of these issues, the rural poverty declined in the past decade and agricultural growth played the key role in that poverty reduction. 9. The structural transformation is still occurring, and Cambodian agriculture is in the midst of rapidly transitioning from a traditional subsistence to a modern commercial sector. This Chapter summarizes the key facts of this transformation using information from MAFF, the Cambodian National Institute of Statistics (NIS), the World Development Indicators (WDI), World Bank reports, the Corporate Statistical Database of the UN’s Food and Agriculture Organization (FAOSTAT), and other sources. 2.1. Agricultural Growth 10. Cambodian agriculture’s growth over the past decade is impressive. Real agricultural GDP grew by 5.1 percent annually between 2002 and 2012 (Figure 1). The average annual growth was 5.3 percent during 2004-2012, the period mostly analyzed in this report. Agricultural value added per worker grew from $700 in 2004 to $1,210 in 2012 (Table 13). The ratio of value added per worker in nonagriculture to agriculture fell from 3.2 in 2004 to 2.1 in 2012, implying a narrowing income gap between agriculture and the rest of the economy. Figure 1: Agricultural GDP growth (constant prices), Cambodia, 2002-2012 Source: World Bank 2014c. 3 11. Taking a longer-term perspective, the composition of Cambodia’s economy has changed dramatically, reducing agriculture’s share and increasing industry’s role. A structural transformation has been taking place. Despite the high growth in agricultural value added, agriculture’s share in GDP steadily decreased from 45 percent in 1995 to 26 percent in 2012, with most of the gains captured by the garment industry (Figure 2). The share of agriculture in GDP, however, has stayed unchanged since 2008. In 2011, the share of industry (28 percent) exceeded the share of agriculture (26 percent) for the first time in the modern history of Cambodia. Figure 2: GDP composition by sector (constant prices), Cambodia, 1995-2012 Source: World Bank 2014c. 12. The analysis in current prices shows a slightly different picture. Calculated in current prices, the share of agriculture in total GDP decreased less, from 48 percent in 1995 to only 34 percent in 2012 (Figure 3). Between 2002 and 2012, the share of agriculture in GDP actually grew, from 31 percent to 34 percent. The difference between the shares of agriculture in GDP measured in current (34 percent) and constant (26 percent) prices implies that agricultural prices grew faster than nonagricultural prices, triggering the high rate of growth of agricultural production. Figure 3: GDP composition by sector (current prices), Cambodia, 1995-2012 Source: World Bank 2014c. 4 13. This high growth in agricultural production is shown in Table 3. During 2002-2012, the rate of gross agricultural production growth in Cambodia was 10 percent per year, one of the highest in the world. Gross per capita agricultural production growth was 8 percent, also among the highest. Table 3: Gross agricultural production, annual growth, selected countries, 2002-2012 Country/Region Average Annual Growth (%) Bangladesh 2.0 Thailand 2.6 China 3.5 Vietnam 4.1 India 4.3 Indonesia 4.4 Lao PDR 5.3 Cambodia 9.6 Asia 3.4 European Union -0.3 World 2.5 Source: FAOSTAT 2014. 14. Cambodia’s agricultural growth was driven by crops, which are dominated by paddy rice. The average growth of crops’ value added was 7.8 percent (Figure 4). Livestock’s value added increased by 3.6 percent per year. The average growth in the fisheries and forestry subsectors was 3.7 percent and 0.7 percent, respectively. Figure 4: Growth of agricultural subsectors (constant prices), Cambodia, 2002-2012 Source: NIS 2014. 15. Crops remain the major agricultural subsector. Due to their high rate of growth, the share of crops in agricultural value added increased from 43 percent in 2002 to 54 percent in 2012 (Figure 5). Cereals, especially paddy rice, accounted for the majority of the crop production value, 5 but the production of non-cereals grew faster than paddy in the last decade (Table 4). The importance of livestock production in agricultural value added declined from 17 percent in 2002 to 14 percent in 2012. Because of its dominance in agricultural value added, this report focuses on crop production. Figure 5: Share of subsectors in agricultural GDP, Cambodia, 2002-2012 Source: NIS 2014. 16. Over the past 10 years, rice production more than doubled through a combination of land expansion and yield increase. In 2002, paddy production was 3.8 million tons (Table 4). In 2012, it reached 9.3 million tons. The annual growth rate was 9.3 percent. The wet season accounts for 75 percent of total paddy production. As just 8 percent of Cambodia’s arable land is estimated to be irrigated during the dry season (the lowest rate in Asia; see Table 5), the country’s agriculture is still largely dependent on rainfall and quite vulnerable to climatic events.9 9 The share of dry season output, however, has been steadily growing, driven by irrigation investments. In 2008, the planting area for the dry season crop accounted for 13 percent of total planting rice area. In 2013, it increased to 17 percent. 6 Table 4: Production, area, yields, and growth of major crops, Cambodia, 2002-2012 2002 2012 Annual Average Growth Rate (%) Production Cultivated Yield Production Cultivated Yield Production Cultivated Yield Crop (ton) Area (ha) (ton/ha) (ton) Area (ha) (ton/ha) (ton) Area (ha) (ton/ha) Rice 3,822,509 1,994,645 1.916 9,290,940 2,980,297 3.117 9.3% 4.1% 5.0% Maize 148,897 80,470 1.850 950,909 215,442 4.414 20.4% 10.3% 9.1% Cassava 122,014 19,563 6.237 7,613,697 337,800 22.539 51.2% 33.0% 13.7% Vegetables 163,175 34,433 4.739 411,435 54,155 7.597 9.7% 4.6% 4.8% Soybean 38,661 33,438 1.156 120,165 70,972 1.693 12.0% 7.8% 3.9% Sugarcane 173,105 9,581 18.068 1,220,255 36,722 33.230 21.6% 14.4% 6.3% Source: MAFF 2014b. Table 5: Irrigated areas, selected countries Country Actually Irrigated Areas Arable Land Share of Irrigated Areas (ha) (ha) in Arable Land (%) 2011-2012 2011 Cambodia 317,225 4,000,000 7.9 China 54,218,976 111,598,500 48.6 Indonesia* 6,722,299 23,500,000 28.6 Malaysia* 340,717 1,800,000 18.9 Myanmar* 2,083,000 10,786,000 19.3 Philippines* 1,879,084 5,400,000 34.8 South Korea 880,400 1,492,000 59.0 Thailand 5,059,914 15,760,000 32.1 Vietnam* 4,585,500 6,500,000 70.5 Note: *Equipped full control irrigation areas as data on actually irrigated areas are not available. Source: World Bank 2014g for data on arable land and World Bank 2014c for data on irrigated areas. 17. Growth in the crops subsector was not limited to rice. Production of several other important crops increased even more than that of rice (Table 4). The most spectacular cases are cassava, sugarcane, and maize, which have all grown more than 20 percent per year. This growth is the result of both area and yield increases. Current yields are still low but their annual growth over the last 10 years was steady, varying between 3.9 percent for soybeans to 13.7 percent for cassava. 18. The rubber subsector has also witnessed remarkable growth. Planted areas increased by 16 percent annually (Table 6) and when these areas go into production, rubber output will likely triple over the next 10 years. 7 Table 6: Production, area, yield, and export of rubber, Cambodia, 2002-2011 Total Planted Immature Mature Production Yield Export Year Area (ha) (ha) (ha) (tons) (kg/ha) (tons) 2002 55,582 18,809 36,773 32,384 880 36,774 2003 53,527 19,831 33,696 32,382 960 32,764 2004 54,209 22,619 31,590 33,770 1,060 33,558 2005 60,406 30,004 30,402 29,464 960 29,950 2006 69,994 37,604 32,390 32,077 990 31,184 2007 82,059 51,568 30,491 32,975 1,080 33,121 2008 108,510 74,197 34,313 37,050 1,080 36,000 2009 129,920 95,785 34,135 37,380 1,095 36,500 2010 181,433 143,027 38,406 42,466 1,100 45,000 2011 213,104 167,942 45,162 51,339 1,137 44,969 Annual growth (%) 16% 28% 2% 5% 3% 2% Source: MAFF 2014b. 19. Most crop products are exported. Rice, cassava, and other cereals are the major agricultural export items. In 2011, formal cereal exports were estimated at 180,300 tons, a five- fold increase compared to 2006 (Table 7). The largest increase in exports was for rice. But the informality of cross-border trade does not allow estimation of the true extent of agricultural exports. In 2013, for example, the formal export of rice was 378,850 tons (Table 9). At the same time, informal cross-border trade of rice is estimated to have been 1,536,000 tons, or five times larger. Most exports of cassava, maize, and vegetables also take place via cross-border trade. Official statistics do not properly capture this information, including imports of vegetables in apparently large volumes. Table 7: Evolution of formal trade of select commodities, Cambodia, 1996-2011 Quantity (‘000 tons) Annual Growth Rate (%) Commodity 1996 2001 2006 2011 1996-2001 2001-2006 2006-2011 Cereals exports 10.4 7.1 38.0 180.3 -7.4 39.9 36.5 Rice exports 5.6 7.0 5.2 174.0 4.6 -5.8 101.8 Coarse grain exports 4.8 0.0 32.6 6.3 -100.0 n/a -28.0 Cereals imports 41.8 59.8 65.2 44.4 7.4 1.7 -7.4 Source: NIS 2014 based on FAOSTAT 2014. 2.2. Agribusiness Development 20. Despite this impressive growth of the primary agriculture, Cambodia’s agribusiness remains weak. Agroindustry is basic and most commodities (paddy, cassava, maize, cashews, cattle, and soy bean) are exported in raw form. A modern agroindustry is emerging for rice, but this is a very recent phenomenon, stimulated by the RGC’s “Promotion of Paddy Production and Rice Export” policy in 2010 and favorable world rice price trends. 8 21. Agribusiness covers several industries. It includes food, beverages, tobacco, and rubber plants. Transport and trade sectors also generate some income from moving/trading agricultural products, estimated at 50 percent and 30 percent of the sectoral value added respectively.10 When added, the agribusiness accounted for 10 percent of GDP and 31 percent of agricultural GDP in 2002 (Table 8). In 2012, these shares were 9 percent and 27 percent.11 Table 8: Agribusiness industry in Cambodia, 2002-2012 2002 2012 Agribusiness GDP (billion Riels, current prices) 5,224 19,000 Share of agribusiness in agricultural GDP (%) 30.8 27.4 Share of agribusiness in total GDP (%) 9.6 9.2 Source: World Bank 2014c. 22. The agribusiness sector consists mostly of small traders and informal agro- enterprises (Figure 6). Even though a modern sector is emerging (e.g., processors, input suppliers, and machinery suppliers), less than half of the agroenterprises have some degree of formality (e.g., registration and accounting systems). Figure 6: Contribution to turnover of different types of agribusinesses, Cambodia, 2011 Source: World Bank and AusAid 2013. 23. The recent expansion of the rice milling industry has been exemplary. In 2010, the RGC issued a policy on “Promotion of Paddy Production and Rice Export,” aiming to export one million ton of milled rice by 2015. Major features of that policy are a strong commitment to open trade,12 reduction of costs of doing business, and protection of foreign investors’ rights. As a result, formal exports of milled rice increased from 12,600 tons in 2009 to over 205,000 tons in 2012, and 378,850 tons by the end of 2013 (Table 9). The “One Window Service” (or one-stop service) 10 There are no official estimates of these shares for Cambodia. These estimates are based on the experiences of the team producing this report. 11 These data are presented in nominal prices, which are not very different from the results in constant prices. 12 Before 2010, only 9 companies had export licenses. The 2010 Rice Policy liberalized exports, and in 2014, 86 companies were already exporting milled rice from Cambodia. Most export is still paddy, across the border to Thailand and Vietnam, but the export of milled rice picked up significantly after the removal of the export licensing restriction. 9 was established in November 2011 to reduce bureaucratic obstacles and export costs to process export documents (such as sanitary and phytosanitary certificates, customs documents, and CAMCONTROL quantity and quality certificates). This reduced the export approval period from 10 days in 2011 to 4-5 days in 2013 (World Bank 2014b). In 2013, 378,850 tons of milled rice were exported to over 50 countries, 91 percent of which went to the EU and Asian countries. About half of formal milled rice exports consist of high-value fragrant rice. The Federation of Cambodian Rice Exporters was created and the industry is becoming increasingly organized. Table 9: Rice commodity balance (‘000 tons), Cambodia, 2008-2013 2008 2009 2010 2011 2012 2013 Total rice production, in paddy 7,175 7,586 8,250 8,779 9,291 9,389 Total rice production, in milled equivalent 4,305 4,552 4,950 5,267 5,575 5,633 Domestic utilization13 2,862 2,937 3,039 3,126 3,212 3,256 Total rice surplus 1,443 1,614 1,911 2,142 2,368 2,378 Total formal export of milled rice 1.5 12.61 105.26 201.89 205.71 378.85 Estimated informal paddy export in milled 100 200 350 1,472 1,600 1,536 equivalent Source: World Bank 2014b . 24. New actors in the milling industry have emerged since 2010. In 2009, only two rice milling companies existed with a capacity of 20 tons per hour (tph). By mid-2012, milling capacity had increased to 322 tph (paddy), and to over 700 tph in 2013 (World Bank 2014b). The polishing capacity of mills also jumped, from 72 tph in 2009 to 520 tph in 2013. Additionally, three rice polishing factories have a combined capacity of 44 tph. This increase in polishing capacity enabled low-quality milled rice from thousands of small mills14 to be upgraded for export. Most new mills are equipped with modern milling and drying facilities, polishing equipment with color sorters, and large warehouses for storing paddy at harvest time (Slayton and Muniroth 2009; 2012). 25. But even this success story has its limits. Most rice is still exported in the form of paddy to Vietnam and Thailand. In 2013, informal export volumes were estimated at 1.7 million tons of paddy to Vietnam and 250,000 tons of paddy and 450,000 of milled rice to Thailand, compared to only 379,000 tons of formal rice exports (Table 9). Most mills continue to have low capacity utilization as a result of underdeveloped supply chain management, poor access to skilled labor, food safety-related services, proper business management and working capital, and infrastructure and logistics weaknesses. They operate on one shift per day, for a limited number of days per year. Consequently, Cambodian mills are less profitable than Vietnamese and Thai mills, where two to three shifts per day and almost a full year of operating days are the norm. 2.3. Agricultural Growth and Poverty Reduction 26. The impressive agricultural growth in the last decade has contributed to significant poverty reduction in Cambodia.15 From 2007 to 2011, the overall poverty headcount fell from 13 Domestic food consumption and other utilization on estimated basis. 14 According to the Ministry of Industry, there were 28,474 rice mills in Cambodia in 2011, of which 16,326 run with less than $3,000 capital. 15 This section is based on World Bank (2013b). 10 50.1 percent to 20.5 percent. During this period, most of the poverty reduction came from improving rural households’ conditions. From all poverty reduction, 3.6 percent was in Phnom Penh, 8.1 percent was in other urban areas, and the remaining 88.3 percent was in rural Cambodia (Figure 7). Furthermore, most of the poverty reduction (80.0 percent) occurred between 2007 and 2009. Figure 7: Poverty rate by region, Cambodia, 2004-2011 Source: World Bank 2013b. 27. The growth elasticity of poverty in Cambodia was among the largest in Asia. Measured in the 2005 purchasing parity power (PPP) for the poverty at $1.25 a day, the growth elasticity of poverty was 5.2, implying that for every 1 percent increase in GDP per capita, poverty fell by 5.2 percent. Among selected countries presented in Table 10, for which most recent poverty data is available, only Thailand had the larger impact of economic growth on poverty. Table 10: Growth elasticity of poverty, selected countries in Asia Country Poverty rate (PPP Growth elasticity of poverty Annualized per capita 2005 $1.25/day) (PPP 2005 $1.25/day) GDP growth rate Lao PDR 28.8 0.5 5.9 Philippines 17.8 0.8 4.2 Indonesia 9.7 2.4 4.5 Vietnam 2.9 3.5 4.6 Cambodia 7.2 5.2 3.3 Thailand 0.1 7.5 1.4 Source: World Bank 2013g. 28. The first factor explaining the reduction in rural poverty is the price of rice. There are more net sellers of rice than net buyers in Cambodia (Ivanich and Martin 2014). The global rice price spike in 2008 was permitted to fully pass through to local prices in Cambodia. As a result, many households improved their income, most of them smallholders. Better prices had two direct effects on household income. First, the value per unit of production increased. And second, total production increased due to more planted area and increased use of inputs (Figure 8). 29. The second factor explaining the reduction in poverty is higher agricultural wages. Because of data limitations, it was not possible to differentiate the sources of wages from different 11 agricultural activities, but there is no doubt that an important share of the wage increases was related to rice production. For example, rice represents about 80 percent of the total planted area in Cambodia (Table 4), and about 80 percent of rural households reported growing rice. Moreover, rice experienced the highest price increase (39 percent) from 2007 to 2009 compared to all other locally produced food products: beef – 25 percent, chicken –24 percent, fish and seafood – 24 percent, fresh eggs – 14 percent, fresh fruits – 27 percent, and vegetables – 22 percent. From 2004 to 2009, agricultural income from wage labor increased 49 percent. Figure 8: Drivers of poverty reduction in Cambodia, 2004-2011 Source: World Bank 2013b. 30. Overall, the direct impact of higher rice prices and the increase in rice production were responsible for almost half of the reduction in poverty. Agricultural wages – also closely related to rice – contributed to 16 percent of total reduction in poverty. At the same time, improved income from nonfarm business was responsible for one-fifth of all poverty reduction. Increases in salaried employment in urban areas accounted for 4 percent of the decline in poverty. Finally, the remaining 14 percent of the reduction was unexplained and occurred mainly between 2009 and 2011 (Figure 8). 31. Despite the significant reduction in poverty headcounts, the vulnerability or the risk of sliding back into poverty remains high in Cambodia. The recent poverty reduction was augmented substantially by the high concentration of people just below the poverty line in 2004 and people just above the poverty line in 2011. Most people who escaped poverty did so by a small margin. It is estimated that a loss of only 1,200 Riels ($0.30) per day would cause Cambodia’s poverty rate to double to 40 percent. Continuation of agricultural and nonagricultural growth and their higher resilience to shocks are needed to reduce that vulnerability in the future. 32. Nutritional security has also lagged behind the progress in poverty reduction . The prevalence of stunting of children under five in 2005 was 44 percent, according to the World Bank. In 2010, it declined to 40 percent and in 2014 to 33 percent, according to the results of the Cambodian Demographic and Health Survey 2014, but it remains high in regional comparison.16 Stunting remained above 40 percent in four provinces: Kampong Chhnang, Kompong Speu, Preah Vihear, and Stung Treng. Stunting is caused not only by lack of food intake but also poor sanitation 16 In Vietnam, the stunting of children under five was 23 percent in 2011. In Thailand, it was 16 percent in 2012. 12 and health care. Yet, irrespective of the multidimensional nature of malnutrition, agriculture can do more to address it. 2.4. Agricultural Transformation 33. Structural transformation is a defining characteristic of the development process, both the cause and effect of economic growth. According to Timmer and Akkus (2008), four quite relentless and interrelated processes define structural transformation: (i) a declining share of agriculture in GDP and employment; (ii) migration from rural to urban areas and a rapid process of urbanization; (iii) the rise of a modern industrial and service economy; and (iv) a demographic transition from high rates of birth and death (common in underserved rural areas) to low rates of birth and death (associated with better health standards in urban areas). The final outcome of structural transformation, already visible on the horizon in rich countries, is an economy and society where agriculture as an economic activity has no distinguishing characteristics from other sectors, at least in terms of the productivity of labor and capital or the location of poverty. 34. Structural transformation clearly took place in Cambodia over the last decade. The share of agriculture in GDP measured in constant prices declined from 28.8 percent in 2004 to 25.6 percent in 2012, while the share of industry in total GDP increased from 26.9 percent to 29.0 percent (Table 11).17 The agricultural labor force decreased by 6 percentage points.18 Both birth and death rates declined, and the urbanization process started. Urbanization has been very slow, however; in 2012, 78 percent of Cambodians still lived in rural areas (Error! Reference source ot found.). Table 11: Indicators of structural transformation, Cambodia, 2004 and 2012 Indicators 2004 2012 Agricultural GDP as % of total GDP 28.8 25.6 Ag. employment as % of total labor force 57.4 51.0 Industrial GDP as % of total GDP 26.9 29.0 Rural population as % of total population 81.0 78.0 Birth rate, crude (per 1,000 people) 26.6 25.9 Death rate, crude (per 1,000 people) 7.8 6.0 Source: NIS 2014 for GDP, employment, and populationand WDI 2014 for birth and death rates. 17 Reliable labor statistics starts only from 2004, collected through the Social Economic Surveys by NIS. This is why the indicators of transformation also start with the year 2004 not 2002 as in the earlier analysis. 18 According to MAFF, the share of agricultural labor in the total labor force increased from 60 percent in 2004 to 71 percent in 2012. This increase contradicts the evidence-based analysis carried out in this report and the results of the NIS Socioeconomic Surveys, which are considered the most accurate and up-to-date in Cambodia. The report uses the NIS data on labor statistics. The ADB also used the NIS labor statistics in its reports prepared in 2014. 13 Box 1: Rural Population in Cambodia In 2012, Cambodia’s rural population was 11.1 million people, or 78 percent of the total population. At the same time, the agricultural labor force accounted for 51 percent of the total labor force. How does the “low” 51 percent agricultural labor share reconcile with the “high” 78 percent of the rural population? What do rural people do if not farming? The answer is that there is a difference between population and labor force, as the latter does not include children, seniors, and the unemployed. According to the socioeconomic surveys of the NIS, in 2012 about 62 percent of the rural population was of working age; i.e., the remaining 38 percent were children younger than 15 years and seniors older than 65 years. The labor participation rate of the rural population was 86 percent, with 91 percent for men and 80 percent for women. And the share of agricultural labor in the total rural labor force was 66 percent. Out of 2,393 rural households in 2012, 1,877 were recorded as farming households in the 2013 Agricultural Census of Cambodia. This means that not all rural households are agricultural and not all members of households participate in the labor market. 35. Where does Cambodia stand in the process of structural transformation in a global comparison? Countries can be divided into groups by their dependence on the role of agriculture. One such grouping was done by the agribusiness team of the World Bank Group (World Bank 2014a), using the framework developed by the 2008 World Development Report. Countries were divided into five groups, from agri-based to most developed (Table 12). In 1980, 41 countries were considered agri-based economies and only 18 countries were considered developed. By 2012, the number of agri-based economies declined to 28 while the number of developed economies increased to 50. Table 12: Country groups based on the role of agriculture in the economy Indicators Agri- Pre- Transitioning Urbanizing Developed Based Transitioning Ag. value added as % of total 40% 11% 12% 9% 2% GDP Ag. employment as % of total 72% 59% 40% 19% 4% labor force Ag. value added per capita, in 387 521 514 682 519 PPP $ Non-ag. value added per 588 4,369 3,836 6,577 27,467 capita, in PPP $ Total GDP per capita, 975 4,879 4,350 7,256 27,986 in PPP $ Ag. value added per ag. 1,236 1,727 2,868 8,770 30,588 worker, in PPP $ Ag. capital stock per ag. 715 658 1,408 5,436 37,711 worker, in PPP $ % people below the $2 36% 32% 14% 6% 0% international poverty line Source: World Bank 2014a. 14 36. In 2012, Cambodia was still an agri-based economy, but it came close to being a pre- transitioning country. Considering the share of agricultural labor in its total labor force and its poverty rates, Cambodia is rapidly approaching the status of a pre-transitioning country (Table 13). The agricultural value added in purchasing power parity (PPP) actually exceeds the average for this group of countries. But for other indicators, such as the share of agriculture in GDP and nonfarm incomes, Cambodia is still very much an agri-based economy. 37. The process of structural transformation in Cambodia will continue. Agriculture is likely to play a larger role in Cambodian economic growth and job creation than it does in neighboring countries at the same stage of development given the agriculture sector’s strong comparative advantages. Agricultural growth can provide excellent opportunities for income generation and poverty reduction for many years to come. 38. Yet future transformation may take different pathways depending on which policy decisions are made now. Cambodia may follow the positive example of Thailand.19 In this neighboring country, agricultural growth during 1960s and 1970s was driven largely through expansion of the land frontier and cheap labor, similar to the current Cambodian situation. Agricultural land per worker increased steadily, which was subsequently converted into productivity improvements through farm mechanization and public spending on infrastructure and human capital. In turn, all of these contributed to significant farm productivity increases, critical for the production of exportable surpluses. Farming systems during that period were largely dominated by traditional smallholders and the export was raw materials of lower-value product segments. But the end of this period also saw the emergence of sophisticated private sector value- added processing. Table 13: Selected economic indicators, Cambodia, 2004 and 2012 Indicators 2004 2012 Ag. value added as % of total GDP 28.8 25.6 Ag. employment as % of total labor force 57.4 51.0 Ag. value added per capita, current $ 120 318 Ag. value added per capita, in PPP $ 447 955 Non-ag. value added per capita, in PPP $ 1,074 1,896 Total GDP per capita, in PPP $ 1,516 2,839 Ag. value added per ag. worker, in PPP $ 1,672 3,600 % people below the $2 international poverty line 60.5 32.6 Source: NIS 2014, WDI 2014, and East Asia Poverty Portal, the World Bank. 39. Thai agriculture lost its competitiveness in the late 1980s due to underinvestment in agricultural public goods and worsening agricultural terms of trade (TOT).20 Exhaustion of the land frontier and an increasing farm population led to a period of declining agricultural land per worker, while the competitiveness of agricultural exports was affected by an overvalued national currency caused by the influx of short-term capital and maintenance of a fixed exchange rate. By the beginning of the early 1990s, Thai farmers were experiencing rising labor costs and shortages of family labor due to increasing rural-urban migration of young agricultural workers to 19 The analysis of the structural transformation of the Thai agriculture is based on Nippon (2013). 20 Agricultural TOT is defined as the ratio of prices of agricultural commodities to the prices of farm inputs. 15 the urban-based manufacturing and service sectors. Average annual agriculture growth rates hovered just over 4 percent throughout the 1980s, dropping to 3.1 percent during 1990-1996. 40. The government and private sector responded to that economic decline. In the 1990s, they invested in productive infrastructure and a system of food quality standards, which triggered the next wave of structural transformation. These investments, coupled with more open agricultural trade policy, made an important contribution to the shift towards mechanized farming and higher-value food processing and exports during this period. Labor use per hectare in rice cultivation declined from 58 person-days in 1987 to just 8-10 person-days according to Isvilanonda et al. (2000) and Bordey et al. (2014). Thai food exports shifted from commodities to higher-value, processed or final products. Between 1992 and 2011, the share of processed food and agricultural products in total agricultural exports increased from 20 percent to 26 percent. For example, most fishery products are now exported in the form of processed products, while that of rice and cereal are exported as small, packaged, uncooked rice and cereal-based finished products, rather than in 100 kg bags. Among rice products, there have been increasing exports of fragrant rice (Hom Mali) and parboiled rice, as opposed to lower-value white rice. The share of processed cassava exports (i.e., starch and modified starch) jumped from 20 percent to 60 percent in the same period. This implies that Thailand is able to obtain more value added from agricultural and food exports as its composition has shifted to more high-value products, supporting further transformation of its agriculture sector.21 41. Yet there are also examples of less successful agricultural transformation in Asia. In the Philippines, for example, the structure of agricultural production and farm labor use have hardly changed over many decades, with traditional low-value crops continuing to dominate. Rice accounted for 16 percent of agricultural value added in 2005, slightly more than in 1980-1990. Its share in total agricultural production has not changed in recent years. The importance of bananas, mango, pineapple, and other tropical crops, in which the Philippines has strong comparative advantage and which are major agricultural export items, has hardly changed over three decades (World Bank 2007). Rice production continues to be labor intensive, with the same labor use in Central Luzon in 2011/12 as was seen in 1966/67, especially for wet season rice (Table 14). The limited dynamism and innovation in the sector hamper agricultural transformation. Table 14: Labor use for rice production in Central Luzon, Philippines (man-days/ha), 1966/67-2011/12 1966/67 1970/71 1974/75 1986/87 1990/91 1994/95 2003/04 2007/08 2011/12 Wet season 69 68 85 71 74 71 61 66 71 Dry season 70 76 98 67 59 68 52 53 57 Source: Central Luzon Loop Survey, 19960-2012, IRRI. 42. Many reasons explain such divergent performance between Thai and Philippine agriculture, but the main ones related to land, farm protection, and the quality of public expenditures. Farmland tenure security is strong in Thailand, while the Philippines’ farmland market is paralyzed by very slow implementation of its land redistribution program (the 21 In the 2010s, however, Thai agricultural policy has become less conducive to structural change. The controversial rice pledge scheme has led to a large budget drain, neglected spending on core public goods, and significant economic distortions on the rice market. This policy has heavily slowed down the successful agricultural transformation. To avoid this trap, Cambodia should not follow this example. 16 Comprehensive Agrarian Reform Program). Delays and legal uncertainties over program implementation have led to disinvestment in the sector, especially in capital-intensive tropical crops and mechanization, thereby promoting farmers’ investments into less profitable, annual crops such as rice. The Philippines also carries out highly protectionist farm policy, in contrast to Thailand, where farm protection is a recent phenomenon. In the Philippines, the agricultural sector has been subject to high import protection and domestic monopolies for decades. Prices of rice, maize, sugar, and poultry are among the highest in the world. This protection keeps many farmers inefficient, allowing them to maintain their incomes at the expense of consumers. Poor quality of agricultural public expenditure and their bias towards the provision of subsidies and other private goods instead of public goods have further hampered agricultural transformation in the Philippines (World Bank 2007). 43. This brief dive into international experience is an important reminder for Cambodia. Future agricultural transformation depends on policy decisions made today. For Cambodia to shape its policy agenda and move to the next level of agricultural transformation, a thorough analysis is required of the past development of the agricultures sector, the drivers of past growth, and the potential drivers of future growth, complemented by lessons from global successes and failures. This analytical work contributes to this analysis. 17 3. CHANGES IN AGRICULTURE: EVIDENCE FROM THE FIELD 3.1. Introduction 44. The agricultural growth of the last decade represented in the national accounts (as presented in Chapter 2) is also evident in the field. Significant changes have occurred in the structure of Cambodia’s farming systems, as revealed through a combination of primary data collection, key informant interviews, and focus group discussions (FGDs). In 2013, 36 FGDs were conducted in 24 villages and interviews held with 179 farmers in 6 provinces and 12 districts, 12 commune councils, and dozens of representatives from MAFF and the Ministry of Land at the district and regional levels.22 45. To ensure the robustness of the results, information from different sources was triangulated. In other words, data from FGDs and individual farm surveys, as well as information collected from farm budgets, were used together to assess the status of agricultural production in Cambodia and to elicit the perspectives of sector stakeholders. This approach allowed better verification and cross-checking of data for farm enterprises and drivers of change. The details of the methodology for primary data collection and interviews are provided in Annex 2 and described briefly below. 46. Rice, cassava, maize, and vegetables best characterize the changes in Cambodia’s farming systems since the mid-2000s. In addition to significant changes in production growth over the past 10 years, each product had specific defining characteristics: (i) rice is the most important crop in terms of GDP contribution, employment, food security, and exports; (ii) maize is characterized by its rapid adoption of improved seeds and integration with the feed industry; (iii) cassava growth is exceptional but its development is subject to intense discussion related to sustainability and biodiversity conservation; and (iv) vegetables are related to nutrition and food safety, and are a good candidate for import substitution. 47. Qualitative methods were used to collect data for the 2013 survey,23 including key informant interviews and FGDs, complemented by quantitative data collection on farm budgets. Data were collected at the national, provincial, district, commune, village, and individual farm level. Structured survey tools were used to gather data from individual farmers for the financial analysis of farm enterprises. A brief discussion of the 2013 survey follows. 48. Provinces, districts, and communes for the 2013 survey were selected based on their contribution to the production of the four commodities. Those visited in the 2005 survey were included in the sample if they fit the purpose of the current research questions. The selection sequence was: (i) selection of provinces; (ii) selection of two districts within each province; (iii) selection of one commune within each district; and (iv) selection of one village within each commune. The six sampled provinces accounted for 46 percent of the national production of rice in 2013,24 83 percent of maize production, 57 percent of cassava production, and 60 percent of vegetable production. 49. At the national, provincial, district, and commune level, key informant interviews were conducted with experts in the field of agriculture, rural development, and land 22 The survey for the World Bank was designed and implemented by Agrifood Consulting International. 23 The survey covered the 2012 wet season and the 2013 dry season, or the 2012/13 marketing season. 24 Broken down by 44 percent of wet season rice production and 53 percent of dry season rice production. 18 management, using: (i) purposive sampling based on information from initial meetings with MEF, SNEC, and MAFF and an initial identification of respondents; and (ii) snowball sampling, which consists of interviewing key informants identified by the respondents themselves. Information was elicited from key informants on: (i) changes in the past 10 years in the agriculture sector, their driving forces, key constraints to improved productivity, and stakeholders’ perspectives for the next five years; and (ii) prices, farmers’ use of inputs including land, and agricultural production. 50. At the village level, one FGD was conducted for each selected crop . The FGD questionnaire elicited: (i) the same information collected from key informants but at the village and farm level, and the impact of changes on farm activities; and (ii) more detailed questions on prices, yields, and use of inputs for agricultural production. At the farm level, data were collected from individual farmers on: farm budgets for the selected crops; qualitative information on changes observed by farmers in the past 10 years on their farms; their perspectives for the next five years; and farm and farmers’ characteristics. 3.2. Crop Production 51. Cambodia’s agricultural landscape diversified over the past 10 years. The cultivated crop area under paddy increased from 2.14 million ha in 2002 and 2.97 million in 2011 (Table 15). But the cultivated area under other crops, especially cassava, increased even more (recall Table 4), resulting in the relative decline of the area under paddy in total cultivated area, from 86 percent in 2002 to 74 percent in 2011 (Table 15).25 Thus, increasingly more farmers are growing non-rice crops compared to the situation 10 years ago. Table 15: Diversification in cultivated area, Cambodia, 2002-2011 Indicators 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 Cultivated crop area 2,473 2,718 2,823 2,939 3,111 3,199 3,321 3,519 3,730 3,990 (‘000 ha) Cultivated area under paddy (‘000 2,137 2,314 2,374 2,444 2,541 2,586 2,616 2,719 2,796 2,969 ha) Share of paddy in total cultivated area 86 85 84 83 82 81 79 77 75 74 (%) Source: MAFF 2014b. 52. More generally, there is strong evidence that Cambodian agriculture is both specializing and diversifying. As Figure 5 shows, agricultural GDP is increasingly dominated by crops. At the same time, within crops there is diversification away from paddy towards other crops. The FGDs reported that the number of wet season rice producers had reduced in 5 villages among the 36 visited (Figure 9.1). Non-paddy crops grew faster than paddy. A shift toward vegetables and upland crops such as maize and cassava was reported by several villages. In one case, a vegetable crop grown 10 years ago was abandoned by farmers. 25 Note that FAOSTAT reports a larger cultivated area in 2002 and a much lower rate of land expansion. Total cultivated area in 2002 was 3,700,000 ha and it grew to 4,000,000 ha in 2011, according to FAOSTAT. 19 Figure 9: Crop production and cultivated land areas, Cambodia 1. Likelihood of changing crops produced in 2. Change in the average cultivated area, 2008- each village 2012 Source: FGDs. Source: Individual farmers’ surveys. 3. Change in the area of cultivated land and 4. Reasons for the change in cultivated land average farm size area Source: FGDs. 53. Farmers diversify towards different crops because of higher profitability, existence of a stable market, cultural reasons, and ripple effects. Indeed, farmers’ reasons for diversifying depend on the type of crop. For example, two out of three farmers chose to grow vegetable and cassava and, to a lesser extent, maize because they were profitable. This finding is supported by the results of the farm budget analysis: farmers had gross margins of up to $505/ha for cassava, $303/ha for maize, and $1,393/ha for vegetable production compared to $295/ha for dry season rice and $245/ha for wet season rice (Table 26). According to the FGD results, more than 80 percent of farmers grew rice for cultural reasons: they were used to doing it and their parents grew rice. Ripple effects were also identified as a reason to diversify to other crops or to adopt new technologies such as improved varieties. Analysis of the data from the FGD shows that regardless of farm size and level of technology use, the reasons to diversify remained similar. 54. Farmers are increasingly moving from the production of nonaromatic to aromatic rice. The 2013 Agricultural Census estimates that one out of ten farmers in Cambodia plants his/her land with aromatic paddy rice. Yet several years ago this share was much smaller. Geographically, aromatic paddy is mostly produced in the Tonle Sap, Plain, and Plateau regions 20 (Figure 10). Meanwhile, traders and MAFF officials estimate that about 30 percent of the wet and dry season crops are of fragrant rice. Figure 10: Regional distribution of aromatic paddy production (% of farmers), 2013 Source: NIS 2014. 55. An important source of growth in agricultural production, irrespective of crop, is land expansion.26 According to MAFF, land expansion contributed to 42 percent of the increase in rice production from 2003 to 2012 (Figure 11). The contribution of land expansion for vegetables, maize, and cassava was 37 percent, 76 percent, and 84 percent, respectively. Cultivated maize areas grew two-fold, from 93,000 ha to 215,000 ha, and production increased three-fold, from 314,000 tons to 950,000 tons. The changes were even more spectacular for cassava: its cultivated areas multiplied by 13 times, from 25,000 ha to more than 337,000 ha, corresponding to a 23-fold increase in production, from 330,000 tons to more than 7.65 million tons. 56. This large agricultural land expansion shown in the national statistics is confirmed by the 2013 farm survey. It indicates that cultivated areas especially increased for upland crops (cassava and maize) (Figure 9.2). This result, combined with the change in the proportion of farmers growing non-rice crops, led to a boost in upland production. 57. There are also indications that the distribution of farm size is becoming more unequal. Figure 9.3 indicates that about half of the FGDs experienced an increase in land area in the past 10 years, regardless of the type of land (irrigated and upland). Respondents reported an increase in average farm size, mostly for medium-size and large farms. On the other hand, half of the respondents reported a reduction or at least stagnation in farm size, especially for small farms. 58. Individual farmer interviews indicated an increase in average farm size over the past five years (Table 16). In many cases, farmers expanded their cultivated upland crops. Areas under maize and cassava increased about two-fold. FGD participants identified RGC policies in expanding infrastructure such as feeder roads and social land concessions as the main contributing factor to upland expansion (Figure 9.4). Improving access to low-population density regions increased the availability of land for agricultural production. Expansion of rice and vegetable 26 Even in the short run, total agricultural area (arable and permanent crop) increased from 3.81 million ha in 2010 to 4.50 million ha in 2013, according to MAFF. 21 cultivated areas as shown in Table 16 was relatively modest, reportedly due to the constraints in availability of good land for vegetable production, technical complexity and labor requirements for vegetable production, and limited irrigation expansion. Figure 11: Contribution of land and yield to average agricultural production growth, 2003- 2012 Source: MAFF 2014b. 59. Higher agricultural prices and better access to credit provided strong incentives for farmers to expand their cultivated land areas. Between 2005 and 2013, the farm-gate price of maize increased by 64 percent, from $140/ton to $230/ton; and even more staggeringly, fresh cassava prices skyrocketed 200 percent, from $20/ton to $59/ton between the two years (Table 90).27 Similar increases were observed for rice, whose prices went up by 90 percent, from $124/ton to $237/ton. These higher prices acted as incentives for farmers to increase their cultivated land. RGC’s policies to spread microfinance institutions and improve access to credit also contributed to the expansion of cultivated land areas. 60. Crop yields increased significantly in the past 10 years as well (Figure 11). Among the four selected crops, maize had the highest change (58 percent), from 2.8 tons/ha to 4.4 tons/ha; cassava yield increased by 44 percent, from 15.1 tons/ha to 21.7 tons/ha, according to the Production Statistics of MAFF. The same positive changes were observed for wet and dry season rice, from 1.9 tons/ha to 2.9 tons/ha for wet season rice and 3.2 tons/ha to 4.4 tons/ha for dry season rice. In the 2010-2013 FAOSTAT data on yields, Cambodia (with Indonesia) had the highest 27 According to 2005 survey and 2013 FGDs. 22 regional yield for cassava, reaching 22 tons/ha (Table 17). On the contrary, it had the lowest average rice paddy yield at 3.0 tons/ha, 80 percent lower than yields in Vietnam and lower than in Thailand (Box 2). Maize yields were comparable to the regional average. Most of FGDs and interviewed experts concurred with the statement that crop yields have increased in the past 10 years. Table 16: Change in farm size, Cambodia, 2008-2012 Average Size of Individual Farms (ha) By Province 2008 2009 2010 2011 2012 Kampong Cham 1.28 1.47 2.18 2.61 3.38 Kandal 0.78 0.86 0.98 0.95 0.96 Takeo 0.63 0.67 0.82 0.83 0.85 Kampot 0.68 0.63 0.64 0.63 0.64 Battambang 3.00 3.37 3.65 4.38 4.84 Bantey Meanchey 5.17 4.80 4.17 5.69 7.92 By Crop Paddy/ Rice 1.89 1.96 1.95 2.06 2.11 Cassava 3.92 3.75 4.33 5.92 7.72 Maize 2.57 2.80 2.88 3.65 4.84 Vegetables 0.60 0.61 0.65 0.62 0.70 By Farm Size Small 0.99 0.73 0.84 0.87 0.88 Medium 1.55 1.83 1.77 2.10 2.38 Large 3.61 3.78 4.14 5.29 7.03 By Technology Modern 2.97 2.85 2.81 3.63 4.65 Traditional 1.02 1.21 1.48 1.61 1.85 Total/ Overall 1.94 1.99 2.11 2.57 3.18 Source: Individual farmers’ survey. Table 17: Average crop yields in selected countries (tons/ha), 2010-2013 Population Country Cassava Maize Rice Paddy Density* Cambodia 22.0 4.1 3.0 256 Indonesia 21.1 4.7 5.0 444 Philippines 10.2 2.8 3.8 782 Thailand 20.5 4.3 3.0 316 Vietnam 17.7 4.3 5.5 809 Note: *Population density is defined as number of people per km2 of agricultural land in 2010-2011. Source: FAOSTAT 2014 data for yields and WDI 2014 data for population density. 23 61. While Cambodia lags behind in rice yield, its average yields for other crops are high in a regional comparison. This is a very positive outcome of the last decade of agricultural transformation given Cambodia’s relatively low population density (see the last column in Table 17). Countries with high population density tend to have higher yields due to constraints on land expansion. Cambodian farmers, however, have managed to achieve yields comparable to those of much more populated countries. Box 2: Rice Yields in Cambodia How large is the rice “yield gap” in Cambodia? Its average yield looks similar to that in Thailand, so the yield gap may seem small. Yet national averages can be deceiving. A large proportion of Thailand’s rice production is of high-value jasmine and glutinous varieties, which inherently have lower yields than conventional varieties. In the Northeast of Thailand (where nearly all jasmine rice is grown, along with glutinous varieties), the average yield of the main crop is 2.3 tons/ha. The more suitable comparator with Cambodia is the wet season crop in Thailand’s Central region , where conventional varieties dominate. Average yield there was 3.8 tons/ha over the 2011-2013 period. This would imply a yield gap of about 30 percent. But what about productivity on Cambodia’s best paddy fields? Does that match the situation among peers and competitors? The answer seems to be “no.” Even in the best managed irrigation schemes in Cambodia, yield may reach 4 tons/ha, rarely more than that. For comparison, in the better growing areas of Thailand Central Region, farmers get 5.7 tons/ha in dry season (Bordey et al., 2014). In Vietnam’s Mekong Delta, average yield at a provincial level is between 6 -7 tons/ha and some smallholder farmers obtain yields well above this. These high-yielding provinces account for most of Thailand’s and Vietnam’s rice exports. This situation implies a very large yield gap, even at the top end of Cambodia’s production. 62. The majority of farmers interviewed in 2013 believed that there is scope for further improvement in crop yields More than half of the interviewed farmers believed that an increase in yield is likely, especially for rice and vegetables, and somewhat less likely for cassava and maize (Figure 12.1 and Figure 12.2). Their expectations about the change in cultivated land areas in the next five years, however, were more conservative. More than half of interviewed farmers expected no change in the area of cultivated land for vegetables and maize (Figure 12.3). About 10 percent predicted a decrease, and one out of three anticipated an increase. Cassava showed a peculiar pattern: almost equal numbers of farmers expected an increase, no change, or a decrease in cultivated area. The response was determined by location – respondents in the Tonle Sap and Mekong regions were not positive about the possibility of increasing cultivated land. In other regions, the option of land expansion is still open, especially for upland crops. Interviews with different stakeholders revealed that many people were aware of the encroachment of cassava crop cultivation into forest land; this is a relatively easy way to increase agricultural land but is receiving increasing attention from policy makers since it puts more stress on already fragile ecosystems. For rice, the overall expectation was “no change,” although more than 20 percent of respondents expected a decrease in cultivated area. 63. Most farmers, particularly small ones, were concerned about the impact of limited land expansion on their agricultural production. More than half of the interviewed individual farmers thought that limited land expansion would have a major negative impact on their production, and as expected, small farms worried most, with two-thirds of them identifying the difficulty of expanding agricultural landholding as a major concern (Figure 12.4). Most of these 24 were rice and vegetable growers. The issue was assessed as not critical by cassava and maize farmers. Figure 12: Projections for land expansion and yield changes, Cambodia, 2013 1. Perspectives on changes in crop yield at the 2. Perspectives on changes in yield by crop at village level (# of FGDs) the farm level Source: FGDs. Source: Individual ffarmers’ surveys. 3. Perspectives on the possibility of change in 4. Perspectives on the impact of limited land cultivated areas by crop expansion on production Source: Individual farmers’ surveys. 64. To address the constraints and reduce the negative impact of limited access to land, respondents made several suggestions:  Increase public investment in infrastructure, especially irrigation and feeder roads. Investment in irrigation would allow farmers to have more control over the agricultural production cycle, thus intensifying the crop cycle, which would increase cultivated areas. The feeder roads would unlock access to available agricultural land in remote areas, thus increasing the possibility of land expansion. These suggestions were made mostly by farmers in Takeo and Kampot provinces.  Strengthen technical assistance to farmers, either through the government or private agents. Limited land expansion may be compensated for by intensification on existing agricultural land. Farmers may produce a similar total quantity by intensification, land expansion, or a combination of both. Better extension service would support farmers to attain higher productivity. 25  Continue and expand the RGC social land concession policy to small farms. This third suggestion focused on changing the policy to ensure that even smallholder farmers can benefit from it and increase their landholding. Respondents identified the suggestion as possible but it needs policy support to be actionable. This response was quite common from respondents in Kampong Cham, Battambang, and Bantey Meanchey provinces. 3.3. Farmland Private land transactions 65. Private land markets were very active over the past 10 years. Results from the FGDs indicate that farmers bought and sold land for a number of reasons, including family needs and external incentives (Figure 13). Transactions were fueled by the higher agricultural prices received by farmers, better access to credit, and the development of public investment in infrastructure. Higher agricultural prices gave farmers higher incentive to invest in land as this is the main asset for agriculture. With better access to financial services over the years, farmers were able to spread land investments across several years through medium-term loans. Government investment in agricultural infrastructure such as irrigation and rural roads was also identified by farmers as a factor that helped boost agricultural land transactions (Figure 13.1). 66. In most cases, farmers who sold land faced social and economic distress (Figure 13.2). Two out of three FGDs reported that farmers who sold land were in default in reimbursing loans from microfinance institutions (MFIs) and banks; about half reported death or serious illness of family members as a reason for selling land. About one-third identified other reasons, such as the need for cash for family obligations (e.g., marriage) and change of activities (e.g., family moving out of agricultural production). 67. People who bought land were farmers expanding their farm size. Almost all FGDs reported that other farmers were the main buyers of land, either to expand their own agriculture or with the expectation of leaving more agricultural land to their offspring (Figure 13.3). In some cases, farmers from other villages were the buyers. The RGC policy on land distribution has improved the access to land to some landless and very small land owners. For example, the Land Allocation for Social and Economic Development Project supported by the World Bank targets poor landless individuals and allocates about 0.1 ha for housing and 2-5 ha for agricultural production. Yet many farmers already established in rural areas are not reached by these operations. 68. The observed agricultural land transactions are likely to result in: (i) increased average farm size; (ii) lower labor/land ratios; and (iii) an increased gap between small and large farmers: a. In the long term, because of their agricultural land transaction strategies, large farmers would accumulate more land with better land quality. According to the FGDs, such farmers sold small plots and those with poor soil fertility and subsequently bought larger and better quality plots for the sake of building up their farm asset. Sellers of these lands often become landless or subsequently owned farms with fragmented land (which may face more difficult farm management) or poor soil fertility (which will result in lower agricultural profitability). b. A further consequence is the increase in the extremes: at one end of the spectrum, there are more very small farms struggling to make an agricultural livelihood economically viable, 26 while at the other end, there are more large farms with better quality assets, able to benefit from new technologies for their agricultural production. “Large” includes farms with agricultural landholding exceeding 3-5 ha, but excludes industrial producers with hundreds to thousands of hectares of cultivated land. Even so, this results in a wider gap between small and large landholder households in rural areas. c. An agricultural landscape dominated by large farms would have the capacity to rapidly move to commercial and productive agriculture from the current agrarian structure. Because of their relatively high volume of production, larger farms can be more easily linked to markets, have lower constraints in terms of access to finance, and are able to make profitable the use of agricultural machines. Figure 13: Private land markets, Cambodia, 2013 1. Reasons for intensive private land markets (# FGDs) 2. Reasons for selling land (# FGDs) 3. Reasons for buying land (# FGDs) Source: FGDs. 69. Landless or marginally landless rural households’ coping strategies include different options. People who sell their land often become farm laborers, move to off-farm activities in rural and urban Cambodia, or migrate to neighboring countries. The decision is determined by several factors, including their level of education and the overall economic environment in Cambodia, especially the capacity of nonfarm sectors to absorb labor from agriculture. The FGD findings indicate that agricultural transformation (the process by which agriculture engages less people and substitutes labor for capital) is taking place in Cambodia. On average, hired labor income on farm activities increased by 2.7 times from 2005 to 2013, from $1.25/day to $4.56/day, or $109/month for 24 days of work per month. In spite of the past growth, the average farm wage in Cambodia remains low in the regional comparison (Table 18). Off-farm wage, represented by prevailing monthly salaries at garment factories, averaged $159/month based on the FGDs, or 45 percent higher than agricultural labor income. More than 80 percent of FGD respondents reported that 27 hired labor is increasingly difficult to find, thus the expectation was that rural wages would continue to increase, and farmers must take that into account in their agricultural production operations. Competition from neighboring countries also fuels the exodus of rural laborers. In 2013, the minimum off-farm wage in Thailand was around $10.30/day, very similar to the rate in Malaysia ($9.80/day) and the Philippines (ranging from $6.80/day to $11.10/day depending on location). Table 18: Wages of labor hired for rice production, selected countries, 2013 Country Daily wage, US$ China 19.3 Thailand 16.5 Vietnam 8.9 Philippines 7.6 Cambodia 4.5 India 4.2 Myanmar 2.4 Note: Dry season rice production. Source: Bordey at al. 2014; data for Myanmar are from the World Bank. Land prices and rentals 70. Irrigation adds value to land. Between 2005 and 2013, increases in land prices showed significant variation across land types. The price of irrigated land increased by 7 times, up from $1,136/ha to $8,182/ha, while prices of rainfed lowland and upland were up more than 2.5 times, from $888/ha to $3,294/ha for rainfed lowland and $945/ha to $3,294/ha for upland. These figures were obtained from commune council interview respondents (Table 19). The higher price of irrigated land was related to its location, particularly in those irrigated schemes where the government concentrated other investments, such as in roads and energy. Irrigated land also offered the option of growing multiple crops over the year, resulting in higher land productivity. Table 19: Changes in land prices between 2005 and 2013, Cambodia 2005 2013 Change in Price ($/ha) ($/ha) (%) Purchase price of 1 ha of land Irrigated land 1,136.4 8,182.0 620 Rainfed land 888.7 3,294.0 271 Upland 945.2 3,294.0 248 Rental price for 1 ha of land Irrigated land 73.5 225.0 206 Rainfed land 63.8 187.0 193 Source: 2005 survey and 2013 commune council interviews. 71. Rental prices of land did not follow the same pattern of high increase as the sale prices of land. Between 2005 and 2013, rental costs for irrigated land increased 200 percent, from an average $73.5/ha to $255/ha. The rental price for rainfed land followed the same pattern. If agricultural production was profitable, then only farmers who faced significant constraints would 28 rent out their land. The rental prices were below the gross margins calculated in the farm budget analysis. For example, with irrigation, modern farmers got up to $442/ha growing dry season rice, a margin that represented almost twice the rental costs for irrigated land ($225/ha). For rainfed rice, farmers got up to $399/ha, again more than twice the income received if rainfed land was rented out at $187/ha. The difference in rental prices and returns to land may be caused by the difference in land quality. Farmers may only rent out low fertility land or land with difficult access, resulting in increased transportation and transaction costs. 3.4. Agricultural Labor Force 72. Cambodia continues to have too many people engaged in agricultural activities, but the last decade witnessed a significant reduction in the agricultural labor force. According to the Social Economic Surveys of NIS, 51 percent of Cambodia’s labor force worked in agriculture in 2012, representing 3.9 million workers out of 7.7 million in the total labor force. In 2004, it was 57 percent. 73. The field evidence confirms that farm labor is being replaced by mechanization services. In 2005, one hectare of wet season rice production required 73 working days. In 2013, labor use declined to 48 days, or by 34 percent (Table 20). Labor use also declined for maize and dry season rice, but increased for vegetable production, which in principle is hard to mechanize. Table 20: Use of labor in farm production by crop, Cambodia, 2005 and 2013 2005 2013 (days/ha) (days/ha) Change Family Hired Total Total (%) Paddy (wet season) 30.30 42.70 73.00 48.29 -34 Paddy (dry season) 24.60 48.70 73.30 27.75 -52 Cassava 13.20 35.20 48.40 48.80 +1 Maize 15.60 26.20 41.80 31.33 -25 Vegetables (mix) 79.80 60.10 139.90 169.85 +21 Source: 2005 and 2013 surveys. 74. There is the scope for further labor use reduction in Cambodia. While its farmers used less labor than in the Philippines, India, and Myanmar, in China and Thailand dry season rice production in the main producing areas requires 10-11 man-days/ha, almost 5 times less than in Cambodia (Table 21). In Vietnam, dry season paddy production requires half of Cambodia’s labor requirement. The lower use of labor in these countries is driven by their higher wages (Table 18), which in turn accelerate mechanization. Mechanization has also begun in Cambodia, and as the experience of other countries shows, the rate of labor replacement by capital will accelerate with further wage increases. 29 Table 21: Use of labor in dry season paddy production, selected countries, 2013 Country Man-days/ha Thailand 10 China 11 Vietnam 23 Cambodia 48 Philippines 69 India 78 Myanmar 110 Source: Bordey et al. 2014 and World Bank estimates for Cambodia and Myanmar. 75. Mechanization and migration affected the structure of rural labor, especially that of women, for both family and available hired labor. Results from the FGDs indicate that women historically played a critical role in agriculture but this role will be less important in the future (Figure 14). The anticipated change is not significant but conveys households’ concern about the increasing difficulty associated with the availability of labor. Two factors are in play: (i) the migration of young women to work in garment factories and young men to work in the construction sector; and (ii) the shift of some agricultural operations from manual to mechanized. Figure 14: Change in the perception about women’s role in agriculture, Cambodia, 2013 Source: FGDs. 76. Young women are attracted to the relatively high, regular pay from garment factories. From the FGD results, the average monthly wage at factories was about $162.5/month, 50 percent higher than the calculated monthly salary for hired workers at $4.56/day, and equivalent to $109/month based on 24 days of work. In addition, garment workers have the convenience of regular and continuous monthly wages. Young men are attracted to construction work, often located in urban areas. Data from the 2013 key informant discussions and the 2005 survey suggest a wage increase of 67 percent, from $72/month in 2005 to $120/month in 2013. Though lower than the wage received by most women working at garment factories, this was still about 10 percent above the agricultural wage. The gap between the agricultural and construction wages, however, declined: in 2005, the ratio was about 1 to 3 in favor of construction. Nonetheless, both agricultural and construction work often require daily tasks, implying that employers do not comply with 30 current labor law, because this kind of wage is not determined under Cambodian labor law. The wage is made by an agreement between the service provider and the employer. Mechanization also contributes to the anticipated decline in the role of women in agriculture. Harvest and post-harvest activities, which previously often involved tasks directed towards women, are currently executed using machinery with male drivers (Box 3). Box 3: Gender Dimensions of Cambodian Agriculture As Cambodian women are increasingly finding alternative employment in the garment sector and other nonfarm sectors, there is increased demand from women for new services in rural areas such as a savings market for remittances and child care options. For women staying in agriculture, access to extension and other agricultural services remains a large constraint to increased productivity, much larger than for men. Women’s needs are often neglected, partially due to the low number of female extension workers and the lack of training and information tailored to women. Mechanization also creates new challenges as female-headed households have less ability to control providers of mechanized services. And hazardous weather events disproportionally impact female-headed households, as their exposure to trainings and programs for disaster risk management remains limited. Source: World Bank 2015. 77. Almost all FGD respondents agreed that migration of young rural workers would continue and the subsequent lack of labor force would negatively affect future farm production and farm budgets. Rural youth leave farms and rural areas for other employment opportunities (Figure 15). However, FGD participants believed that households would benefit from migration through an increase in remittances. Increases in working capital and investment in agriculture and off-farm activities are examples of positive impacts of migration. Nevertheless, respondents did not agree with the statement that young workers leaving the rural households may become a source of good practices and useful information on nutrition and health of children. The same pattern was true for the perceived capability of returning migrants to provide advice on agricultural techniques or commercialization of agricultural products. 31 Figure 15: Impact of youth migration on agricultural households, Cambodia, 2013 Source: FGDs. 3.5. Use of Agricultural Inputs 78. Most farmers seemed to be aware of the type of inputs that could boost agricultural productivity. Improved seeds, urea, insecticides, and herbicides were adopted by more than 50 percent of farmers in several villages (Figure 16.1). Manure, compost, and other fertilizers such as DAP and NPK had lower levels of adoption. 79. High agricultural prices were the main driving force behind the high adoption of modern inputs. In order of importance, the reasons driving the high adoption of modern technologies were identified as: the high prices of agricultural products; better infrastructure; and better access to finance (Figure 16.2). The private sector is very active in supplying inputs in all parts of the country. Farmers adopt new technologies based on their profitability expectations, which take into account the perceived production and price risks. Data from the 2005 and 2013 surveys show that cassava prices increased 200 percent, from $19.79/ton to $59.38/ton; maize prices increased 64 percent, from $140/ton to $230/ton; wet season rice prices rose from $124/ton to $237/ton and dry season rice prices from $118/ton to $210/ton; and vegetable prices increased 60-233 percent depending on the type of crop. The expectation of higher prices for the following agricultural year led more farmers to adopt modern technologies targeting higher yields. Additional use of modern inputs was also a result of higher application rates. Initiatives from private firms are discussed in Section 3.7 (extension services), and were undoubtedly a reason for higher use of modern inputs. Better access to finance resolves farms’ financial gap during the plantation season. Often, the lean season coincides with the cultivation period, which in the absence of adequate financing would result in lower use of agricultural inputs. 80. Farmers identified high price of inputs as the main constraint to adoption of modern inputs. From 2006 to 2013, prices of urea increased by 86 percent (from $0.17/kg to $0.66/kg) and prices of DAP increased by 96 percent ($0.18/kg to $0.74/kg). The most significant increases were observed for improved seeds: around four-fold for rice (from $0.16/kg to $0.66/kg) and around six- to seven-fold for maize (from $0.6/kg to $4.0/kg). For common fertilizers, a valid assumption is that these prices are for the same quality of products (urea, DAP). However for 32 improved seeds, the very high price increases are likely the result of both a change in commodity prices and a change in seed quality. Figure 16: Access to inputs, Cambodia, 2013 1. Importance of type of inputs by technology 2. Reasons for use of agricultural inputs adopters Source: FGDs. 81. Given the simultaneous increase in input and output prices, using higher quantities of modern inputs to get higher yields does not automatically result in higher gross margins per hectare. Indeed, modern inputs, farm assets, and farmers’ skill together constitute a package for higher yield, and the absence of any one of them precludes attainment of the expected results. For example, an adequate rate of chemical fertilizer application would not lead to the expected increase in yield if farmers did not have a minimum level of irrigation control. Similarly, poor weed control would preclude farmers from getting higher yields even with the use of fertilizers. A few cases identified from the farm budgets collected during this survey illustrate the situation where modern farmers with a higher use of inputs had a lower gross margin. For example, a modern wet season rice producer spent $132/ha on fertilizers and reported gross revenue of $499/ha and a gross margin of $132/ha. In contrast, traditional wet season rice farmers spent $39/ha on fertilizers and reported an average gross revenue of $474/ha with a gross margin of $152/ha. Seeds 82. Individual farmers’ use of improved seeds increased over the past 10 years. For the four selected crops, farmers’ use of improved seeds increased continuously at a relatively high growth rate (Figure 17.1). Indeed, improved seeds are often one of the most cost-effective technologies to boost productivity. Cambodia can take advantage of the research results in neighboring countries with similar agro-ecological characteristics and the same market preferences. For example, several new rice varieties developed in Vietnam or Thailand can be grown in Cambodia without huge investments in research or seed multiplication.28 The same applies to other crops such as cassava and maize. However, Cambodia needs more adequate border control and effective quarantine to ensure safe introduction of new varieties of crops and 28 Although this might be the source of intellectual property rights litigation in the case of use of seeds developed by companies in Thailand or Vietnam. 33 compliance with intellectual property rights and plant variety protection (World Bank 2011 and IFC 2014). Figure 17: Access to improved seeds, Cambodia, 2013 1. Change in the use of improved seeds by crop 2. Sources of information on improved seeds 3. Perceived impact of improved seeds on 4. Perspectives on the use of improved seeds by agricultural productivity crop Source: Individual farmers’ survey. 83. Farmer-to-farmer was the most frequently used source of information about adoption of improved seeds. About two-thirds of interviewed individual farmers reported adoption after discussion with and/or seeing the impact of improved seeds on other farmers’ plots (Figure 17.2). Extension agents from different sources (RGC, NGOs, and private suppliers) occupied the second position, with many of them using a farmer-to-farmer approach to increase outreach. Extension services need to provide more support in the future because few farmers were aware of the existence of new crop varieties on their own. 84. Cambodian farmers seem to have started using higher seed quality. The increasing use of improved seeds and the amount spent by farmers to buy seeds seem to contradict the conclusions from technicians that farmers still face limited availability of good quality seeds (ADB 2014a; World Bank 2011). For example, in 2013, cassava producers spent $228/ha on cuttings; maize producers spent $76/ha on seeds; and on average, dry season rice producers spent $108/ha on seeds. However, because farmer-to-farmer is the preferred way to adopt seeds, the quality of some seeds is indeed not high because it is often obtained from relatives and other villagers, or used from the previous harvest for planting the following year. 34 85. About half of the surveyed farmers believed that the use of improved seeds would increase productivity by more than 20 percent (major impact); the other half believed that the use of improved seeds would increase productivity between 10-20 percent (moderate impact). The trends were similar for rice, cassava, and vegetables, with slightly less than half of farmers expecting a major impact and the rest anticipating a moderate impact (Figure 17.3). However for maize, two out of three farmers were quite confident about the productivity impact of using improved seeds. Stakeholders reported that the use of hybrid seeds was common for maize farmers. The supply chain has the advantage of support from large importers, who provide a minimum level of technical advice on the use of hybrid maize seeds. 86. Farmers were very positive about the continued use of improved seeds over the next five years. Two out of three interviewed individual farmers claimed they would continue to use them; the remaining claimed they were likely to use improved seeds in the next five years (Figure 17.4). However, a small group of cassava farmers was unsure whether they would use improved seeds. Farmers suggested that the RGC should increase access to improved seeds through strengthening of technical assistance from MAFF and provide quality seeds at a reasonable price. Manure/Compost 87. Manure and compost use did not change significantly over the past 10 years. The trends were similar for all four selected crops (Figure 18.1). Often, farmers reserved the use of organic fertilizers for vegetables (almost all farms used manure or compost) and rice production (above half). Non-reconstitution of organic matter on upland cultivation, especially cassava, was a big concern. With land expansion, cassava is cultivated from massive clearings of forest land, which in most cases is still rich in organic matter, reducing the need for manure or compost. Cassava can grow in different types of soil even with little or no additional fertilizer. However, in the medium-to-long term, Cambodia could face productivity issues linked to soil fertility depletion. 88. Availability of manure may be a challenge in the future. Farm budgets collected in 2013 show that farmers bought manure for agricultural production and that expenditures were as high as $300/ha for vegetable producers, representing about 17 percent of their total variable costs. The share went as high as 25 percent for small modern watermelon producers. The drive towards mechanization might further contribute to reducing the stock of cattle used for land preparation and as a source of manure. 35 Figure 18: Use of manure, Cambodia, 2013 1. Change in the use of manure/compost by crop 2. Sources of information on the use of manure and compost 3. Perceived impact of manure and compost on agricultural productivity Source: Individual farmers’ survey. Chemical fertilizers 89. Most farmers used fertilizer for rice and vegetable production. The FGDs reported an increase of about 30 percent in the number of users of urea for rice production over the past five years. The change was relatively lower for vegetables and maize, and farmers barely used chemical fertilizers for cassava production (Figure 19.2). The average expenditures from the 2013 farm budget data show that farmers used 153 kg/ha of fertilizers for dry season rice, 66 kg/ha for wet season rice, and 193 kg/ha for vegetables. The results also display that farmers barely used chemical fertilizers on maize (30 kg/ha in 2013) and almost none on cassava (8 kg/ha). 90. Over time, the use of fertilizers per hectare increased. For example, for dry season rice production, the rate of use of chemical fertilizers increased 50 percent for medium-size farms (89 kg/ha to 127 kg/ha) and four-fold for large farms (40 kg/ha to 192 kg/ha). Other types of fertilizers such as DAP received increasingly more attention from farmers. The FGD results show a steep increase in the number of villages where farmers used DAP. A breakdown by crop shows that farmers used fertilizers mostly for rice and vegetable production (Figure 19.3 and Figure 19.4). 36 Figure 19: Use of fertilizers, Cambodia, 2013 1. Sources of information on the use of 2. Overall trends in fertilizer use chemical fertilizers 3. Change in the use of urea 4. Change in the use of DAP 5. Perspectives on the number of users of 6. Perspectives on the intensity of use of chemical fertilizers fertilizers (quantity per hectare) Source: Individual farmers’ survey. 91. The rates of use of fertilizer for dry season rice, cassava, and maize production increased. Fertilizer use for wet season rice, however, did not show a similar increase. Table 22 illustrates the changes in the use of chemical fertilizers between the 200529 and 2013 surveys. 29 The 2005 survey was conducted by the Agribusiness Institute of Cambodia (ABiC), which was used for the World Bank’s Cambodia Agrarian Structure Study and is used in this report for comparisons with t he 2013 survey. 37 Table 22: Illustrative use of fertilizers, Cambodia, 2005 and 2013 2005 2013 Farm Size Farm Size Crop Small Medium Large Small Medium Large Wet season rice (kg/ha) 53 94 93 72 65 63 Dry season rice (kg/ha) 133 89 40 133 127 192 Cassava (kg/ha) - - - 12 - 8 Maize (kg/ha) 35 - - 74 7 9 Source: 2005 and 2013 surveys. 92. In spite of the increased use, the rate of fertilizer use in Cambodia is still below the regional average. From FAOSTAT data, average fertilizer nutrient consumption for Cambodia was about 15 kg/ha30 in 2011, low compared to Vietnam’s 160 kg/ha and Thailand’s 126 kg/ha (Figure 20). More recent surveys indicate a higher use of chemical fertilizers in Cambodia, but still below the level of regional neighbors.31 Figure 20: Trends in chemical fertilizer use (kg nutrient/ha), select countries, 2005-2011 Source: FAOSTAT 2014. 93. In addition to unfavorable input/output prices and lack of knowledge, the low average level of application may be caused by farmers’ adverse selection due to a high risk of obtaining tainted, low-quality fertilizers in the marketplace. Based on their past purchases, farmers may not be willing to pay the market price for the perceived low-quality fertilizers available in the market. While there are cases of low-quality and sometimes fake fertilizers sold in Cambodia, it does not seem to be a major issue (IFDC 2010). The trend is increasing both in the number of farmers using chemical fertilizers and the amount of fertilizers used per hectare. About three out of four rice and vegetable farmers expected the number of farmers using chemical 30 Calculated as kg of fertilizers per ha of arable land. 31 Field data from a recent (2013) feasibility study conducted by CelAgrid for ADB (TA7833) reported that the average application of chemical fertilizer is 122 kg/ha, 96 kg/ha, and 98 kg/ha, respectively, in rice, vegetable, and corn fields. 38 fertilizer to increase in the next five years, and about half anticipated an increase in the dose per hectare (Figure 19.3 and Figure 19.4). Although only less than half of cassava farmers expected to use chemical fertilizers, the amount used per hectare is expected to increase. The most remarkable changes were anticipated for maize: more than half of the respondents anticipated an increase in the number of farmers using fertilizer and more than two out of three anticipated an increase in the quantity used per hectare. Other chemicals 94. The use of insecticides is increasing in Cambodia. Vegetable production requires higher use of pesticides; the proportion of users topped at nine out of ten farmers. For maize and rice, the number of users also increased but at a lower magnitude: one-third of respondents used insecticides. For cassava, less than 20 percent of interviewed individual farmers used insecticides (Figure 21.1). Farmers anticipated that the number of insecticide users would increase mostly for vegetable and maize production. In 2013, farmers spent up to $82/ha on insecticides, with vegetables having the highest expenditure. The figures were much lower for dry season rice ($12/ha), and very low for cassava, wet season rice, and maize (less than $4/ha). However, farmers expected the number of pesticide users to continue to increase and remain high for vegetable production. This situation raises concerns of food safety in areas where farmers are not fully aware of the dangers linked to pesticide use and due to improper handling by traders. Figure 21: Use of pesticides, Cambodia, 2013 1. Change in the number of insecticide users 2. Change in the number of herbicide users 3. Perspectives on the number of insecticide 4. Perspectives on the number of herbicide users users Source: Individual farmers’ survey. 39 95. The use of herbicides is also increasing, especially for upland crops. Cassava and maize farmers were the most prominent users of herbicides (Figure 21.2). Weed infestation was a major constraint in large-scale upland production; with the increasing difficulty in finding rural workers and the increase in cultivated land areas, farmers tended to use herbicides. The figure depicts a sharp increase in the use of herbicides in the last five years, a trend expected to continue for the next five years, especially for cassava and maize (Figure 21.3 and Figure 21.4). For irrigated rice, manual rotary weed control equipment allows for good weed control and has the advantage of requiring only a few days of labor per hectare. In the farm budgets collected in 2013, expenditures on herbicides ranged from a low of $4.6/ha for wet season rice production to $33.8/ha for cassava. Expenditures on herbicides for maize ($20/ha) and dry season rice ($12.2/ha) were in between. 3.6. Use of Agricultural Machinery 96. The past 10 years witnessed a spectacular increase in the use of agricultural machinery in Cambodia. Changes were observed in land preparation, planting, and harvesting. Data from the Department of Agricultural Mechanization show that the number of combine harvesters increased three-fold from 2006 to 2010, and the number of threshers increased two-fold (Table 23). The 2013 survey shows that increasingly more farmers used tractors (both 4WD tractors and hand power tillers) for land preparation for upland crops (i.e., cassava and maize) and they expected to continue to use this equipment in the next five years (Figure 22.3, Figure 22.4, Figure 22.5, and Figure 22.6). For rice and vegetables, their current use was relatively low (10-20 percent) but farmers indicated that their use would increase. However, the use of agricultural machinery may be constrained by the low purchasing power of smallholders, the high cost of agricultural equipment, and the lack of well-trained operators and mechanics for agricultural machinery. Table 23: Number of agricultural equipment units, Cambodia, 2006-2010 Equipment 2006 2007 2008 2009 2010 Combines 325 399 430 836 947 Tractors 4,247 4,475 4,611 5,495 6,200 Threshers 7,795 8,036 8,237 13,798 14,390 Power tillers 29,706 34,639 38,912 53,220 66,548 Irrigation pumps 127,610 131,702 136,061 164,974 166,633 Source: Department of Agricultural Mechanization. 97. Farmers use different sets of equipment based on their farm size, their agricultural assets and equipment levels, and the availability of services. Manual tractors will continue to be common in rice and vegetable production despite the uptake of regular tractor use. Manual tractors are an intermediate technology between draught animals and tractors, and are highly effective for puddling, preparatory tillage, and seed bed preparation for rice production. They are also the most efficient form of mechanization for fragmented and small land plots. However, farmers thought that the level of use of power tillers would decrease, except in vegetable production. In terms of costs, 2013 data collected from FGDs show that tractor rental for land preparation cost about $60/ha and two out of three FGDs reported the service as relatively easy to find; animal traction cost a bit more, at $71/ha, and only one out of three FGDs said that service 40 providers were easily available. For harvest and post-harvest operations, some services were still hard to get. About half of the FGDs reported having difficulty getting combines to their fields. On the contrary, rice reapers and threshers were relatively more available, with only one out of four FGDs stating difficulty in getting them. Figure 22: Use of agricultural equipment, Cambodia, 2013 1. Number of FGDs reporting the use of 2. Main reasons driving the use of agricultural agricultural equipment in their villages equipment Source: FGDs. Source: Individual farmers’ survey. 3. Change in the use of and perspectives on 4. Change in the use of and perspectives on tractors for land preparation power tillers 5. Change in the use of and perspectives on 6. Change in the use of and perspectives on drum seeders (rice and maize) other equipment for rice production Source: Individual farmers’ survey. 98. Data collected from the FGDs are in line with the increasing trend in the use of agricultural equipment shown by official statistics. Irrigation pumps and manual tractors were among the earliest equipment adopted by Cambodian farmers. The uptake of manual tractor use started about 10 years ago, and increased until late 2010 (Figure 22.1). Irrigation is the key to rice 41 production and power tillers are versatile enough to work on muddy plots. The use of tractors, adapted to medium-size to large farms, also started to increase in the last 10 years. The economic land distribution policy and the expansion of agricultural land, resulting in higher average farm size, will contribute to this continued growth. The trends for harvest and post-harvest equipment use show a typical pattern: rice farmers first move from labor-intensive tasks to intermediate mechanization using rice reapers for harvest and threshers for post-harvest; ultimately they move to a highly capital-intensive approach using combine harvesters. The result is a plateau in the number of threshers and rice reapers and an increase in the number of combines (as illustrated in Figure 22 and Table 23). 99. Combines were often rented by farmers from business entrepreneurs, and the use of combines has taken off. The high purchase price of the machines makes renting the preferred option for small and medium-size farms. Several FGDs reported having access to such services, with an average payment of $95/ha, 10-15 percent higher compared to the use of reapers and threshers for harvest and post-harvest activities (Table 24). However, some farmers complained about the quality of rented combines and reported that their use brought additional work. With poor-quality harvesters and/or unskilled drivers, the harvested paddy grain is mixed with a lot of broken straw, requiring further cleaning processes. Others criticized the extra costs incurred due to inexperienced drivers, who spend too much time maneuvering at the end of the field while farmers pay for the equipment by the hour. Table 24: Average costs of and difficulty finding agricultural machinery, Cambodia, 2013 % of FGDs Reporting Tasks $/hectare Difficulty Finding Equipment (if rental) Land preparation by animal 76.5 68 Land preparation by power tiller 59.8 33 Land preparation by tractor 74.3 20 Rice reaping 43.0 27 Rice threshing using mechanical thresher 40.3 8 Combine harvester 95.5 46 Source: FGDs. 3.7. Agricultural Services Extension services Availability and access 100. In the past 10 years, RGC and NGO extension agents provided technical advice in two out of three villages visited during the FGDs. Commercial firms were also present, but to a lesser extent, in one out of three villages. Technical advice is a source of information on recommended and proven scientific farming techniques. Agricultural extension encompasses several dimensions, including: economic dimensions such as increases in farming income, agricultural productivity, farm financial management, and food preservation, and social dimensions such as improvement in the health of family members, and leadership. Farmers may get technical advice from several sources, comprising government employees, NGOs, and the 42 private sector. More than 10 years ago, the proportion of villages receiving extension services was less than 10 percent on average for all four crops (Figure 23.1). Services from the RGC and NGOs rapidly increased to cover more than half of these villages and then slowed down in the last five years to around 70 percent (25 FGDs out of 36). At the opposite end of the spectrum, private extension services grew at a slower pace in the early 2000s before rising to a higher and more regular level, currently reaching about half of the interviewed villages. 101. Specific crops such as rice and vegetables received more attention from RGC and NGO extension agents. About half of the interviewed farmers reported receiving technical advice on these two crops (Figure 23.2 and Figure 23.3). The RGC agents mostly focused on rice and vegetables and neglected other crops such as maize and cassava, for which extension services were available in only one out of ten villages. NGOs’ extension services also targeted mostly rice and vegetables, but they started to add cassava and maize about five years ago. Private commercial advisors were mostly interested in developing vegetable production and had reached about one out of three interviewed villages (Figure 23.4). Satisfaction, perception of impacts, and perspectives 102. Farmers considered RGC and NGOs’ agricultural extension services more satisfactory than services from private companies. About nine out of ten FGDs32 expressed satisfaction with the extension services from RGC and NGOs (Figure 23.5). On the contrary, services from private commercial advisors were perceived by one out of three FGDs as unsatisfactory. The main complaints were about: a. Difficulty acquiring the inputs required. For example, advice promoting the advantage of a specific new variety of crops was not followed by activities to make the promoted seeds available. b. The insufficient number of visits conducted by extension agents. More frequent visits would help farmers, as they mentioned gradual improvements in their knowledge and skills from repeated discussions with agricultural extension agents. Farmers wanted extension agents to follow-up and monitor their progress after the initial visit. c. Confusion regarding commercial activities and dissemination of technical messages. Complaints were on the type of messages, which in some cases are too focused on selling products. Some respondents even reported that they felt obliged to buy the product after meeting with these agents. Some farmers suggested that the government should intervene when the performance of products sold by extension agents is not as effective as advertised. 32 FGDs with valid responses. 43 Figure 23: Access to extension services, Cambodia, 2013 1. Number of FGDs reporting access to 2. Development of extension service from RGC extension services by type of provider agents Source: FGDs. Source: Individual farmers’ survey. 3. Development of extension service from NGO 4. Development of extension service from the agents private sector Source: Individual farmers’ survey. Source: Individual farmers’ survey. 5. Satisfaction with the quality of extension 6. Impact of lack of access to technical advice service received (# of FGDs) on agricultural production Source: FGDs. Source: Individual farmers’ survey. 103. Farmers reported that extension would be more effective if technical agents had a better understanding of farm characteristics. Respondents reported that in some cases, innovations proposed by technical agents did not correspond to the existing reality on the farm. For example, agents promoted techniques requiring good water control to farmers who did not have irrigated land; or disseminated technical packages that would not be profitable because of high initial investments. Farmers wanted extension agents to ultimately discuss adoption based on the specific assets, soil quality, and fertility associated with their farms. In an ideal situation, such tailored advice would be provided to farmers, but this approach may face high costs of intervention 44 and may only be possible for medium-size and large farms, excluding the more numerous small farmers. In other countries, producers’ organizations are an effective way to alleviate the unit costs of intervention. They have the advantage of focusing on limited types of crops, if not one crop; and can tackle similar problems encountered by most farmers within the area of interventions. However, the effectiveness of farmers’ organizations in Cambodia has been limited so far (see Chapter 3.8). 104. Relatively few farmers were aware of the impacts of the lack of access to technical advice on their agricultural production. Several individual farmers did not provide answers to this question. Depending on the type of extension service received, the rates of nonresponse were about two out of five respondents for those dealing with private firms, and about half for farmers working with RGC or NGO extension agents. There was also a difference across crops. For rice, maize, and vegetables, nine out of ten farmers identified the lack of extension services as having a moderate to major impact on their agricultural activities. Cassava growers, on the other hand, did not see the lack of extension services as a major issue in their agricultural production (Figure 23.6). A consequence of such lack of awareness is farmers’ low willingness to pay for technical advice. This will be an issue if the RGC decides to privatize agricultural extension. Credit Availability and use 105. Financial services from several sources comprising commercial banks, MFIs, community saving groups, and money lenders were increasingly available to farmers . In the past 10 years, increased availability of financial services from MFIs was one of the main changes in rural Cambodia, with the proportion of villages having access to credit increasing from 25 percent to above 90 percent (Figure 24.1). 106. In addition, the types of service providers have diversified with the establishment of commercial banks in small towns and community savings and loans institutions in villages (World Bank and AusAid 2013). For FGD participants, the availability of commercial banks and community savings and loans services increased 10-fold compared to the situation 10 years ago. However, improved access was not strong enough to satisfy all needs; hence, the continued use of financial services from money lenders. 107. Farmers’ choices of financial service provider varied according to the size of their farm and the type of crop produced. Maize and cassava growers used financial services from commercial banks and MFIs. More than half of the interviewed farmers growing cassava and maize reported access to MFIs’ financial services; and one out of three used financial services from commercial banks (Figure 24.2 and Figure 24.3). Most of these farmers were located in Kampong Cham and Takeo. Commercial banks were not used much in the other four visited provinces. On the other hand, commercial banks were barely used by rice and vegetable growers: the proportion of users remained below 10 percent over the past 10 years. These farms, often small, preferred working with community savings and loans institutions. In sum, respondents were almost unanimous in identifying MFIs as the most available and accessible financial service for farmers. For example, ACLEDA, which currently has commercial bank status, started as a microfinance provider and has been the largest microfinance institution in Cambodia for over a decade. 108. Community savings and loans institutions are being implemented in more and more villages, but to date, few farmers have had access to their services. Slightly more than half of 45 the visited villages reported receiving financial services from community savings and loans institutions, most of them in the past five years (Figure 24.1). However, very few interviewed farmers reported using services from this type of local institution, which was supposed to serve farmers’ needs (Figure 24.4). Most users were: rice and vegetable growers; located in Battambang, Takeo, and Kampot provinces; and modern technology users. The Cambodian Community Savings Federation has provided financial and nonfinancial services to low-income families, especially women since early 2000. Cambodian Community Savings Federation is currently working in Battambang and Bantey Meanchey provinces and provides services to more than 3,000 farmers. 109. Despite the existence of numerous formal financial service providers, farmers still take loans from money lenders, presumably because money lenders do not require collateral and higher risk loans can be obtained. Individual farmers receiving services from money lenders increased from 10-15 percent about six years ago to 20-50 percent in 2013. Cassava and maize farmers appeared to rely more on money lenders than did other producers (Figure 24.5). These upland crop growers, however, showed higher access to commercial banks and MFIs. Satisfaction, perception of impacts, and perspectives 110. About nine out of ten FGDs viewed community savings and loans institutions as providing the most satisfactory quality services. MFIs came in second place, with two out of three responses “very satisfied” or “satisfied.” Commercial banks and money lenders occupied the last place, with less than half of respondents reporting satisfaction.33 Satisfaction encompasses several aspects, but the most important constraints often cited by farmers were the high interest rate, the excessive delay between time of application and availability of loans, and the required collateral. For community savings and loans institutions, some farmers criticized the constraints linked to the “social organization,” such as frequent meetings and the limited amount that can be borrowed by each member (Figure 24.6). For several respondents,34 lack of access to credit would negatively impact farm production, regardless of the type of crop, size of farms, or region (Figure 24.7). 111. Only a handful of farmers reported no impact from the unavailability of MFIs and money lenders. However, there were a large number of nonresponses, varying from half of respondents regarding MFIs to about 80 percent regarding commercial banks. In general, nonrespondent farmers could be categorized as those who did not use financial services, either by choice or because of nonavailability of such services. 33 These proportions excluded nonresponses, which in some cases may be large (30 percent nonresponse for extension service from RGC, and 50 percent for extension services from private commercial firms). Responses may be missing because respondents did not receive the services, or respondents did not know the answers even if they had received services. 34 Information based on individual farmer interviews. 46 Figure 24: Access to financial services, Cambodia, 2013 1. Change in the availability of financial services 2. % of farmers receiving financial services from at the village level commercial banks Source: FGDs. Source: Individual farmers’ survey. 3. % of farmers using financial services from 4. % of farmers using financial services from MFIs community savings and loans institutions Source: Individual farmers’ survey. Source: Individual farmers’ survey. 5. % of farmers using financial services from 6. Satisfaction with the quality of financial money lenders services received Source: Individual farmer’s survey. Source: FGDs. 47 7. Perception of the impact of lack of credit on agricultural production Source: Individual farmers’ survey. Access to other services for agriculture Price information 112. Over the past 10 years, the RGC made significant efforts to provide regular price information to farmers. The RGC uses two main ways of information dissemination: mass media, including public radio and TV; and interpersonal communication using extension agents. For the past five years, increasingly more villages have been able to receive information on prices, often from RGC programs (Figure 25.1). These government programs focus on rice and to a lesser extent on upland crops. In addition, dissemination of price information by private businesses became more common in about half of the villages visited during the FGDs. They were more focused on diffusion of information related to vegetables, even though the increase in the number of people getting information slacked off in the past two years. 113. Despite the increased access to information, farmers were not satisfied with the quality of service received. Only a handful of FGDs reported “very satisfied” and “satisfied” perceptions; the majority reported “not satisfied” (Figure 25.2). The most common reason was discrepancies between the announced prices on TV and the real prices at the market place. Government agents do not conduct their assessment carefully so disseminated prices are often higher than the actual selling prices, resulting in disappointment when farmers face buyers. Ultimately, the businessman/buyer has the last word. Farmers feel cheated and do not know what to do or where to get advice. On the other hand, farmers are frustrated when the announced prices are low, unduly below the production costs for some commodities. The system is then judged not useful since it is the bearer of bad news. Some farmers suggested that the government pass laws to control the price of agricultural products and disseminate information through the system. Indeed, the inadequate quality of prices disseminated through official media was reported by the Cambodian Institute for Cooperation and Peace in 2011: “Cambodian farmers do not get access to updated accurate market price data for their products.” 48 Figure 25: Access to information, Cambodia, 2013 1. Number of FGDs reporting access to 2. Satisfaction with the quality of information information by type of provider received Source: FGDs. 3. % of farmers receiving price information 4. % of farmers receiving price information from RGC activities from other value chain actors Source: Individual farmers’ survey. 114. Use of mobile technology has started to improve access to information . In 2013, 95 percent of the interviewed farmers owned a mobile phone. About a quarter reported using their mobile phone to request information, mostly on price and commercialization; most of these were located in Takeo and Kampot provinces. The use of mobile phones to get technical advice on technologies is still at the early stages, with only very few cases in Kampong Cham, Kandal, Battambang, and Bantey Meanchey provinces. These communications were often between private actors (farmers, sellers, suppliers); the percentage of farmers receiving calls or SMSs related to agriculture (price or techniques) from the RGC and NGOs remained below 10 percent. Business development services and contract farming 115. Services providing support to farmers with respect to business planning and contract farming are not yet developed in rural Cambodia. Only a handful of villages reported having these services available to them. For example in the past two years, only three villages in Takeo, Kampot, and Battambang provinces reported having the opportunity to work with business development service providers. The services provided focused on vegetables and upland crops, excluding rice. However, the quality of the support received from business development service providers was disappointing, as reported by several FGDs. 49 116. Contract farming in Cambodia is still limited. Only 6 out of the 36 FGDs provided an answer to the question related to contract farming. Groups that provided an answer were satisfied with the quality of services offered by the contractor. Contract farming is appreciated because it provides farmers with an easy way to sell their products, although some farmers complained about cases where other buyers offered higher prices. Among the individual farmers interviewed in 2013, only a handful of rice and vegetables producers in Takeo were involved in contract schemes. 117. Farmers with contracts were able to get higher prices and higher gross margins . The 2005 survey shows that contract farmers had a higher gross margin on rice ($478/ha versus $229/ha) and tobacco ($682/ha versus $132/ha). Even under harsh climate conditions such as drought, contract farmers had higher productivity due to better management practices. The 2013 survey did not specifically look at farmers with contracts. However, data collected on farm budgets show that the few farmers with contracts producing wet season rice had higher yield compared to noncontract farmers (3.5 tons/ha versus 2.84 tons/ha), received higher output prices ($325/ton versus $284/ton); and subsequently had a higher gross margin ($330/ha versus $236/ha). However, they had less cultivated land area, with an average of 0.8 ha versus 1.9 ha for noncontract farmers. Overall, the survey data showed no huge difference between the cost structures of farmers involved in contract farming and those of other farmers. Note that these averages should be treated with caution since they are from very small samples and are not statistically representative (i.e., 3 individual contract farms out of 33 rice farms interviewed in 2013).35 3.8. Farmers’ Organizations 118. Most farmers are not yet engaged in farmers’ organizations (FOs). FGD participants were not able to report the main reasons for using FOs, resulting in low response rates. This could indicate either a lack of awareness or an unsuccessful past experience in participating in an FO. In the selected FGDs, there was no report of creating cooperative FOs. In some cases, being a member of FOs has a direct advantage. For example, most farmers should be willing to be part of a water users association to get the benefits of irrigation and drainage. The low response rate (1 out of 36 FGDs) may be caused by the lack of irrigation infrastructure in the sampled villages. Similarly, a credit group is a valuable tool available to smallholders, alleviating the barrier to access to credit. In this sample, 4 out of 36 FGDs reported having such a group in their village even though there is an overall assessment that access to credit increased for farmers in the past years. On the other hand, higher numbers of FGDs reported having FOs focused on input supplies (6 cases including equipment supply) and production (10 cases) (Figure 26). The move toward mechanization and intensification of agricultural production would entice producers to be more interested in FOs (CDRI 2012). The practice depicted by the current study shows the emergence of private mechanized service providers. Having production groups would reduce the costs and improve the quality of the work for FOs based on machinery and would reduce the transaction costs with higher assurance on the quality for FOs based on inputs supply (seeds, fertilizers). 119. The governance and capacity of existing FOs need to be improved so that farmers appreciate the benefits of membership. About half of the FGDs recognized that FOs are very helpful; a quarter reported that they received very little benefit from FOs; and another quarter stated that FOs can do better. Suggestions on improvement from the FGDs included more education and awareness raising for farmers, more technical training to empower farmers to 35 The 2013 sampling method did not use contract farming as selection criteria. 50 bargain with other value chain actors, and formalization. Although Cambodia has a variety of FOs (Box 4), most of them are not operational. The RGC is trying to promote the formation of agricultural cooperatives that could lead to sustainable business models. Cambodia’s experience with cooperatives so far is still limited and it is too early to fully assess their success. Figure 26: Reasons for creation of FOs and their perceived benefits, Cambodia, 2013 Source: FGDs. Box 4: Farmers’ Organizations in Cambodia In Cambodia, five classes of FOs exist, as follows: (i) Farmers’ groups, with 5-15 members; (ii) Farmers’ associations, with > 15 members; (iii) Farmer Communities, for users of natural resources like forests, fisheries, and Water Users Groups; (iv) Federations, where different groups form an overarching common group; and (v) Cooperatives, called either farmer coops or agricultural coops, established at either the village or commune level MAFF’s current focus is on forming coops because they are supposed to lead to a more sustainable business model than farmers’ groups. Cooperatives are subject to a Royal Decree (“Compilation of legal framework for agricultural cooperatives 2012”) that includes a model statute for each cooperative to follow in setting its operating rules. There must be a minimum of 30 members to form a coop. Any one member can have a maximum of 20 percent of the coop’s shares. Each member has an equal vote, regardless of his/her number of shares. Activities of cooperatives can be many and varied, but must be limited to agricultural activities (e.g., input supply, marketing, production, credit provision, and money savings). The Royal Decree follows seven international principles of cooperatives, as identified by the International Cooperative Alliance. The Royal Decree is currently in the process of being upgraded to a law (i.e., by passing through the National Assembly). There is a prescribed five-step process for establishing a coop. The formation of a coop must be voluntary rather than forced upon farmers. Five pilot coops were established in five provinces in 2003. The Royal Decree makes provision for formation of coops, to be registered by MAFF, normally via the relevant Provincial Department of Agriculture (PDA). It is estimated that there are currently 360-370 cooperatives in Cambodia. The PDA is responsible for taking farmers through the five steps of forming a coop, and the PDA Director signs for registration of a coop once the steps are completed. MAFF provides its Coops Unit with a budget to help farmers obtain the training they require and covers the legal costs of establishing coops. Sometimes project donors contribute toward the costs of establishing coops, but farmer members need to cover their running costs. The PDA cannot interfere with the internal workings of a coop. However, each coop must hold an Annual General Meeting every year and send a report to the PDA. 51 3.9. Agricultural Infrastructure Irrigation 120. In addition to the RGC’s investment in irrigation infrastructure, some farmers are making individual investments in irrigation. Wokker et al. (2014) show that the elasticity36 of water irrigation in Cambodia may vary from 6 percent for the wet season to 12 percent for the dry season, measured in terms of rice production. This relatively low return does not stop Cambodian farmers from paying for and investing in irrigation (Figure 27.1). Sales of irrigation pumps rose 30 percent in the last five years, and the RGC’s Rectangular Strategy aimed to increase irrigated land by 120,000 ha from 2008 to 2013. For the 2013 dry season rice production, farmers spent an average $53/ha for irrigation, with the cost peaking at $115/ha for farmers in Takeo province. 121. Farmers stressed the importance of irrigation in their assessment of the impacts of flooding and drought on agricultural production. However, farmers were more concerned about drought than flooding. About half of the FGDs assessed flooding as having a major impact on their activities, while three out of four said the same for drought. Indeed, these natural shocks affect farmers differently depending on whether their landholdings are lowland or upland and on the agricultural season (wet or dry) (Figure 27.2 and Figure 27.3). Fear about the impact of drought led dry season rice producers to invest in small pumps and to allocate 11-20 percent of their total variable costs to irrigation. Indeed, the additional cost of water irrigation from the use of individual water pump equipment resulted in adequate yield and higher gross revenue but negatively affected farms’ gross margins. In general, farms using individual irrigation equipment received 50-150 percent lower gross margins than similar size farms with no irrigation costs. For modern farmers, regardless of farm size, irrigation expenditures constituted 18-20 percent of total variable costs (see Table 74 - Table 77 in Annex 3). 122. Aside from irrigation pumps, individual investment in water control is still underdeveloped. Irrigation pumps were very popular in several villages. Farmers identified pumps as an efficient way to overcome the lack of hard irrigation infrastructure such as a dam, while simultaneously reducing the likelihood of production lost due to drought and in some cases flooding. Pumps are used to drain water from flooded fields. Farmers also used pumps during the dry season, thus increasing the possible number of harvests per year as well as crop productivity. On the other hand, farmers were not using more modern irrigation technologies such as drip irrigation and greenhouses, often used for vegetable production in other countries, except in a few isolated cases. 36 Defined as the percentage change in production due to a percentage change in irrigated water. 52 Figure 27: Irrigation, Cambodia, 2013 1. Change in the use of and perspectives on pump irrigation Source: Individual farmers’ survey. 2. Perceived impacts of flooding (L) and drought (R) on agricultural productivity Source: FGDs. Other agricultural facilities 123. Expansion of cultivated areas and mechanized agricultural production increased the need for drying and storage facilities. Results from the FGDs show that about one in three villages was concerned with drying and storage facilities (Figure 28.1 and Figure 28.2), a constraint that mostly affected medium-size to large farms. The increased need for storage and drying facilities was directly caused by the mechanization of harvest and post-harvest activities. With mechanization, farmers face a high volume of wet products within a short period of time. To reduce losses and preserve product quality, farmers need to have adequate storage and drying facilities. Drying facilities became a priority for some villages in the past two years. During the FGDs, farmers complained about the low prices offered by middlemen and crop collectors for wet products. Without adequate drying and storage facilities, farmers were obliged to sell at the prevailing prices, often determined by the buyers. 124. Farmers were also concerned about the lack of social infrastructure, especially sanitation and potable water (Figure 28.3 and Figure 28.4). In its Rectangular Strategy, the RGC states that “water supply and sanitation” is a critical element of rural development. The government 53 planned to increase access to sanitation from 23 percent of the rural population in 2008 to 33 percent in 2013, and access to drinking water from 40 percent in 2008 to 50 percent in 2015. The overall health of farmers’ family members will be improved by greater use of adequate sanitation facilities and better access to drinking water. This will indirectly increase agricultural labor productivity by reducing the number of days of illness, for example. The entire value chain may benefit from this infrastructure due to greater prevention of contamination of fresh fruits and vegetables. Regulations in importing countries sometimes require a minimum level of infrastructure in sanitation and access to drinking water at production and processing units before products are eligible for export. Figure 28: Agricultural facilities, Cambodia, 2013 1. Change in the use of agricultural facilities 2. Perceived impact of lack of storage facilities on agricultural production 3. Perceived impact of lack of adequate sanitation facilities (L) and water distribution (R) on agricultural production Source: FGDs. 54 4. CHANGES IN FARM BUDGETS AND FARM PROFITABILITY 4.1. Introduction 126. Chapter 3 described the transformational changes taking place in Cambodian agriculture. Over the last decade, farmers have gotten better access to modern inputs and finance, and they get extension advice from various sources. They use more inputs as a result, and labor is being replaced by mechanized services. Farmers have better access to markets. All these improvements have increased the adoption of yield-increasing technologies. On the other hand, the quality of public programs has not been satisfactory. Farmers want more and better agricultural extension, combined with public and private delivery mechanisms. They want better price and market information. They seek to gain benefits from collective action. And they also want lower production and marketing costs to be achieved through better quality of irrigation and other infrastructure investments. This hints at the issues related to total factor productivity (TFP). Farmers use more inputs to maximize gross revenues but do not necessarily receive adequate public agricultural services to use these inputs as efficiently as possible to maximize incomes. 127. The 2013 survey collected detailed farm production information to understand the impact of the above changes in the agriculture sector on farm incomes. This information includes revenues, variable and fixed costs, gross margins, and returns to labor for each farm budget model. Surveyed farmers were selected based on their crops, their level of use of modern technology, and the area of cultivated land allocated to specified crops. Wet season rice, dry season rice, cassava, maize, and vegetables were chosen for study as they best characterize the changes in Cambodia’s farming systems since the mid-2000s. 128. Farms were characterized as small, medium or large. This categorization varies by the crop (Table 25). For cassava, small is a farm cultivating less than 5 ha, medium is between 5 ha and 10 ha, and large is over 10 ha. For maize and rice, small is below 1 ha, medium is between 1 ha and 3 ha, and large is over 3 ha. Vegetables are quite difficult to categorize because most farms would be considered small. It was decided to categorize farmers with less than 0.3 ha as small, between 0.3 and 0.5 ha as medium, and over 0.5 ha as large. The purpose of the categorization was to assess the likely impact of farm size on farm productivity and profitability. Table 25: Categorization of farm size by cultivated crop areas, Cambodia Farm Size Commodity Small Medium Large Wet season rice Less than 1 ha Between 1 and 3 ha More than 3 ha Dry season rice Less than 1 ha Between 1 and 3 ha More than 3 ha Cassava Less than 5 ha Between 5 and 10 ha More than 10 ha Maize Less than 1 ha Between 1 and 3 ha More than 3 ha Vegetables Less than 0.3 ha Between 0.3 and 0.5 ha More than 0.5 ha 129. Figure 29 presents the average farm size of the individual farmers interviewed in the 2013 survey. The average size was: 1.80 ha for wet season rice farmers; 4.15 ha for dry season rice producers; 8.25 ha for cassava farmers; 4.82 ha for maize farmers; and 0.41 ha for vegetable farmers. The average size of large cassava farms, for example, was above 17 ha, while for large 55 vegetable producers, the average farm size was only 2.37 ha. Average farm size by crop for the current study may differ from the national figures37 because of the use of purposive sampling for the selection of farms. This survey selected similar numbers of farms (one to two) using the criteria of farm size and level of technology use. Figure 29: Average land by farm size and crop, Cambodia, 2013 Source: 2013 survey. 130. In addition to the grouping by size, farmers were grouped into two categories of technology use: traditional and modern. This grouping was based on initial criteria (such as use of improved seeds, use of fertilizers and chemicals, mechanization, and irrigation) that were then used to carry out a factor analysis to identify similar farmers. An example of factor analysis output is presented in Box 5. For the data collection and analysis, two to three farmers were selected within each category as defined by crop, farm size, and technology use. This sample selection process selected a possible range of farmers who may not be statistically representative. 37 Overall, 48 percent of rural households in Cambodia hold less than 1 ha of agricultural land, though the difference between provinces can be significant according to the MAFF crop data assessments of 2003 and 2004. 56 Box 5: Factor Analysis and Modern versus Traditional Farmers The decision to categorize farmers as using traditional or modern technologies is not straightforward. No clear separation exists between users and nonusers of modern technologies in agriculture. Most farmers adopt at least one modern technology, and in practice, farmers are in a continuum of use for each technology. For example, very few farmers do not adopt chemical fertilizers, but for those who do the quantity applied varies from very little to very high. In this analysis, principal components factor analysis was used to reduce the optimal combination of several independent variables into one score index that ranks farmers along one dimension. The goal is to characterize farmers as modern or traditional based on their level of use of modern inputs, labor, and services. The following factors were retained for the analysis: use of modern inputs (improved seeds, fertilizers, and pesticides), use of agricultural services (land preparation, harvest, post-harvest, irrigation), and use of labor (modern farms have lower use). These variables were used to compute the rank of each farm into one latent variable, the projection of each farm from a multi-dimensional to a one-dimensional space. The analysis was conducted by crop. 131. The main elements of a farm budget are gross revenue, variable costs, gross margins, and returns to labor, defined as follows: a. “Gross revenue” is the amount received by farmers assuming that the entire production is sold at the farm gate price determined by farmers. Additional revenues obtained through the sale of byproducts are also taken into account. For example, in the case of paddy, gross revenue includes sales of straw for paddy rice. b. “Variable costs” are the amount spent by households for the production of the crop, excluding the costs of family members’ labor or the use of households’ agricultural assets for agricultural production. Total variable costs do not take into account the depreciation of households’ agricultural equipment. In general, variable costs include: input expenditures (improved seeds, manure, fertilizers, and chemicals – herbicides and insecticides); expenditures on hired labor (land preparation, sowing, plantation, and transplantation, weed control, crop management, labor for irrigation, manual harvest and post-harvest tasks); and costs of services (land preparation, plantation, irrigation, harvest and post-harvest, and transportation). c. “Gross margin” is gross revenue minus variable costs (Box 6). It is the remuneration of farmland, labor, and capital, as well as the profit received by farmers from production activities. Given that very few farmers rent land and own machinery, gross margin in Cambodia can be considered a remuneration to labor and farm management, which allows the calculation of the returns to own labor, the next indicator. d. “Returns to labor” is the gross margin divided by man-hours of own farm household labor spent on the production of specific crops. 57 Box 6: Gross Margin versus Value Added The concept of farm gross margin is similar to the concept of value added but it is not identical. The difference is in the treatment of hired labor. Value added is defined as gross revenue less intermediate inputs, while gross margin is gross revenue less intermediate inputs and hired labor. Depending on changes in the cost of hired labor, value added can differ from gross margins, sometimes significantly. 132. The results of the 2013 survey were compared with the 2005 study,38 which combined primary data with data from secondary sources, including the 1992/93 and 1997/98 NIS socioeconomic surveys’ datasets, market price data collected by Marketing Office/MAFF, and various studies undertaken in Cambodia. The primary data sources comprise focus discussions with key informants at the district level in representative agro-ecological zones. The list of key informants included local NGOs working in agriculture, forestry, and livestock development, a sample of producers (small, medium, and large) of major commodities, marketing agents, processors, and representatives from government agencies based at the district level. 133. Change over time was assessed by comparing data from the 2005 and 2013 surveys. These changes are presented in nominal and real terms. The GDP deflator was used to obtain real term results. The cumulative GDP deflator between 2005 and 2013 was 35 percent, according to the International Monetary Fund (IMF). Price data were converted into US$, using the exchange rate of Riels 4,000 to 1 US$. 4.2. Results of the 2013 Survey 134. Farmers got the highest total revenue per hectare from vegetable production ($2,843/ha), followed by cassava ($1,297/ha) and dry season rice ($992/ha). Wet season rice ($756/ha) and maize ($744/ha) were in between. Modern farms tend to generate larger revenues per hectare than traditional farms for all crops; the largest difference is for vegetables and the smallest is for dry season rice (Figure 30). Annex 3 presents detailed farm budgets for each crop and farm type. 38 The 2005 survey was conducted as a part of the World Bank’s Cambodia Agrarian Structure Study (2008). The 2005 survey collected 2004 wet season and 2005 dry season data (the 2004/05 marketing season), and the 2013 survey collected 2012 wet season and 2013 dry season data (the 2012/13 marketing season). 58 Figure 30: Gross revenues ($/ha) by crop and farm practice, Cambodia, 2013 Note: The revenue from vegetables for modern farms does not fit into the scale of this figure, and thus is not shown. Source: 2013 survey. 135. In general, farmers using modern technologies had higher yields and often sold their products at higher prices. The price of paddy for wet season modern producers was $242/ton versus $235/ton for traditional farmers; modern maize farmers received 21 percent more ($235/ton versus $195/ton); and cassava producers got slightly higher prices at $161/ton versus $157/ton. The combination of higher yield and higher prices of outputs resulted in higher gross revenue for modern farmers. 136. Similar to revenues, total variable costs were highest for vegetable producers. They averaged $1,481/ha. Total variable cost for maize was $440/ha, wet season rice $510/ha, dry season rice $696/ha, and cassava $791/ha, with differences between traditional and modern farms (Figure 31). Total variable costs are a critical element in determining the choice of crops, the cultivated areas, and the production strategy. If finance is a constraint for upland farmers, then maize would be the best choice because it requires less upfront operational costs compared to cassava. Similarly, growing dry season rice on irrigated land is less financially demanding than growing vegetables, assuming the same cultivated area. High production costs of vegetables explain the slow uptake in vegetable production in Cambodia. 59 Figure 31: Total variable costs ($/ha) by crop and farm practice, Cambodia, 2013 Note: The costs for vegetables for modern farms do not fit into the scale of this figure, and are thus not shown. Source: 2013 survey. 137. Variable costs are influenced by many factors, including farmers’ financial position, technology choices, and input prices. Hired labor accounted for the largest share of total variable costs for wet season rice and vegetable producers (Figure 32). Inputs and mechanized services were important for all crops, except vegetables. On average, expenditures of modern farmers exceeded those of traditional farmers, with the magnitude varying by type of crop: 10 percent for cassava, 40 percent for maize, 20 percent for dry season rice, and more than 200 percent for vegetables (Annex 3). These costs were mostly related to purchase of modern inputs and hiring labor. Figure 32: Breakdown of cost structure by crop, Cambodia, 2013 Source: 2013 survey. 138. In 2013, farmers producing vegetables got the highest nominal gross margins per hectare. For one hectare under cultivation, farmers had a gross margin of $1,394 for vegetables, $506 for cassava, $296 for dry season rice, $304 for maize, and $245 for wet season rice (Table 26). Gross margins were much higher for farmers using modern technologies, except for dry 60 season rice, while the impact of farm size on gross margins was not uniform (see more discussion on economies of scale below). Table 26: Gross margin ($/ha) by crop, technology use, and farm size, Cambodia, 2013 Farm Size Commodity Small Medium Large Total Overall Sample ($/ha) Wet season rice 207 270 240 245 Dry season rice 307 283 297 296 Cassava 544 438 512 506 Maize 244 336 330 304 Vegetables 1,575 747 306 1,394 Users of Modern Technologies ($/ha) Wet season rice 522 415 341 400 Dry season rice 276 174 283 255 Cassava 774 788 609 695 Maize 282 504 522 396 Vegetables 2,239 - -292 2,155 Users of Traditional Technologies ($/ha) Wet season rice 117 236 71 178 Dry season rice 312 349 415 336 Cassava 493 298 416 422 Maize 119 192 253 212 Vegetables 611 747 505 633 Source: 2013 survey. 139. Returns to labor varied from $4.62/day to $9.58/day, depending on the crop (Table 27). The returns to labor were well above the prevailing wages paid to hired labor in 2013, averaging $4.5/day. Overall, one day spent working on cassava production yielded $9.43 in return. For wet season rice, the return was $4.62/day, slightly higher than the daily cost of hired labor for medium-size wet season rice farms, but there was a huge difference between modern farmers ($12.67/day) and traditional farmers ($2.86/day). Dry season rice provided the highest returns to labor ($9.58/day), because of the very low share of labor in its cost composition, especially for large farms. Other crops showed a pattern similar to that observed for wet season rice: higher returns to labor when farmers used modern technologies. 61 Table 27: Returns to labor ($/day) by crop, technology use, and farm size, Cambodia, 2013 Farm Size Commodity Small Medium Large Total Overall Sample ($/day) Wet season rice 2.99 4.83 8.2 4.62 Dry season rice 6.15 8.27 22.83 9.58 Cassava 10.71 6.82 10.62 9.43 Maize 5.89 9.51 12.12 8.83 Vegetables 7.1 12.88 3.8 7.21 Users of Modern Technologies ($/day) Wet season rice 8.73 9.24 27.98 12.67 Dry season rice 90.61 74.36 55.3 60.04 Cassava 13.75 8.58 12.17 11.2 Maize 6.32 12.09 38.1 10.56 Vegetables 7.54 - - 7.54 Users of Traditional Technologies ($/day) Wet season rice 1.62 4.05 1.23 2.86 Dry season rice 5.4 6.54 5.43 5.85 Cassava 9.95 5.6 8.94 8.45 Maize 3.83 6.42 7.81 6.75 Vegetables 5.42 12.88 6.31 6.53 Source: 2013 survey. 140. In most cases, small farms generated similar gross margins per hectare as larger farms. Only maize production and, to a limited extent, dry season rice production for traditional technology adopters showed higher returns as farm size increased (i.e., economies of scale). In these two cases, large farms had higher gross margins per hectare compared to medium-size and small farms (Table 26). For maize production, large farms using modern technologies had about twice the gross margins of small farms ($522/ha versus $281/ha) and medium farms about three- fourths higher ($504/ha versus $281/ha). Similar trends were observed for farmers practicing traditional techniques, with $253/ha, $191/ha, and $118/ha for large, medium-size, and small farms, respectively. 141. For vegetable production and modern farmers growing wet season rice, gross margins per hectare were higher for small farms. Small vegetable farms had the highest gross margin per hectare, with an average of $2,239/ha when adopting modern technologies. Even for farmers using traditional techniques, gross margins calculated from small farms’ budgets were 17 percent higher than large farms’ gross margins. Lower returns for larger farms may be due to several factors. The theory of indivisibility of assets does not hold for most agricultural tasks. For example, during land preparation farmers have a choice between using draught oxen and ox-plows, power tillers, and small mechanized equipment, and tractors that are available in different sizes and powers. For harvest, farmers’ options range from the use of manpower to the use of combines; reapers and threshers lie in between. Farmers may adopt specific strategies depending on 62 household endowment to maximize either production (for food-insecure households) or profit (for commercial smallholders). Decreasing economies of scale may also be the consequence of small farms’ higher efficiency compared to large farms. For farmers with low literacy, management difficulties in addition to increased risks may reduce efficiency as scale expands. For example, a large farm that cultivated 3 ha of watermelon in Battambang ended up with a total gross loss of $876 ($292/ha). Indeed, vegetable production is prone to several critical phases that require timely and adequate management decisions. Poor management of frequent pest attacks, lack of irrigation water during the growing period, or issues of product perishability during harvest and post-harvest periods will reduce the efficiency in vegetable production. Under these constraints, farmers are more vulnerable to production and price risks. 142. Yet the analysis of the returns to labor shows clearly increasing returns as farm size increases. Returns to a day of labor for wet season rice production increased from $2.86/day for small farms to $12.67/day for large farms (Table 27). The average return for wet season rice was $4.83/day. Maize production showed very similar trends: medium-size farms exhibited 61 percent higher returns to labor compared to small farms and large farms twice as high, at $12.19/day. Dry season rice production showed the highest impact of farm size on returns to labor with $6.15, $8.27, and $22.83 for small, medium-size, and large farms, respectively. For cassava, the size of farms seemed not to affect returns to labor. Returns to labor for large farms seem to benefit from the reduced labor for management and crop monitoring. It may also benefit from better access to capital, resulting in higher use of labor-saving equipment. Large farms may also have lower costs of transactions and transportation, both for inputs (consolidation of purchase) and commercialization, which would affect assembly costs (consolidation of sales). 143. The returns to labor for small farms producing rice during the wet season were below the prevailing wage rate. This has a direct impact on tackling poverty reduction because wet season rice still constitutes 83 percent of total cultivated rice land, comprising most smallholder farmers. This is critical for poverty reduction policy because returns to labor determine farmers’ wealth, and rice cultivation occupies more farmers in Cambodia than any other crop. 144. A higher level of technology use enhances returns to labor. Farmers using a higher level of technology had much higher returns to labor, regardless of farm size (Table 27). The ratio was about 1 to 2 for maize, 1 to 10 for dry season rice, and 1 to 4 for wet season rice. Vegetable and cassava producers had the least impact of technology use on returns to labor, although modern cassava farmers had a 30-50 percent higher return than traditional farmers. 4.3. Comparison of the 2005 and 2013 Survey Results 145. In nominal terms, all 2013 gross margins were above the 2005 values, except for maize (Table 28 and Figure 33). A comparison of the gross margins between the two periods shows that vegetable producers got five times more than the $284/ha received in 2005; the gross margin for cassava increased 2.5 times from $198/ha; and dry season rice doubled from $195/ha. However, maize was reduced by 50 percent, from $574/ha to $304/ha. 63 Table 28: Comparison of key farm budget indicators, $/ha, Cambodia, 2005 and 2013 2005 2013 Change Total Revenue ($/ha) ($/ha) (%) Cassava 375 1,297 246 Maize 708 744 5 Wet season rice 345 756 119 Dry season rice 398 992 150 Vegetables 433 2,843 556 Total Variable Costs Cassava 178 791 346 Maize 134 440 229 Wet season rice 186 510 174 Dry season rice 202 697 245 Vegetables 149 1,449 893 Gross Margins Cassava 198 506 156 Maize 577 304 -47 Wet season rice 159 245 55 Dry season rice 195 296 51 Vegetables 284 1,394 380 Source: 2005 and 2013 surveys. 146. These changes were driven by changes in yields and prices of outputs and inputs. From 2005 to 2013 (survey data), cassava prices increased by 200 percent, paddy rice prices by 90 percent, maize prices by 64 percent, and vegetable prices from 60 percent (eggplant) to 233 percent (watermelon). During the same period, prices of inputs almost tripled for seeds and labor ($1.2/day to $4.5/day), and doubled for fertilizers (urea from $0.30/kg to $0.56/kg and DAP from $0.35/kg to $0.68/kg). 147. The structure of costs, especially for the use of labor, also showed radical changes , with sharp declines of 20 percent (maize) to 70 percent (dry season rice). The lower labor use was a response to rising wages and availability of mechanization services in rural areas. The exception was vegetables, where the labor use increased five-fold, from 38 days/ha to 176 days/ha. Since the 2005 survey included a mix of vegetables while the 2013 survey focused on four main vegetable crops,39 this huge shift may be due to the change in the types of vegetables produced. 148. In 2013, farmers used more modern inputs per hectare, especially improved seeds and fertilizers. In 2005, surveyed farmers applied an average 100 kg/ha of fertilizers (all combined) for dry season rice production. This quantity was 2.5 times higher in 2013, reaching 250 kg/ha. The same pattern was observed for vegetables, with an increase from 110 kg/ha to 295 kg/ha. The rates of application for wet season rice (120 kg/ha) and maize (50 kg/ha) did not change much. However, the rate of application decreased for cassava from an average 30 kg/ha to 10 kg/ha in 2013. 39 Cucumber, mustard, eggplant, and watermelon. 64 Figure 33: Comparison of nominal gross margins ($/ha), Cambodia, 2005 and 2013 Source: 2005 and 2013 surveys. 149. In nominal terms, the sum of gross margins for all selected crops more than doubled, from $469 million in 2005 to $1,067 million in 2013. The cumulative increase was 103 percent, or 13 percent per year. This increase was driven by both the expansion in land area (37 percent) and the per hectare gross margins (63 percent) (Table 29). The smallest increase in total margin was observed for maize: 90 percent from 2005 to 2013; the largest increase was observed for cassava: 1,180 percent. Table 29: Comparison of total crop budgets, Cambodia, 2005 and 2013 2005 2013 Crop Unit Cultivated Total Unit Cultivated Total Margin Land Margin Margin Land* Margin ($/ha) (ha) ($ million) ($/ha) (ha) ($ million) Wet season rice 159 2,121,591 337 245 2,484,832 609 Dry season rice 195 321,939 63 296 495,465 147 Cassava 198 30,032 6 506 337,800 171 Maize 577 90,732 52 304 215,442 65 Vegetables 284 35,762 10 1,394 54,155 75 Total 180 2,600,056 469 298 3,587,694 1,067 Note: * Land area is used for the year 2012. Source: 2005 and 2013 surveys. 150. In real terms, however, the growth in total gross margins was smaller. It rose by 68 percent during 2005-2013, or 9 percent a year (Table 30). This increase was due to the 38 percent change in cultivated crop land and 30 percent increase in per hectare gross margin. Unit margins increased for all crops except maize. 65 Table 30: Changes in crop margins in nominal and real terms, Cambodia, 2005 and 2013 Change in Change in Change in Change in Change in Cultivated Total Total Unit Margins Unit Margins Crop Land Margins Margins in Between 2005 in Real Between 2005 Between 2005 Real Terms and 2013 (%) Terms (%) and 2013 (%) and 2013 (%) (%) Wet season rice 54 17 71 19 36 Dry season rice 52 54 106 17 71 Cassava 156 1,025 1,180 121 1,145 Maize -47 137 90 -82 55 Vegetables 391 51 442 356 407 Total 65 38 103 30 68 Note: The cumulative GDP deflator from 2005 to 2013 was 35 percent. Source: 2005 and 2013 surveys. 151. There are differences in the sources of growth across crops. Although cassava’s per hectare profitability grew annually by 15 percent, the most growth in total gross margin came from land expansion (Table 31). The problem with that growth is that land expansion in cassava is unsustainable, since it is carried out through exploitation of soils that were previously forest land. Most growth in maize and dry season rice also came from land expansion. But in the case of wet season rice and especially vegetables, the increase in total gross margin was mostly due to the increase in per hectare profitability. On average, the total gross margin for all crops considered increased 8.2 percent per year, due to 4.7 percent land expansion and 3.4 percent in per hectare margins. Table 31: Annual changes in crop margins in real terms, Cambodia, 2005 and 2013 Change in cultivated Change in unit gross Change in total gross area between 2005 and margin between 2005 margin between 2005 and 2013 (%) and 2013 (%) 2013 (%) Wet season rice 2.1 2.4 4.5 Dry season rice 6.7 2.1 8.8 Cassava 128.1 15.1 143.2 Maize 17.2 -10.3 6.9 Vegetables 6.4 44.5 50.9 Weighted 4.7 3.4 8.2 average Source: 2005 and 2013 surveys. 152. The increase in agricultural production and farm profitability has led to an increase in the returns to own farm household labor. The returns to labor in real terms increased by 84 percent for wet season rice, 254 percent for cassava, and 278 percent for dry season rice 66 (Table 32).40 But the returns to labor fell for maize and vegetables. The significant increase in the returns to labor for rice and cassava was a result not only of higher total gross margins but also of the reductions in labor used in production. The reduction in labor used ranged from 19 percent for maize to 69 percent for dry season rice (i.e., the structural transformation in action). Vegetables were the only crop where labor use increased over the observed period. This led to the decline in returns to labor in both nominal and real terms. Labor is being increasingly replaced by mechanized services in Cambodia. Table 32: Changes in returns to labor in nominal and real terms, Cambodia, 2005 and 2013 Change in Change in Nominal Terms Real Terms Labor ($/ha) 2005 2013 (%) (%) Cassava 104.81 227.40 117 54 Maize 46.49 142.85 207 144 Wet season rice 101.81 220.21 116 53 Dry season rice 107.88 126.50 17 -46 Vegetables 45.83 802.10 1,650 1,587 Labor (days/ha) Cassava 87.34 49.87 -43 Maize 38.74 31.33 -19 Wet season rice 84.84 48.29 -43 Dry season rice 89.90 27.74 -69 Vegetables 38.19 175.90 361 Returns to labor ($/day) Cassava 2.26 9.43 317 254 Maize 14.82 8.83 -40 -103 Wet season rice 1.87 4.62 147 84 Dry season rice 2.17 9.58 341 278 Vegetables 7.43 7.21 -3 -66 Note: The cumulative increase in the consumer price index between 2005 and 2013 was 63 percent. Source: 2005 and 2013 surveys. 153. As a result of increasing returns to labor, agricultural wages in general rose, converging with nonfarm wages. Agricultural wages grew by a factor of almost three between the 2005 and 2013 surveys (Error! Not a valid bookmark self-reference.).41 Over the same 40 The adjustment factor for converting the returns to labor from nominal to real terms is the consumer price index (inflation), which includes prices of both food and non-food items. This index shows changes in the purchasing power of farmers, and is higher than the GDP deflator used to deflate gross margins. The GDP deflator includes only intermediate inputs used for production. 41 This agricultural wage increase between 2005 and 2013 surveys is much higher than the 48 percent agricultural wage increase from 2007 to 2009 reported in the 2013 poverty assessment. There are several reasons for this difference. First, the surveys have different household coverage. Second, the poverty assessment does not include the 2005-2007 and 2009-2013 time periods. The increase was especially rapid between 2011 and 2013 when wages grew again after the slowdown during 2009-2011 in the aftermath of the global financial crisis. 67 period, nonfarm wages increased by only 60 percent. As a result, the gap between farm and nonfarm sector wages declined. Thus agriculture appears to have provided excellent income- earning opportunities in Cambodia over the last decade. Figure 34: Farm wage, off-farm wage, and returns to agricultural labor ($/day), Cambodia Source: 2005 and 2013 surveys. 68 5. COMPETITIVENESS OF CAMBODIAN FARMS 5.1. Introduction 154. Chapter 4 described the farm income situation in Cambodia. One of the findings is the rising production costs, especially for maize and rice. In most cases, the rise in production costs is compensated by higher revenues. The question, however, is how the increase in production costs affects farm competitiveness. 155. Farm competitiveness in Cambodia is analyzed using the Domestic Resource Cost (DRC) approach developed by Tsakok (1990). DRC measures a country’s efficiency in using domestic resources to produce specific goods. The key question DRC answers is whether a country has a comparative advantage in producing a certain good or if import is a better option so that domestic resources may be used to produce other commodities that result in higher profit. A DRC less than one suggests comparative advantage and efficiency in production: the domestic opportunity costs of producing the selected good are less than the domestic value added generated by the production process. Conversely, a DRC greater than one implies that production is inefficient and that foreign exchange would be better spent importing the product rather than producing it domestically. 156. The DRC is computed based on the following assumptions: a. The foreign exchange premium is set to zero. In Cambodia, the foreign exchange rate is not a source of distortion because of the dollarization of the economy (i.e., US dollars and Riel are freely interchangeable within the country). b. The capital recovery costs of agricultural equipment are not computed. The rental costs of equipment paid by farmers, including for irrigation using pumps, are split 80:20 percent, where 20 percent represents the cost of fuel, a tradable commodity. For transportation and irrigation, the estimated ratio is 60 percent nontradable and 40 percent tradable. The assumption is that the opportunity cost of equipment can be approximated by the rental fee because of the existence of a competitive market for agricultural equipment rental. c. Land price is set to zero so that the social profit includes returns to land and to management. d. A first set of DRCs is computed using average distances of 70-100 km from the production site to the export location for agricultural outputs, and conversely, from the off-load port to input suppliers’ stores. This is the average distance between Battambang, Kandal, or Bantey Meanchey to the nearest border. Transportation cost is assumed to be Riel 32,000/ton ($8/ton). One to two provinces per crop were selected to compute the DRCs, taking into account the specificity of farm budgets in each region. e. The reference international prices used in the analysis are those of Thailand Hom Mali grade A rice, Thailand Bangkok 5 percent broken rice, maize yellow grain, and dried cassava chips. 157. The computed DRCs illustrate Cambodia’s high comparative advantage in agricultural production. The country has an advantage in producing most agricultural products. All of the DRCs are below one, with some very low, such as those for maize and cassava production, revealing a strong advantage in using domestic resources (Table 33). Least competitive 69 is wet season paddy production, the costs of which should be monitored closely to avoid losing its comparative advantage in the future. Table 33: DRCs for selected agricultural crops, Cambodia, 2013 Commodity DRC Dry season rice (paddy) Kandal 0.69 Dry season rice fragrant (paddy) Kandal 0.43 Dry season rice (paddy) Bantey Meanchey 0.61 Milled rice (IRRI 5% broken) 0.54 Wet season rice (paddy) Takeo 0.78 Wet season rice (paddy) Bantey Meanchey 0.54 Wet season rice (paddy) Kampot 0.86 Wet season rice fragrant (paddy) Kampot 0.61 Milled rice fragrant Kampot 0.39 Cassava (dried chips) All 0.39 Cassava (dried chips) Battambang 0.35 Maize (grain) All 0.35 Maize (grain) Bantey Meanchey 0.35 Maize (grain) Kampong Cham 0.22 Source: Own estimates based on the 2013 survey data. 5.2. Domestic Resource Cost for Rice 158. Comparative advantage depends on the variety of cultivated rice but, overall, rice is produced competitively across the country. Wet season rice production for IRRI varieties in Takeo has a DRC of 0.78 (yield of 2.88 tons/ha), the closest of Cambodia’s computed DRCs to inefficient use of domestic resources (i.e., a DRC greater than one). However, farmers producing fragrant rice have a greater comparative advantage, with a DRC of 0.61 in Kampot (yield of 2.5 tons/ha). Despite the lower yield for fragrant rice, the main difference in DRCs is due to the prices of these varieties. The same pattern is observed for dry season rice. IRRI varieties have a DRC of 0.61-0.69 (yield of 4.8 tons/ha), more than 50 percent higher compared to fragrant variety rice with a DRC of 0.43 (yield of 4.0 tons/ha). 159. How large is the reserve in comparative advantage? In other words, how sensitive is it to future changes in costs, prices, and types of technology use? For dry season rice, for example, a 50 percent decrease in the cost of agricultural services would result in a 29 percent decrease in the DRC, from 0.60 to 0.43; i.e., Cambodian dry season rice would enjoy a relatively higher comparative advantage. This simulation is based on farmers using water pump irrigation; a different DRC would arise in the presence of public investment in adequate irrigation. An increase in the price of modern inputs (seeds, fertilizers, and other chemicals) would worsen the comparative advantage of dry season rice production (DRC increases from 0.60 to 0.79). In all of 70 the sensitivity analyses, the simulated DRCs change but remain below the threshold of one, indicating Cambodia’s advantage in producing rice (Table 34). 160. By location, Bantey Meanchey province has a greater comparative advantage in producing dry season rice (DRC=0.61) than Kandal province (DRC=0.69). The ratio of tradable components to variable costs in these two provinces differs significantly (26 percent of tradable components for Kandal and 44 percent for Bantey Meanchey). In fact, farmers from Bantey Meanchey allocate more chemical fertilizers and less labor for dry season rice production, resulting in a higher dependence on tradable inputs. Access to and cost of irrigation water, however, may change the DRC for Bantey Meanchey, as farmers spent close to 19 percent of total variable costs on irrigation compared to less than 3 percent for farmers in Kandal. Table 34: Sensitivity analysis of impacts of changes of production costs on DRC for rice Dry Season Rice Decrease in the Cost of Services Increase in the Cost of Inputs % change in the factors 0% 10% 20% 50% 0% 10% 20% 50% DRC 0.60 0.56 0.53 0.43 0.60 0.63 0.67 0.79 % change in DRC 0% -7% -13% -29% 0% 5% 10% 30% Wet Season Rice Decrease in the Cost of Services Increase in Yield % change in the factors 0% 10% 20% 50% 0% 10% 20% 50% Simulated yield (kg/ha) 2,880 3,168 3,456 4,320 DRC 0.98 0.95 0.91 0.80 0.98 0.88 0.79 0.61 % change in DRC 0% -4% -8% -19% 0% -11% -19% -38% Source: Own estimates based on 2013 survey data. 161. Wet season rice production’s sensitivity analysis was conducted using the price for IRRI high-yielding varieties, a pathway Cambodia may take to improve its rice value chain. A 10 percent increase in rice yield (from actual 2.88 tons/ha to 3.17 tons/ha) corresponds to an 11 percent decrease in the DRC (DRC=0.88), making the country more competitive in producing high-yielding rice varieties. A 50 percent increase in yield (4.3 tons/ha) results in a 38 percent decrease in the DRC (DRC=0.61). The decrease in the DRC is not linear; for example, doubling average rice yield to 5.76 tons/ha would only result in about a 50 percent decrease in the DRC, and this sensitivity analysis does not take into account the additional cost of inputs required to attain this high level of productivity. Still, even with a 20 percent increase in yield (DRC=0.79), the DRC for high-yielding varieties remains above the DRC for fragrant rice (DRC=0.61), indicating that it is advantageous for Cambodia to focus on fragrant rice. To get to similar DRCs with fragrant rice production, the yield for IRRI varieties would need to increase by about 50 percent, from 2.88 tons/ha to 4.3 tons/ha (Table 34), an increase that would require additional costs. 162. Takeo (DRC=0.78) shows an advantage over Kampot (DRC=0.86). Takeo’s DRC is 15 percent lower, as its farmers use slightly more modern inputs than their peers in Kampot, and thus were able to get higher yield (20 percent more at 3.1 tons/ha). Again in this case, the comparative advantage is driven by the difference in yield. 71 5.3. Domestic Resource Cost for Cassava and Maize 163. For non-rice crops, Cambodia’s highest comparative advantage is in maize production (DRC=0.35), followed by cassava production (DRC=0.39). For cassava, nontradable costs represent more than 87 percent of total variable costs, indicating that few policy adjustments related to tradable inputs are available to improve the comparative advantage of this crop (Table 33). However, cassava production still uses relatively high amounts of labor, so changes in labor cost may affect cassava’s comparative advantage. A sensitivity analysis showing the effect of change in labor cost is presented in Table 35. Every 10 percent increase in the cost of hired labor would result in an increase of cassava’s DRC by 3-4 percent; a 50 percent increase would lead to a loss of about 17 percent in Cambodian cassava production’s comparative advantage. Labor cost is likely to increase in the near future if Cambodia continues its path toward agricultural transformation, stimulating more competition between labor available for agricultural versus nonagricultural work. 164. Battambang province has a better-than-average DRC (0.35 versus 0.39). Cassava yield is slightly higher in this province compared to the average (20.4 tons/ha versus 18.4 tons/ha). Cassava production is also less dependent on changes in the prices of tradable inputs (fertilizers, pesticides). A 10 percent increase in the world prices of these inputs will only increase cassava’s DRC by 1 percent. Table 35: Sensitivity analysis of impacts of changes in production costs on cassava and maize DRCs Cassava Decrease in Yield Increase in the Cost of Labor % change in the factors 0% 10% 20% 50% 0% 10% 20% 50% Simulated yield (kg/ha) 18,650 16,785 14,920 9,325 DRC 0.39 0.43 0.49 0.81 0.39 0.40 0.41 0.45 % change in DRC 0% 12% 27% 111% 0% 3% 7% 17% Maize Increase in the Cost of Inputs Increase in Yield % change in the factors 0% 10% 20% 50% -50% -20% 20% 50% Simulated yield (kg/ha) 1,970 3,152 4,728 5,910 DRC 0.35 0.36 0.37 0.40 0.89 0.46 0.28 0.22 % change in DRC 0% 2% 5% 14% 154% 31% -20% -38% Source: Own estimates based on 2013 survey data. 165. Cambodian farmers use relatively high amounts of modern inputs for maize production. About 40 percent of the production costs for maize are tradable commodities, implying that the comparative advantage of Cambodia for this crop closely depends on the world market behavior. Overall, doubling the price of modern inputs (seeds, fertilizers, and chemicals) would result in a 14 percent increase in maize’s DRC (0.40); i.e., Cambodia still has a strong comparative advantage in producing maize. 166. To keep maize’s comparative advantage, the focus needs to be on maintaining its relatively higher yield (3.9 tons/ha), which is currently not far from the average yields in neighboring countries. A 20 percent decrease in average yield would result in a loss of 31 percent of maize’s comparative advantage (DRC=0.46); and a 50 percent decrease (to 1,970 kg/ha) would 72 cause the country to lose most of its comparative advantage (DRC=0.88). On the other hand, any increase in yield is expected to improve its advantage in producing maize (DRC= 0.22 with a 50 percent increase in yield, to 5.9 tons/ha). 167. Bantey Meanchey province has the least comparative advantage in maize production (DRC=0.35), while Kampong Cham province has the highest comparative advantage (DRC=0.22). The ratio 1:1.6 of the comparative advantage is partially related to the observed yields in these provinces: 3.9 tons/ha versus 5.3 tons/ha for Bantey Meanchey and Kampong Cham, respectively. 5.4. Summary 168. All commodities analyzed indicate Cambodia’s comparative advantage in agricultural production. The comparative advantage is the strongest for fragrant rice, maize, and cassava. However, wet season rice production has an almost neutral advantage. Fragrant rice has a much higher comparative advantage than high-yielding IRRI varieties, and dry season rice has a comparative advantage relative to wet season rice. 169. The advantage of agricultural production is strongly affected by price variation in the world market. A 20 percent decrease in rice prices or a 20 percent increase in tradable input prices would make wet season rice production uncompetitive (Table 36). Other crops have more room to accommodate rising costs and falling prices, however, indicating a strong foundation for continued strong agricultural growth in Cambodia for many years to come. Table 36: Sensitivity analysis of impacts of changes in tradable input and output prices on DRCs of all crops (except vegetables) Wet Dry Season Season Cassava Maize Rice Scenario Rice Current status DRC 0.60 0.98 0.39 0.35 10% increase in world price of DRC 0.53 0.87 0.35 0.31 commodities Change -11% -11% -9% -10% 20% decrease in world price of DRC 0.83 1.32 0.49 0.46 commodities Change 39% 35% 27% 33% 20% increase in world price of tradable DRC 0.66 1.01 0.39 0.36 inputs Change 10% 3% 1% 4% Source: Own estimates based on 2013 survey data. 73 6. DRIVERS OF PAST GROWTH 170. Cambodia’s agricultural growth of the past decade can be interpreted as the outcome of a number of factors. These positive factors or drivers can be divided into policy, connectivity and investments, and incentives (Table 37). This list is not exclusive but it includes most important drivers. Table 37: Drivers of agricultural growth during 2002-2012 Policy Connectivity and Investment Incentives (Food Prices and Wages) • Market-oriented agricultural • Improved connectivity to • Higher food prices and trade policies conducive markets, technology, and originating from world for private sector investments finance market structural changes • Trade integration with the • Private sector investments, • Expansion of domestic global economy and in including FDI market due to the economic particular with ASEAN growth countries • Increase in labor cost originating from migration of labor out of agriculture Source: Own assessment. 171. The most powerful factor behind the past agricultural growth was private sector performance in Cambodia. It was the energy, resources, and entrepreneurship of farmers, traders, and other businesses that made the growth possible. Cambodia’s agricultural and trade policy has been generally market oriented and conducive to private sector development. Although the costs of doing business in Cambodia are still high (World Bank 2014e), they would have been much higher had the RGC introduced farm output or input subsidies or restricted exports at the time of the food price spikes in 2007-2009. The market-oriented policy allowed farmers and entrepreneurs to make undistorted decisions about production and trade and to respond to incentives. It allowed the higher global rice prices that prevailed in the past decade to pass through to local markets. It helped trigger a strong farm supply response to higher demand for agricultural commodities such as rubber, cashews, and beef originating from neighboring countries, and facilitated farmers’ diversification towards upland crops such as cassava and maize in response to higher prices. 172. In the last decade, global agricultural markets experienced structural changes. Due to a combination of factors that resulted in global demand for food exceeding global supply, global food prices significantly increased after 2007-2008. The World Bank Food Price Index in real terms increased by 43 percent between 2003-2006 and 2009-2013 (Table 38). The world market prices of cereals, coffee, and rubber increased the most. The price increase in global markets has passed through to Cambodia. As a result, in recent years the food CPI consistently exceeded non- food CPI in spite of the rising agricultural production (Figure 35), pointing to the external factors of food inflation in Cambodia. 74 Table 38: Global agricultural prices, selected commodities, $/ton (in real 2010 prices) 2003-2006 2009-2013 Change (%) Maize, Yellow, US Gulf 128 230 80 Rice, 5% Thai 298 513 72 Wheat, Hard Wheat, US Gulf 190 266 40 Rubber, Asia, TSR 20 1,720 3,170 84 Coffee, Robusta, New York 1,220 1,940 59 Beef, Australia/New Zealand 2,810 3,500 25 World Bank Food Price Index 71.22 102.17 43 Source: World Bank Pink Sheets. Figure 35: CPI and food price index in Cambodia, 2002-2012 Source: NIS 2014. 173. Located in the heart of the Greater Mekong subregion, Cambodia benefits from connectivity and trade with its regional partners. Most surplus paddy, cashew nuts, sesame seeds, cassava, and cattle find a market in neighboring Vietnam and Thailand. However, most of this trade is informal and consists of unprocessed products. An estimated 9 million tons of cassava, 3 million tons of paddy, 1 million tons of sugarcane, and perhaps another million tons of agricultural, livestock, and fishery products move across borders, representing about 1 ton of agricultural export per capita. At the same, time considerable volumes of pigs, vegetables, and oil are imported. 174. The formal exports of rice mainly go to the EU. Cambodia primarily exports quality rice to European countries, benefiting from access to the EBA, which provides price benefits to Cambodia relative to competitors India, Vietnam, and Thailand. Vietnam and Thailand, two major rice exporters, are highly diversified globally in terms of their destination markets. In the group, only Vietnam lags behind in terms of quality rice exported. India and Thailand are renowned for their fragrant rice (Basmati and Jasmine) and Cambodia also now produces and exports high- quality fragrant rice. However, Vietnam recently started producing and exporting quality fragrant rice and its share of fragrant rice in total rice exports is expected to increase rapidly in the years to come. 75 175. Among these four countries, Cambodia has the most clearly formulated rice policy . The RGC has set clear targets for exports and is promoting development of transportation (by upgrading the Sihanoukville port) and power infrastructure. At the same, time it is promoting investment in rice milling and in a number of programs to support better seeds and irrigation. As a result, formal rice exports increased from 1,500 tons in 2008 to 378,000 tons in 2013 (recall Table 9). 176. Due to private investments, Cambodia’s rice milling capacity increased remarkably in recent years. Cambodia’s modern rice milling (i.e., the larger mills) capacity increased sevenfold, from 96 tph in 2009 to over 700 tph in late 2013. The polishing capacity of mills also jumped, from 72 tph in 2009 to 564 tph in 2013, with an average milling rate of 64 percent. All of the investment in large mills came from the private sector, at least 35 percent of which was from joint ventures with foreign investors. In theory, the existing milling capacity could process almost the entire paddy surplus in Cambodia. However, the high cost of fuel and electricity renders Cambodia’s average milling cost about 30 percent higher than that of Thailand and Vietnam. Moreover, rice mills lack the working capital to purchase and store paddy. 177. Cambodia’s rapid economic growth since the mid-1990s has led to significant changes in the country’s labor market especially in terms of increased employment, shift in employment structure, and improved outcomes for workers. Agriculture employed about 81.4 percent of the labor force in 1995, but this share had shrunk to 51 percent by 2012 (recall Table 11). Employment in the industry sector increased from 3 percent of the total employed workforce in 1995 to 19 percent in 2012, while employment in the services sector doubled from 16 percent to 32 percent. Labor market outcomes improved more or less in parallel with the growing trend towards labor migration. In 2008 there were 2.5 million recorded internal migrant workers, 89,545 migrants registered as working abroad officially and an estimated 180,000 working abroad unofficially. 178. The options for labor in Cambodia, particularly youth labor, are to migrate from agriculture to nonfarm activities, to migrate from rural areas to urban areas (for example, to garment factories for young women and construction work for young men), or even to migrate out of the country. These options result in increased choices for job seekers in rural areas in addition to agricultural work. As a result, agricultural wages have increased by a factor of four over the past 10 years (As a result of increasing returns to labor, agricultural wages in general rose, converging with nonfarm wages. Agricultural wages grew by a factor of almost three between the 2005 and 2013 surveys (Error! Not a valid bookmark self-reference.). Over the same period, nonfarm wages increased by only 60 percent. As a result, the gap between farm and nonfarm sector wages declined. Thus agriculture appears to have provided excellent income- earning opportunities in Cambodia over the last decade. 179. Figure 34). This has pushed farmers towards increased adoption of mechanization. 180. Outside of the agriculture sector, connectivity has improved in terms of both communication and roads. With a population of 14.6 million, Cambodia had 19 million mobile phone subscribers in 2012. Currently transportation costs for rice and other agricultural commodities are around $10-13/100 km/ton (Figure 36). Compared to an estimated $15/100 km/ton in 2009 (ADI 2009), transportation costs decreased due to road improvements, more trucks available, and more competition amongst local transport companies. Yet transport costs in Cambodia are still higher than in neighboring countries such as Vietnam, where it is about 76 $7/100 km/ton and Thailand where it is around $5/100 km/ton. Costs are also higher on rural roads due to the underinvestment. The ADB survey of 18 communes in Battambang, Kampong Thom, and Takeo provinces in 2013 found only one to have an asphalt road (ADB 2014a). It should be noted that both Vietnam and Thailand rely significantly on river transport for moving rice and other crops, which is significantly cheaper. Using its waterways, Vietnamese transport costs are only $3/ton compared to $15-17/ton to move the same distance by truck in Cambodia.42 Figure 36: Transportation costs in Cambodia and neighboring countries, $/100 km/ton Source: Own estimates and ARPEC. 181. While past agricultural growth in Cambodia resulted in an impressive increase in production and exportable surplus of a number of commodities, it had limitations. To promote future growth, limitations that need to be addressed include: (i) the reliance on crop land expansion; (ii) the increase in the number of very small farms; (iii) the slow job creation outside agriculture; (iv) relatively low productivity growth; and (v) weak development of agroprocessing and other agribusinesses. 182. First, the large share of the past growth was achieved through the expansion of cultivated land. This expansion was stimulated by the sharp increase in agricultural prices and the improved market opportunities. Between 2008 and 2011, cultivated rice area grew 4.5 percent annually. After agricultural prices leveled and even began declining, land expansion slowed down. In 2012, cultivated paddy area grew only 0.4 percent and in 2013 even declined 0.4 percent (World Bank 2014b). Without increasing farm profitability through higher productivity, it will be very hard to increase agricultural profits in the future. 183. Even at higher farm profitability, repeating the past land expansion in the future is not feasible given the RGC’s plans to revert deforestation. The target is to have 60 percent of land area forested, from the current 57 percent (World Bank 2014e).Thus growth of cultivated land in the future is likely to be just a fraction of what it was in the past. This is good news, because past land expansion seemed to have contributed to land degradation. Degradation has been especially high in the upland areas used for cassava production, for which little fertilizer has been applied to maintain soil nutrients. Land degradation can be defined as the persistent reduction or 42 The Phnom Penh Post “Waterways May Reduce Export Costs,” March 12, 2013, quoting the Alliance of Rice Producers and Exporters of Cambodia (ARPEC). 77 loss of land ecosystem services. Using global information on the changes in the normalized difference vegetation index between 1982 and 2006, which serves as a proxy for land degradation, Cambodia was found to be among the countries with the largest share of land degradation hotspots. About 55 percent of its crop land and 42 percent of its total land are estimated to have degraded in terms of net productivity and vegetation cover (Le et al. 2014). About 60 percent of Cambodians are estimated to reside in land degradation hotspots, and human-induced activities are the source of most degradation. 184. Second, the past growth was accompanied by an increase in the average farm size and an increasingly unequal agricultural land distribution. Large farms are becoming larger and small farms are becoming smaller.43 This change in distribution of farmland occurred independently of economic land concessions. The increasing disparity is accompanied by land consolidation and more commercial farming. Even though no general economies of scale in land use are found in Cambodia, economies of scale in returns to labor have increased as a result of mechanization, and larger farms seem to be integrated more easily into modern food value chains. 185. Thus, the main concern is that the returns to labor for these small farms will remain low. For example, the returns to labor in small farms producing rice during the wet season are estimated to be below the prevailing wage rate. The returns to a day of labor for wet season rice production increase from $2.86/day for small farms to $12.67/day for large farms. Dry season rice production has the highest impact of farm size on returns to labor, at $6.15, $8.27, and $22.83 for small, medium-sized, and large farms, respectively. This has a direct impact on the ability to attack poverty, because wet season rice still constitutes 83 percent of total cultivated rice land, comprising most smallholder farmers. Very small farms are less willing and able to mechanize production processes, falling behind larger farms in income generation. 186. Another concern is that while farm size increases, landlessness may increase. If landlessness is the result of rural households moving out of agriculture into other profitable employment opportunities, then landlessness is not a major issue. Yet when landlessness is associated with poverty and vulnerability, future growth strategies need to address the impact of increasing landlessness on poor and vulnerable households through a combination of measures including safety nets, insurance, and capacity building. 187. Third, one of the most worrying aspects of the past growth of agriculture in Cambodia is that, despite growing yields and production, per hectare profitability of farming has increased slowly. The average gross margin per hectare in real terms increased by only 3.4 percent between 2005 and 2013, due to the increase in yields and the gradual shift of production to more diverse mix of crops more profitable than wet season paddy. Yet there is more room to increase yields, especially of paddy, which would need to come from more efficient use of resources rather than just higher use of intermediate inputs, labor, and machinery. For farming to be attractive, the increase in revenue needs to exceed the increase in costs. In the situation with limited potential for farmland expansion, this puts emphasis on productivity, profitability, and competitiveness to underpin future growth. 188. Achieving higher agricultural productivity depends on many factors, with investments in the delivery of core public goods and good quality of public programs among the key factors. But most productivity-enhancing public programs such as agricultural extension, 43 It should be noted that even though farm size can increase over the next decade or so, the vast majority of Cambodian farms will remain less than 5 ha in size. 78 agricultural research, vocational training, and input quality control are underfunded in Cambodia. Farmers rely mainly on extension advice from the private sector and NGOs, which are often limited to specific commercial products. Some inputs are fake and their quality control by public agencies is not adequate or present at all, reducing farmers’ willingness to purchase more inputs. Other programs such as irrigation receive sufficient funds but are mostly used to construct primary structures making Cambodia’s irrigation coverage the lowest in Asia (recall Table 5). The effectiveness of these investments and impacts on farm incomes appears to be very low. 189. Fourth, during the past 10 years, the cost of labor has increased tremendously, creating a number of adjustments in production, including accelerated mechanization, reduction of labor use, and adoption of labor-saving technologies. This is consistent with the key patterns of agricultural transformation from a traditional to a modern agriculture sector. However, if continued, the increasing cost of labor will also imply that agriculture will not be able to absorb the increasing labor force and therefore new sources of labor demand have to be identified not only in urban areas and outside of the country (the two traditional “solutions” for surplus labor in agriculture), but in the rural nonfarm sector as well. Outside the garment and footwear industry and construction sectors, nonfarm job creation has been slow. 190. Finally, the past growth of agriculture was mostly related to the production sector; very little value added was generated by agriculture and in fact an enormous surplus of commodities was exported in raw form. The RGC has made a concerted effort to change this situation in the case of rice, but the objective of one million tons export by 2015 is not likely to be achieved. Contract farming and productive partnerships between farmers and agroenterprises are very limited; sustainable farmer organizations are few despite the promotion of agricultural cooperatives. Yet the development of agribusiness and, more generally, of an agri-based rural, nonfarm economy must be an indispensable element of any future growth strategy. 79 7. DRIVERS OF FUTURE GROWTH 191. Some drivers underpinning future agricultural growth in Cambodia will be similar to those of the past. For example, GDP growth will increase domestic demand for more diverse food and encourage further outmigration of labor from agriculture. According to the World Bank (2013a), Cambodia’s GDP will continue to expand rapidly up to 2030, though at a slower pace (slightly over 7 percent) than over 2000-2010 (8.2 percent). GDP is expected to increase in real terms from its present value of $13 billion to $23 billion by 2020, and to nearly quadruple to $50 billion by 2030 (Table 39). With population growth expected to decline from 1.4 percent to 1.1 percent over this projection period, real GDP per capita would increase from $957 in 2012 to $2,733 by 2030. Table 39: Selected economic projections, Cambodia, 2012-2030 Indicator Future 2012 2015 (Short- 2020 2030 (Long- Growth (Baseline) term (Medium- term rate (%) Projections) term Projections) Projections) Real GDP, $ billion 7.2 14.2 17.5 24.7 49.5 Population, million 1.1 14.8 15.3 16.2 18.0 Real GDP/capita, $ 6.1 957 1,140 1,526 2,733 Urban population, 5.1 3.3 3.8 4.8 8.0 million % of urban population 4.0 22 25 30 45 Source: Own estimate based on World Bank 2013a. 192. Rising global and regional demand for food will continue to provide markets for increased Cambodian exports (Table 40), which will be facilitated by the continuation of market- oriented agricultural and trade policies. Public investments will help continue to reduce the costs of doing business and attract private investments to create jobs. 193. Yet there will be important differences. First, it is expected that future economic growth in Cambodia will be accompanied by accelerated urbanization. Total population will rise from 14.8 million in 2012 to 18.0 million by 2030; assuming an urban population growth rate of 5.1 percent (consistent with the past decade’s growth), the urban population will more than double from 3.3 million to 8 million over the projection period. This implies Cambodia’s rapid urbanization rate increase from 22 percent to 45 percent (Table 39). 194. Rapid urban population growth has a number of implications for Cambodia’s agri- food system. An increasingly urban population will need to be fed by a relatively smaller rural population. While in 2012, Cambodia needed one agricultural worker on average to feed 2.8 persons, in 2030, each agricultural worker will need to feed 4.5 persons. For the urban population, the food consumed will be richer in protein and micronutrients and will require more stringent quality standard; consumers will want food that is more convenient to prepare and consume. Increasing urbanization also implies more pressure on agricultural land in peri-urban areas, often the most fertile land. Food distribution systems appropriate for urban areas need improved marketing and distribution infrastructure. An increasing amount of food will be distributed via modern outlets and supermarkets, will be more processed, and will be sold in a variety of packaged forms. This will put pressure on very small farms, which often find it difficult to integrate into 80 modern value chains. Agroindustry will also need to develop to provide enhanced features such as storability, convenience, improved packaging, and more diversified products. Table 40: Caloric food demand projections, world and selected regions/countries Region/Country 2009 2030 Annual growth rate Actual Projections % World 2,831 3,050 0.4 East Asia 2,694 3,190 0.9 East Asia (less China) 2,663 2,965 0.5 China 3,036 3,739 1.0 Cambodia 2,382 2,667 0.5 Japan 2,723 2,613 -0.2 Lao PDR 2,377 2,662 0.5 Malaysia 2,902 3,249 0.5 Mongolia 2,434 2,725 0.5 Myanmar 2,493 2,792 0.5 Philippines 2,580 2,889 0.5 South Korea 3,200 3,583 0.5 Thailand 2,862 3,205 0.5 Vietnam 2,690 3,012 0.5 Note: Caloric food demand refers to the amount of food expressed in kilocalories (kcal) per day, available for each individual during the reference period. Source: Jamora 2014 using FAOSTAT. 195. Due to increasing income and urbanization, diets will be more diversified, and richer in protein and micronutrients. Not only in Cambodia but also in East Asia and the world as a whole, more animal and fish products and more processed foods will be consumed (Figure 37). More people will be aware of health issues related to nutrition and food safety (cholesterol, diabetes, allergies, chemical residues, antibiotics in animal feeding, etc.). Rice consumption per capita in Cambodia will likely remain high but lower than its current level, implying an increasing availability of rice for export. Rice quality will have to increase not only for export markets but for domestic and urban markets as well. 196. Rising ASEAN and global demand for food will offer market outlets for Cambodian exporters, but penetrating foreign markets will be more challenging . The further increase in cross-border trade will require Cambodia to have an adequate capacity and network of agreements in place with major trading partners (especially China) for food safety and quality standards, phytosanitary and quarantine, double tax agreements, transit, and harmonized trade and commerce legislation (e.g., anti-dumping, labeling, trademark protection). Cambodia’s agri-food industries must become competitive with those of neighboring countries. Adequate physical and institutional infrastructure is needed to achieve such competitiveness, including more stable and cheaper access to energy, improved transportation network and logistics systems, highly effective information and communication systems, and financial institutions to support trade and investment. 81 Figure 37: Value of the projected food demand in East Asia, $ billion, 2030 Source: Jamora 2014 using data from IMPACT/IFPRI model and OECD. 197. Exporting much more rice, especially of low quality, will be difficult . Total per capita consumption of rice is projected to decline by 2030 due to global income growth and substitution effects (Table 41). Future global consumption rates are heavily affected by future rice consumption in China and India, but consumers are projected to demand more higher-quality rice (Slayton and Muniroth 2012b). Structural changes in diets will continue to reduce the relative importance of rice in favor of meat, fruits, vegetables, fish, and milk (Alexandratos and Bruinsma 2012; Sharma 2014). Table 41: Global rice demand projections (kg per capita) per year Sources/assumptions 2020 2025 2030 FAPRI 2005 65 62 61 FAO 2003 63 62 61 Abdullah et al. 2008 Scenario 1: China -0.33%/India -0.93% 59 58 57 Scenario 2: China -1.77%/India -1.44% 54 51 48 Scenario 3: China -3.55%/India -3.03% 49 46 43 Source: Jamora 2014. 198. Future agricultural TOT is projected to be less favorable than in the last decade. As shown in Table 38, global food prices increased significantly in the past decade. In the future, however, prices of most agricultural products are projected to decline (Table 42). Prices of grains are projected to decline 12 percent in real terms between 2010 and 2025, with rice prices in 2025 projected to decline by 20 percent compared to 2015. On the other hand, energy prices are expected to keep increasing, though fertilizer prices are expected to fall in the future. Overall agricultural TOT, defined as the ratio of prices of agricultural outputs to prices of inputs for agricultural production proxied by the simple average of fertilizers and energy, is projected to decline. Rising costs of energy will drive up costs of irrigation and mechanization, while declining agricultural prices will put pressure on profits and competitiveness. In this context, profitable farming systems will depend on access to modern knowledge and technologies and the efficient use of more 82 expensive inputs, along with resilience to weather variations and climate shocks. In other words, high attention will need to be paid to TFP. Table 42: Global agricultural terms of trade, index projections (in real 2010$)44 2010 2015 2020 2025 Agricultural prices 100 97.4 90.1 83.1 inc. Food 100 102.2 93.1 84.3 Fats and oils 100 106.4 96.1 86.2 Grains 100 104.3 96.0 87.9 Other food 100 94.9 86.5 78.4 inc. Beverages 100 87.5 77.2 67.9 inc. Raw materials 100 78.7 75.2 71.6 Fertilizer prices 100 89.4 81.2 73.3 Energy prices 100 120.0 113.5 108.9 Agricultural TOT 1.00 0.930 0.926 0.912 Source: World Bank Commodity Price Forecast, July 2014. 199. Future agricultural growth will depend less on land expansion . Reaching the RGC target of 60 percent of land under forest requires afforestation programs, strict compliance with land and forest protection laws and regulations, attention to sustainable management of natural resources, and improvements in the land tenure system. The sustainability of environmental resources such as land and forests has been neglected in the past decade of growth in Cambodia. Further depleting natural resources for sustained growth is not a good option. Failing to take into account the environment, even at the early stages of growth, can be a big mistake, and one that is extremely expensive to fix in the future (World Bank 2013a). Cambodia is endowed with rich environmental resources that have benefited its economy and people greatly, and it is important to ensure that this remains the case in the future. 200. The agricultural labor force will continue to decline, while agricultural wages increase. As in other countries, Cambodia’s rural youth are increasingly choosing employment in sectors other than agriculture. Labor shortage provides incentives for increased mechanization. Outmigration might also accelerate the process of increasing farm size. As the agriculture sector transforms to a more commercial and competitive industry, it will attract more investment and support more semi-skilled and higher-paid employment. Yet without strengthening land tenure security and reducing the cost of capital, adoption of farm mechanization in Cambodia will be slow, putting upward pressure on farm production costs. Modern farm machinery and new farming technologies, such as precision farming, are increasingly available to Cambodian farmers, but they can be more efficiently used on larger farms. The existing land tenure insecurity precludes this from evolving at a natural pace, as it creates distortions in the land tenancy market that constrain small absentee land owners from renting their land to capable and entrepreneurial farmers. 44 Agricultural price index is the sum of food, beverages and raw materials. Fats and oils include coconut oil, copra, groundnuts, palm oil, palm kernel oil, soybeans, soybean meals, and soybean oil. Grains include barley, maize, rice, sorghum and wheat. Other food includes bananas, oranges, sugar, meats, and fish. Beverages include cocoa, coffee, and tea. Raw materials include cotton, rubber and tobacco. Fertilizers include DAP, urea, TSP, potassium, and phosphate rock. Energy includes coal, crude oil, and natural gas. 83 However, land speculation, which could be caused by low land taxes stemming from outdated land valuation prices, is another potential motivation for increasing farm size. 201. More budget resources will be available to spend on agricultural programs in Cambodia but more resources will also be required for other purposes to sustain growth and boost shared prosperity. Competition for public funds will motivate ministries to use funds more efficiently; those doing so will receive and spend more on their programs. Fiscal discipline will require clearly definition of what to finance (public goods) and what not to (private goods) from the budget. Except irrigation, many critical agricultural programs (such as research, extension, food safety, input quality control, etc.) are underfunded. Improvements in the quality of agricultural programs and the ability to attribute their impacts on agriculture sector outcomes through impact evaluations will determine the funding that MAFF and Ministry of Water Resources and Meteorology (MOWRAM) receive from the national budget and donors to support further agricultural development. 202. Finally, more attention needs to be paid to implementation. Most of the messages above are not new to Cambodia and are rightly identified in the RGC’s new five-year Rectangular Strategy III. The key challenge for Cambodia is implementation, which will require increased inter-ministerial coordination, leadership, and most importantly, monitoring of specific targets linked to those reforms. Successful implementation of the reforms could help Cambodia reach a GDP growth path of 8 percent and sustain its successful agricultural transformation. The driving forces discussed in this chapter and their impacts on future agricultural growth are summarized in Box 7. 84 Box 7: Driving Forces Underpinning Future Agricultural Transformation in Cambodia 1. Economic growth in Cambodia a. Stimulate more demand for food b. Pull labor out of agriculture c. Increase labor costs d. Lower cost of capital and increase farm demand for mechanization e. Create a larger fiscal space for public investments in agriculture 2. Urbanization in Cambodia a. Put pressure on land available to agriculture, converting land to urban purposes b. Lead to declining per capita rice consumption c. Create higher demand for more diversified (change of diet) and processed foods d. Require more attention to food safety and quality e. Create challenges to integrate small farms into value chains f. Increase demand for farm labor force 3. Rising ASEAN and global demand for food a. Lead to declining per capita rice consumption of low-quality rice b. Create more demand for higher-value, better processed and marketed rice c. Raise demand for higher-value, processed foodstuff d. Require high attention to safety and quality to comply with the rules of buyers and avoid non- tariff barriers 4. Declining global agricultural TOT a. Without increasing TFP, lower farm prices will reduce farm profits and incomes b. Call for more and better spending on core agricultural public goods 5. Deceleration of agricultural land expansion in Cambodia a. Increase pressure for sustainable management of natural resources (land, water, forests) b. Bring more attention and urgency to reduced land degradation c. Call for building institutions and clear rules for natural resource management d. Bring urgency to raise agricultural productivity to sustain farm incomes 6. Fiscal discipline a. Despite a larger fiscal space due to continued economic growth, more budget resources will be required for various purposes. Public in general would also require better transparency and accountability on the use of funds. This would lead to stronger fiscal discipline. b. Define the role of government and key public investments to support agricultural transformation under fiscal constraints 85 8. A LONG-TERM VISION FOR CAMBODIAN AGRICULTURE 203. The RGC’s long-term vision, including a 2030 vision for the country, is focused on rapid development that is resilient, inclusive, and sustainable. Continuing its strong economic performance will be a central pillar to achieve the country’s aspirations. With the appropriate policies, Cambodia’s economy can more than triple in size by 2030. Beyond economic growth, the aspirations for Cambodia’s progress towards 2030 are threefold: (i) for progress to be resilient to shocks such as economic crises and natural disasters; (ii) for progress to be inclusive, ensuring continued poverty reduction, shared prosperity with an expanding middle class, and social cohesion; and (iii) for progress to be environmentally sustainable, securing future benefits from the country’s rich land, forest, and water resources for future generations. Overall, the RGC’s long- term vision mentioned in the strategic documents consists of seizing opportunities to make Cambodia a more prosperous, inclusive and successful nation (World Bank 2013a). 204. Agricultural transformation is the key to realizing a 2030 Vision. The agriculture sector was instrumental in stimulating growth, reducing poverty and boosting shared prosperity in the past and it remains a powerful source of pro-poor growth in the future, given its large share in the economy, labor force, and exports. The sector also has an important role in achieving environmental sustainability and resilience objectives as it is the largest user of land, water, and other natural resources and is the most vulnerable to climate change. The Agricultural Sector Strategic Development Plan 2014-2018 prepared by MAFF aims to continue the past growth in agricultural value added at 5 percent in the future. Achieving the 5 percent growth is possible but challenging as can be judged from the recent growth slowdown. In 2013, the growth in agricultural value added was only 1.6 percent and it is expected to be even smaller in 2014. 205. Four lessons from agricultural transformation around the world are relevant to the formulation of Cambodia’s vision: a. At early stages of development, agricultural growth is the main engine of poverty reduction because most of the poor are in rural areas. b. TFP growth is a main source of long-term agricultural growth. TFP requires not only land, labor, and capital accumulation, but also technology, innovation, efficiency, human resource capacity, and governance. TFP growth contributed 40-70 percent of agricultural output growth in Asia over the last 50 years (Fuglie 2010). c. Increasing urbanization presents challenges foe which traditional farming systems are not well prepared. Helping small farmers integrate into modern value chains is a feasible response for governments to these urbanization challenges. d. As an economy moves closer to middle-income status, development of the rural nonfarm sector (including agri-food manufacturing, services, marketing, and logistics) becomes increasingly important in closing the gap between rural and urban incomes. 206. Cambodian agriculture will need to use these lessons to continue its successful transformation. As the country moves to middle-income status, it will continue to experience agricultural transformation, whereby the economic structure of its society will change from one based on agriculture to one based on agribusiness, industry, and services. During this process, agriculture’s importance in the economy, as measured by its share of labor and GDP, will decline (Box 8). At the same time, agriculture will become more productive: returns to agricultural labor and land will increase and agricultural GDP will grow in the absolute terms. 86 Box 8: Indicators of Agricultural Transformation Agriculture becomes less “important” in the economy  Declining share of agriculture in labor  Declining share of agriculture in GDP Agriculture becomes more productive  Increasing agricultural labor income  Increasing agricultural GDP per unit of labor and land 207. By 2030, Cambodia’s agriculture needs to become “less important” but “more productive.” The challenge is to continue the past agricultural growth of 5 percent per year via higher resource use efficiency rather than cultivated land expansion. With 5 percent annual agricultural growth by 2030, the share of agriculture in GDP is projected to decline to 23 percent in current prices and 17 percent in real prices (Table 43). The labor share in agriculture will decline to 32 percent compared to 51 percent in 2012. Cambodia may reach the status of a “transitioning economy” based on the framework presented in Table 12. More productive agriculture (land and labor productivity) would not only increase farm incomes in absolute terms, but would also significantly reduce the income gap in current prices between agriculture and non-agriculture to 1.6 by 2030, starting from a high ratio of 2.1 in 2012. Table 43: Indicators of agricultural transformation at 5 percent agricultural growth Indicator Unit Past Future 2012 2015 2020 2030 Growth Growth (Baseline) (Short (Medium (Long Rate Rate term) term) term) (%) (%) Agriculture as a share % 1.3 -2.1 34% 32% 28% 23% in GDP (in current prices) Agriculture as a share % -1.9 -2.1 26% 24% 22% 17% in GDP (in real prices) Labor share in % -1.5 -2.6 51% 47% 41% 32% agriculture Agricultural $/capita 4.5 6.5 1,210 1,460 2,000 3,700 GDP/agricultural labor force Agricultural $/ha 0.3 3.9 1,325 1,485 1,800 2,640 GDP/cultivated land area Ratio of nonfarm Ratio -1.2 -1.6 2.1 1.9 1.8 1.6 GDP/nonfarm labor and agricultural GDP/farm labor Notes: Based on assumptions indicated in Annex 4. Source: Own estimates. 87 208. A 5 percent agricultural GDP growth is not assured, however. Future growth may well be only 3 percent, especially if needed policy actions and investments are not made (“doing nothing”). The costs of doing nothing would be very high for Cambodians. For illustration purposes only, it is estimated that the 3 percent growth in agricultural value added would reduce overall GDP,45 make agriculture less productive sector, keep more people in the sector,46 and, most importantly, increase the income gap between farm and nonfarm workers (Table 44). The cumulative reduction in agricultural value added by 2030 can reach 29 percent, and the income gap between agriculture and non-agriculture can stay the same, at 2.1, compared to 1.6 at the 5 percent growth in agricultural GDP. Thus, achieving the 5 percent agricultural growth is important. Table 44: Changes of indicators of agricultural transformation at 3 percent compared with 5 percent growth Selected indicators Future Growth Rate 2020 (Medium 2030 (Long- 5% 3% Term) (%) Term) (%) GDP growth 7.2 6.0 -8.6 -18.3 Agricultural GDP growth 5.0 3.0 -14.3 -29.3 Agriculture as a share of GDP -2.2 -3.0 -6.4 -13.7 Agriculture as a share of total -2.7 -2.3 3.3 7.7 labor force Agricultural GDP/ agricultural 6.4 4.0 -16.7 -33.7 labor Ratio of nonfarm GDP/nonfarm labor and agricultural GDP/farm -1.54 -0.02 15.0 32.7 labor Source: Own estimates. 209. A long-term vision for Cambodian agriculture would need to aim to continue higher growth but with more emphasis on intensification and farm income convergence and lower costs to the environment. The proposed vision statement can be the following: “A sustainable and competitive agriculture sector that fosters productivity growth and contributes to increased farm and nonfarm incomes.” 210. This vision statement is largely in line with the policy goal of the Agricultural Strategy Development Plan for 2014-2018, which is “enhanced agricultural productivity, diversification and commercialization, thereby contributing to reduced poverty and accelerated economic growth while, at the same time, ensuring environmental protection and sustainable natural resource management.” The additional point emphasized by the proposed vision statement is the increase 45 In 2013, the share of agricultural GDP in total GDP was 24 percent. Thus, in the short run a 2 percent reduction in agricultural GDP would lead to a 0.48 percent reduction in overall GDP. The longer-run effect is likely to be much higher given the strong links between agriculture and trade, transportation, and food processing sectors in Cambodia. For the hypothetical calculations of the implications of the 3 percent growth in agricultural GDP, the long-term GDP growth is reduced from 7.2 percent to 6.0 percent. 46 The impact of lower agricultural growth on agricultural labor is not known. It is, however, likely that lower growth in the long-term agricultural GDP would keep more people in the sector than 5 percent growth. At 3 percent agricultural growth, the agricultural labor force is assumed to decline by 1 percent annually, compared to 1.4 percent at 5 percent agricultural growth. 88 in farm incomes and their convergence with nonfarm incomes, which would indicate successful structural transformation. 211. The key elements of the proposed Vision are sustainability, competitiveness, productivity, and income growth: a. Sustainability: Sustainable agriculture includes many dimensions, including environmental sustainability. Progress in achieving environmental sustainability will be judged by the success in securing future benefits from the country’s rich land, forest, and water resources for future generations. It implies reversing the expansion of cultivated land at the expense of forest and other environmentally sensitive land, increasing soil organic matter, reversing land degradation through good agricultural practices and climate-smart agriculture, and increasing agricultural water productivity. b. Competitiveness: Having farms able to produce at competitive prices will not be enough to maintain high agricultural growth rates and increase farm incomes. Even future farm competitiveness is at risk if policy actions do not address pertinent constraints, including poor infrastructure, weak governance, weak quality and safety control, low technology, limited capacity and human resources, and difficulty in accessing credit and doing business. These constraints harm not only farms but also agribusinesses. Competitiveness implies a production and distribution system oriented towards the market and meeting consumers’ demand effectively by providing higher value. Competitive products are not necessarily cheaper, but meet the preferences and budgets of consumers, using efficient value chains. Competitiveness is based on comparative advantage, productivity, and profitability, but goes one step further: rather than focusing only on cheaper costs, it looks at higher value added, quality, and safety. c. Productivity growth: Consistent with the insights of agricultural transformation, a more dynamic agriculture is accompanied by higher productivity growth. The agricultural land expansion that took place in previous decades is not a viable option in the future. Productivity growth needs to become a central element of agricultural development in the country. Growth of agricultural-based activities will have rural nonfarm effects and imply nonfarm employment. Higher economic growth in the agriculture sector will in turn contribute to higher GDP. Combined with a reduced population growth rate and reduced growth of agricultural labor, higher GDP per capita in agriculture will reduce poverty and improve living standards in rural areas. d. Income growth: A desirable agricultural transformation implies growing incomes for farmers and improved integration with the rest of the economy, primarily through strong linkages with the agribusiness sector, reflected in growing returns to agricultural labor. As a result of labor migration from agriculture to other sectors, overall productivity growth, and a more diversified (profitable) mix of agricultural products, the gap between agricultural and non-agricultural incomes will decline. The growth of agricultural income will increasingly depend on the development of agroprocessing, storage, trade, food service, production services, and agri-tourism. The income growth of the smallest farms may lag behind that of larger farms, calling for special attention to this group of farmers. 212. The key elements of the proposed vision require indicators and targets for agricultural transformation. Targets indicate what is intended and possible to achieve as a result of the agricultural strategy’s implementation. The proposed indicators and targets are presented in 89 Table 45. They are based on the underlying data summarized in Annex 4. Targets are listed for the short term (3 years), medium term (8 years), and long term (18 years). Each target is accompanied by indicators that allow progress (or lack thereof) toward the target to be measured. 90 Table 45: Proposed indicators and targets for a Long-Term Vision for Cambodian Agriculture Annual Growth 2012 2015 2020 2030 Rate Vision’s Indicator Unit Past Future Baseline Short Medium Long Comments Elements (2004- Term Term Term 2012) Forest cover % of land -1.1% 0.4% 56% 57% 58% 60% The 2012 baseline and the 2030 target are from NDSP. Past growth rate is from MAFF. Degraded million ha NA -3.8% 7.8 6.9 5.7 3.9 Baseline data based on Sovuthy land Pheav, Director, Department of Ag Land Resources Management. Targets to be achieved through soil conservation, reforestation, and soil Sustainability nutrient management programs. Soil organic % NA 8.0% 1.0% 1.3% 1.9% 4.0% Baseline data from Rice Production matter in Cambodia edited by Harry J. Nesbitt SRI. Targets to be achieved by no tillage, intercropping, organic fertilizer, and other methods. Water liters / kg NA -1.2% 2,500 2,409 2,264 2,000 Baseline based on Wokker et al. productivity rice 2014. Targets to be achieved by more efficient use of irrigation. Agricultural $ million 19% 14% 400 587 1,113 4,000 Past growth is based on Exports stat.wto.org. Agribusiness % of Ag. 6.4 9.7% 27.4% 31.3% 38.9% 60.3% For agribusiness to grow from 30% GDP GDP to 60% of agricultural GDP, growth Competitiveness of agribusiness GDP has to be 9.7 % (if agricultural GDP grows at 5%). Capital-ag $/ha NA 3.9% 1,000 1,122 1,361 2,000 Baseline based on FAOSTAT. land ratio Ag land Ag. GDP 0.3% 3.9% 1,325 1,485 1,800 2,640 Baseline based on data from WDI. Productivity productivity /cultivate Target derived from assumption on 91 Annual Growth 2012 2015 2020 2030 Rate Vision’s Indicator Unit Past Future Baseline Short Medium Long Comments Elements (2004- Term Term Term 2012) d area, ag GDP growth of 5% and a $/ha moderate increase in cultivated land of 1% per year. Agricultural Ag. GDP/ 4.5% 6.5% 1,210 1,460 2,000 3,755 Baseline based on data from WDI labor ag labor and NIS. Target derived from productivity force, assumption on ag GDP growth of $/person 5% and decrease in ag labor by 1.4%. Returns to $/day 12.0% 6.3% 6.0 7.2 9.8 18.0 Baseline based on data from the agricultural 2013 survey. Target derived from labor the conservative assumption of future growth to be only a half of the past growth. Income Gap between Ratio -1.2% -1.6% 2.1 1.9 1.8 1.5 Derived from the assumptions on nonfarm per ag GDP growth, ag labor growth, capita income non-ag GDP growth, and non-ag and farm per labor growth. capita income 92 9. OPTIONS FOR SUPPORTING A LONG-TERM VISION FOR AGRICULTURE 9.1. Introduction 213. Realizing the Long-Tem Vision for Cambodian Agriculture requires a conducive policy environment and high-quality public programs. These policies and programs must take into account the factors underpinning future agricultural development described in Chapter 7. A number of priority areas exist, as well as trade-offs. Analysis of the farm budget and profitability database for Cambodian agriculture offers a useful platform to compare policies and programs in terms of their impact on farm gross margins, net present value (NPV), returns to agricultural labor and extra demand for labor, demand for fertilizers, and impacts on competitiveness. 214. Four sets of policy scenarios are developed to support Cambodia’s agricultural transformation over the next twenty years and movement toward a Vision 2030: a. Environmental sustainability: Ensuring sustainable land expansion through land use planning; reverting agricultural land to forest, aiming to reach the country’s target of 60 percent forest cover by 2030; and maintaining soil fertility through optimal use of available land and use of fertilizer and organic matter. b. Productivity growth: Increasing adoption of improved technologies through extension services and use of more and better quality inputs; and investing in effective irrigation by ensuring strong linkages with extension programs and enforcement of operations and maintenance guidelines. c. Competitiveness: Analyzing the effect of price changes of inputs and outputs and making adjustments to maintain competitiveness of Cambodian agriculture. d. Income growth: Increasing farm households’ income through land consolidation (i.e., higher total available income per capita because of larger assets), higher productivity, and better informed choice of higher-value crops and technologies. 9.2. Scenarios for Environmental Sustainability 215. Land expansion has been a major factor of agricultural growth in Cambodia. Over 2002-2012, cultivated land for crops covered by the survey (rice, maize, cassava, and vegetables) increased by 50 percent. The growth in upland crops was even more dramatic: cultivated land increased by a factor of 2 for maize and 13 for cassava. Even for rice, where the increase in land was more modest than the average, the contribution of cultivated land to rice production growth was over 40 percent. 216. For upland crops, especially cassava, FGDs and expert interviews revealed that cultivated land expansion comes at the expense of forest land. Moreover, as soil nutrients are mined and not replenished, land expansion results in soil fertility degradation. In other areas of the country, cultivated land expansion puts pressure on water resources and may encroach on pasture land. This consideration, in line with any biodiversity conservation program that Cambodia may have, would argue against continuing a development strategy based on land expansion. 217. The majority of farmers interviewed during the 2013 survey believed that their total cultivated land area will not change. Farmers were relatively conservative regarding their expectation of cultivated land areas in the next five years. More than half of interviewed farmers 93 expected no change in the area of land cultivated with vegetables and maize (Figure 38). About 10 percent predicted a decrease, and one out of three anticipated an increase. Cassava showed a peculiar pattern, with almost equal numbers of farmers expecting an increase, no change or a decrease in cultivated areas. Figure 38: Perspectives on the possibility of change in cultivated areas by crop, Cambodia Source: 2013 survey. 218. The response was determined by location. Respondents in the Tonle Sap and Mekong regions were not positive regarding the possibility of increasing cultivated land. In other regions, the option of land expansion is still open, especially for upland crops. During interviews with different stakeholders, many people were aware of the encroachment of cassava crop cultivation into forest land. This is a relatively easy way to expand agricultural land, but its practice is receiving increasingly more attention from policy makers since it puts more stress on already fragile ecosystems. For rice, the trend was “no change,” although more than 20 percent of respondents were pessimistic, expecting a decrease in area cultivated. 219. As land expansion cannot play the same role in production growth in the future, alternative growth strategies such as yield increase, crop and technology choices, and competitive market advantage will have to play a more important role. The analysis of environmental sustainability is divided into two types: (i) “weak environmental sustainability,” whereby the increase in cultivated area for agricultural production continues but at a slower pace, and (ii) “strong environmental sustainability,” whereby cultivated land area stays the same or decreases over time. Weak environmental sustainability 220. Three scenarios (L1, L2, and L3) are developed. They consider a slower cultivated land expansion growth rate compared to past trends and calculate the resulting total production, gross margins, and demand for inputs. Except land, all other factors are held constant at their 2013 observed values. To make the simulation results comparable, the overall increase in cultivated land for the selected crops is set at about 450,000 ha by 2030. This leads to a total cultivated land area of 4.01 million ha in 2030, an increase of 11-12 percent depending on the scenario, compared to 3.58 million ha for the selected crops in 2012. 94 221. The scenarios’ assumptions are summarized in Table 46. Scenario L1 considers maintaining upland crops and vegetables’ areas constant at their 2013 levels and the land area under rice growing at 25 percent of its past growth rate. This scenario focuses on rice production, reinforcing its role in the economy at the expense of diversification. Scenario L2 assumes a land expansion of about one-tenth of its past growth rate for all crops. This scenario treats all crops equally. Scenario L3 sets upland crops and vegetables’ cultivated land expansion at slightly less than 14 percent of their past growth and keeps cultivated rice land at its current level. The policy then emphasizes diversification toward expansion of upland crops, which may present an environmental challenge. Table 46: Assumptions about land expansion for Environmental Sustainability scenarios L1-L3 Baseline L1 L2 L3 Cassava 0.0% 0.0% 10.0% 14.0% Maize 0.0% 0.0% 10.0% 14.0% Wet season rice 0.0% 25.0% 10.0% 0.0% Dry season rice 0.0% 25.0% 10.0% 0.0% Vegetables 0.0% 0.0% 10.0% 14.0% Source: Own estimates. 222. In scenario L1, rice production increases by 17 percent from the baseline of 9.6 million tons to 11.2 million tons. This is the result of land expansion for rice by one-fourth of its past growth rate. Gross margins for rice increase by 15.8 percent, from $756 million to $876 million (Table 95A47). The overall gross margin for the combined five crops increases by 11.2 percent, from $1.02 to $1.19 billion (Table 47). 223. In scenario L2, the assumption is that land expansion for all crops continues at 10 percent of past growth rates. Rice production increases by 6.4 percent from the baseline 9.6 million tons to 10.17 million tons (Table 96A). Assuming cassava maintain its level of productivity, a strong assumption due to the low use of organic and chemical fertilizer for this crop, cassava production increases by 74 percent, from 6.3 million tons to 10.96 million tons; maize production increases from 849,000 tons to 1 million tons (17.8 percent increase). Rice is still the most important crop in terms of gross margins even through its share falls by 7 percent, from its 71 percent baseline. The overall gross margins for the five crops increase by 18.2 percent (Table 47), reaching $1.26 billion by 2030. 224. In scenario L3, which focuses on upland crops and vegetables production, rice production remains similar to the baseline at 9.6 million tons. Cassava production is more than twice the baseline, rising from 6.3 million tons to 13.6 million tons; and maize production increases by 25.8 percent, from 0.85 million tons to 1.07 million tons (Table 97A). The share of upland crops and vegetables in overall gross margins increases from 29 percent to 42 percent, reaching $542 million. The share of gross margins from rice drops by 13 percentage points, from 71 percent to 58 percent. Overall gross margins increase by 21.6 percent (Table 47), reaching $1.30 billion, the highest among the three weak environmental sustainability scenarios. 47 Table numbers with “A” are presented in Annex 4. 95 Table 47: Simulation results for Weak Environmental Sustainability scenarios L1-L3 Simulation Change NPV Returns Change in Change in Likely in Gross ($ to Ag. Demand Demand Effect on Margin, million) Labor for Ag. for Competiti $ million ($/day) Labor Fertilizers, veness (%) million tons/year (DRC) days (%) (%) Baseline 1,023 10,220 6.41 166.5 464,861 L1 Rice areas increase by 11.2% 10,644 6.42 11.2% 17.1% Neutral 25% of past land expansion L2 Equal increase of all cultivated 18.2 10,882 6.70 13.1% 8.2% Neutral land by 10% L3 Increase of upland areas by 21.6 10,958 6.86 13.7% 2.6% Neutral 14% Source: Own estimates. 225. These simulations suggest that the policy toward crop diversification (scenario L3) results in the highest gross margin and NPVs. Moreover, diversification toward upland crops favors more regionally-balanced development. However at the same time, there is a serious danger of soil degradation and forest land loss because of its focus on upland crops, particularly for cassava plantation (more than 400,000 ha of new upland brought into agricultural production). The scenario focusing only on rice (scenario L1), at the current level of per-hectare gross margins, results in the worst economic outcome. 226. For each of the three scenarios, demand for labor is roughly comparable, 11-14 percent higher than the baseline. On average, 20 million days of additional demand for labor are generated, which is related to the increase in cultivated land area. By 2030, the total demand for labor in scenario L1 is expected to reach 185 million days; the demand in scenario L2 is higher at 188 million days; and the demand in scenario L3 is 189 million days due to the labor requirement for cassava production (Table 101A). The average returns to labor do not change much and remain close to $6.5 per day (Table 103A). 227. The policy focusing on rice requires the highest amount of chemical fertilizers. It results in a 17 percent increase, from 465,000 tons to 544,000 tons; i.e., an additional demand of 80,000 tons (Table 102A). Scenario L3 leads to the lowest increase in demand for chemical fertilizers (2.6 percent at 476,000 tons) while scenario L2 results in an additional demand of about 40,000 tons for chemical fertilizer reaching 477,000 tons. These demands were estimated assuming the status-quo application and adoption rates of chemical fertilizers as in 2013. Strong environmental sustainability 228. Under the strong environmental sustainability group, three scenarios were developed. Two scenarios (L4 and L6) maintain the same agricultural land area until 2030 and one scenario (L5) reduces the agricultural land available. 96 229. Scenario L4 keeps the current land use for agriculture but assumes a decrease in cassava yield. Assuming that current farm practices are maintained, characterized by low-to-no use of organic and low use of inorganic fertilizers for cassava, a drop in yield is likely. It is estimated that cassava yield can be as low as 12.6 tons/ha when cultivated on low fertility soil. It is assumed under this scenario that such a low yield will be attained by 2030, corresponding to an annual decrease in cassava yield of 2.4 percent from 2013 to 2030. 230. The RGC’s objective is for 60 percent of land to be forested by 2030, up from its current 56 percent. To meet that goal, scenario L5 takes back land from agriculture through reforestation. This policy target results in about 710,000 ha of non-forest land converted into forest, increasing forest areas to 10.83 million ha from the current 10.12 million ha. Scenario L5 assumes that 25 percent of the required increase comes from cultivated agricultural land in upland areas and the remainder from barren, non-cultivated land. This corresponds to an annual increase of 40,000 ha of forest cover, with about 10,000 ha converted from agriculture. 231. Scenario L6 assumes status quo in the area of agricultural land but simulates the impact of nutrient restitution by putting upland fields into fallow one out of every three years. This practice is expected to restore organic matter and maintain soil fertility at an acceptable level, allowing for the possibility of increased yields. The required yield level is estimated to maintain the current agriculture sector performance with reduced land under cultivation. 232. Soil depletion on upland crops (simulated on cassava production) reduces the annual cassava gross margin by 31 percent (Table 98A). This decline is the direct effect of lower yield, from 18.65 tons/ha in 2012 to 12.80 tons/ha in 2030, corresponding to a production of 4.3 million tons in 2030 compared to the current 6.3 million tons. Overall gross margin for the five crops drops by 5 percent, resulting in an NPV loss of $213 million by 2030 (Table 48). 233. Reducing agricultural land to contribute to the country’s target of 60 percent forest cover in 2030 (scenario L5) also reduces the gross margin. The overall gross margin and NPV decline by 7.5 percent and 4.7 percent in 2030, respectively (Table 48). This decline is the direct effect of lower cultivated land areas for cassava and maize (Table 99A). Because the hypothesis is reduced upland cultivated areas, the share of cassava and maize in total gross margin decreases from 22 percent to 17 percent; the gross margins from rice and vegetables do not change because of the initial assumption. The results show that reducing upland areas harms the agriculture sector in the absence of options to increase agricultural productivity and profitability. 234. In scenario L6, the cultivated upland area is decreased by allocating one-third of the current area to fallow to restore soil fertility. With this assumption, the results show the possibility of maintaining the 2013 level of productivity level until 2030 even with the decrease in cultivated land area. Under scenario L6, overall gross margin remains stable at $1.02 billion despite the lower areas under cassava and maize. The total annual cultivated land declines by 5.2 percent but this amount of land (about 187,000 ha in 2030) becomes fallow and is not reforested (Table 100A). The annual cultivated area under cassava drops by one-third, from 337,800 ha to 223,515 ha, and area under maize drops by the same proportion, from the current 215,442 ha to 142,553 ha. A better production system putting into fallow one-third of the available upland area results in improved soil fertility, optimal pest management and likely higher productivity. 235. However, scenario L6 makes a strong assumption about yield growth. To overcome the impact of lower cultivated areas, farmers need to efficiently manage crop production and adopt agricultural techniques that will result in increased yield. The unchanged gross margin obtained 97 with this scenario requires an increase in cassava yield of 50 percent, from its current 18.65 tons/ha to 25.91 tons/ha, and a similar increase for maize yield from 3.94 tons/ha to 5.90 tons/ha (Table 100A). Table 48: Simulation results for Strong Environmental Sustainability scenarios L4-L6 Simulation Change NPV Returns Change in Change in Likely in Gross ($ to Ag. Demand Demand Effect on Margin, million) Labor for Ag. for Competitiv $ million ($/day) Labor Fertilizers, eness (%) million tons/year (DRC) days (%) (%) Baseline 1,023 10,220 6.41 166.5 464,861 L4 Loss of soil Worsened/ fertility -5.0% 10,007 6.09 0.0% 0.0% DRC increased L5 Decrease of upland crop -7.5% 9,901 6.23 -4.7% -1.2% Neutral area by 2.4% a year L6 Fallow 1/3 of Improved/ upland crop -0.2% 10,211 6.72 -4.7% -1.2% DRC area decreased Source: Own estimates. 9.3. Scenarios for Productivity Growth 236. Considerable technological change has occurred in crop production in the past ten years. A lot of this technological change has the typical features of the green revolution (improved seeds, use of inorganic fertilizers and other chemicals, and irrigation), but there is also an increasing reliance on mechanization, driven by higher labor costs and scarcity of available rural labor. 237. Cambodian farmers are aware of the importance of improved and higher-quality seeds. They think that seeds have a high impact on productivity. Farmer-to-farmer communication is the main source for adoption of new and improved seeds: according to farmers, it was even more important than information obtained from technical agents of the extension system. Farmers are positive about the continuous improvement in seeds. This is not necessarily because of the effort of related institutions such as the research and extension organizations in Cambodia, the public extension system, NGOs or private companies. A key role in improved seed dissemination and adoption is played by contact with farmers and traders in neighboring countries. In practice, many of the improved seeds in Cambodia originated from research and development systems of Vietnam and Thailand. While access to this research and development effort is a positive externality for Cambodia, the seed trade remains largely uncontrolled and issues of intellectual property rights are not addressed. 238. Chemical inputs are increasingly used. But Cambodian farmers use much less chemical inputs than neighboring countries. This lag in adoption of chemical inputs might have some positive effects in terms of reducing environmental impact and chemical residues on food. 98 However, the lower use of chemicals in Cambodia appears to be less the result of a conscious choice inspired by environmental or food safety concerns than the outcome of a poorly functioning input distribution system characterized by high input prices and often inputs of dubious quality. In some cases, most notably cassava and to a lower extent maize, the very low use of fertilizer (whether chemical or organic) has serious negative impacts on soil fertility and sustainability. 239. The increasing use of inputs, however, does not appear to have significantly increased farm profits. In other words, improved technology use has not been accompanied by efficiency improvements or increased TFP. Over the past decade, the overall profitability of agriculture (measured by gross margins per ha in real terms) has not increased for the majority of farmers. The profitability of those engaged in rice might have been stagnant or declining in spite of the noted changes in output prices and use of modern technology. For cassava, the increase in output prices was higher than the increase in the unit cost of production. The problem is that the land expansion in cassava is itself unsustainable, since it is carried out through exploitation of soils that were previously forest land. 240. Labor competition is increasing and labor costs are mounting. Wages for basic agricultural tasks have increased by a factor of three over the past decade, implying changes in production techniques and production cost structure. Farmers are using more services (e.g., mechanization), more modern inputs, and less labor. This movement favors larger farmers who have access to capital and penalizes small farmers for whom labor is a main asset. Migration of youth and women to nonfarm rural sector and urban sector employment continues unabated. The subsequent lack of labor in rural areas puts pressure on rural wages, which has a negative impact on the cost of production; at the same time, it accelerates mechanization. 241. According to the surveys’ findings, rural wages increased tremendously over the past decade, from $1.25/day in 2005 to $4.56/day in 2013. The estimated monthly salary for rural labor is about 70-80 percent of the industrial wage, depending on the season. The most dramatic changes in the production system regard labor use in cassava and dry season rice production, where mechanization has contributed to labor substitution in operations such as land preparation, weed control, harvesting, and threshing. The increasing use of labor in vegetables production might be associated with the need to provide better care, the use of more intensive practices such as several cycles over the year, and the trend toward more labor-intensive vegetables such as broccoli, lettuce, and cucumber. 242. As a result, labor use is decreasing, except in vegetables production (Figure 39), while returns to labor are increasing for all crops, with the exception of maize (Figure 40). Perhaps in the case of maize, early adopters of hybrids were rewarded with higher margins. Figure 39 also shows the higher variability of the returns to labor in 2005, ranging from $1.87/day (wet season rice) to $14.82/day (maize). The gap across crops become less acute in 2013 (between $5.08/day and $10.65/day), indicating improved allocation of available labor in rural areas. 243. It is important to note that the share of service in the cost of production is indicative of a new type of farming emerging in Cambodia, namely one better integrated with the nonfarm service sector. The idea of a self-sufficient farmer is changing. Even though a farmer might be able to produce more than what s/he consumes (at least for some specific commodities), s/he is highly dependent on labor and services provided outside of her/his household. The trend is for the Cambodian farmer to become a farm manager, namely somebody who in the effort to maximize profitability out of land will use inputs, services, and knowledge more than her/his own 99 labor. The implicit concern in all these major changes in the cost and use of labor in agriculture is that neither agriculture nor other sectors of the economy will be able to absorb the increasing labor force. 244. To simulate the effect of factors that could improve TFP, four scenarios are developed. The first scenario (P1) simulates technical change (moving the farm technology frontier up). The second scenario (P2) is about technical efficiency (reducing the productivity gap by shifting farms from traditional to modern practices). The third scenario (P3) combines technical change and technical efficiency. The fourth scenario (P4) simulates the effects of irrigation on TFP. Figure 39: Labor use in production Figure 40: Return to labor ($/day), (days/ha), Cambodia Cambodia Source: 2005 and 2013 surveys. Increased adoption of modern farm technologies 245. Scenarios P1 through P3 assume that increasingly more farmers will become modern technology adopters, resulting in fewer farmers using traditional practices. In these simulations, the level of chemical fertilizer application, the use of improved seeds, and labor allocation in the farm budgets are used as measure for modern technology adoption and contribute both to the higher adoption of new technologies (technical change) and to a larger share of farms using existing improved technologies (technical efficiency), thereby increasing an overall TFP. The basis for the simulation is an annual 4 percent shift from traditional to modern farming until 2030. 246. As the initial distribution of farms by technology use is not known with precision, the information collected during the FGDs was used to make assumptions. Modern technology adopters appear to account for 35-40 percent of total farms in Cambodia. Details of modern input use and the distribution of farms by level of technology use are presented in Table 104A and Table 105A. The projected trend in the number of modern and traditional farms is presented in Figure 41 and shows that by 2030 at least 75 percent of farmers will be categorized as users of modern technologies. This requires at least a 4 percent shift per year from traditional to modern farming in 100 terms of cultivated land. The simulation shows that by 2019, the total area cultivated by modern farmers exceeds that cultivated with traditional techniques; by 2030, there is a clear dominance of farmers using modern technologies in Cambodia. Figure 41: Shift in land used by traditional and modern farmers, Cambodia Source: 2013 survey. 247. Modern farmers differ from traditional ones in the following manner . For rice and cassava, modern farmers use about twice the amount of chemical fertilizers that traditional farmers do (Table 104A). Maize production presents the largest gap (2.6 kg/ha versus 104 kg/ha), and vegetable production a ratio of about 1 to 3.5 (traditional versus modern). The table shows different levels of adoption of improved seeds between modern farmers and those using traditional varieties: the difference is largest for rice and vegetables, with modern farmers using a higher amount of improved seeds per hectare. Table 105 details the level of labor use per hectare across farms using traditional and modern technologies, the wage rate assumed as in 2013. It is assumed that every year, 4 percent of farmers practicing traditional technology shift to modern technology. The breakdown of modern farmers by crop is as follows: 35 percent for cassava, 40 percent for maize and rice, and 40 percent for vegetable production. With the retained growth rate of 4 percent, the overall proportion of farms adopting modern technologies reaches 75 percent in 2030, an increase of 35 percentage points from the current average of 40 percent. 248. Farmers using modern technologies are more productive and profitable . In the baseline, modern cassava producers generate 65 percent more in gross margins compared to traditional farmers; for maize, the gap is higher, at 87 percent. For dry season rice production, the gross margins between modern and traditional farmers are quite close, at $387/ha versus $336/ha. For wet season rice and vegetables, the differences are huge: 124 percent for rice and 240 percent for vegetables. In scenarios P1-P3, it is assumed that modern farmers do not incur additional irrigation costs. 249. Scenario P1 consists of keeping the farm distribution by technology use at 2013 status. However, due to access to better input quality and extension services, modern farmers have higher efficiency, which translates into productivity growth twice as high as that of traditional farmers by 2030 (Table 49). 101 250. Scenario P2 simulates the effects of an annual shift of 4 percent of land cultivated with traditional techniques to land cultivated with modern techniques. However, there is no additional efficiency gain; i.e., the productivity gap between traditional and modern farmers remains the same as the level recorded during the 2013 baseline. 251. Scenario P3 combines the effect of an annual 4 percent shift on land cultivated with improved efficiency of modern technology users by using a productivity gap twice the baseline. This means that modern farmers should be able to get higher gains from the use of any inputs through better crop production systems, improved knowledge and skills, and better information on the quality of inputs. Table 49: Assumptions about Productivity Growth scenarios P1-P3 Baseline P1 P2 P3 % decrease in no. of traditional 0.0% 0.0% 4.0% 4.0% farms % increase in no. of modern farms 0.0% 0.0% 4.0% 4.0% Increase in yield to ensure effectiveness of modern farms Cassava 0.0% 1.1% 0.0% 1.1% Maize 0.0% 1.1% 0.0% 1.1% Wet season rice 0.0% 1.0% 0.0% 1.0% Dry season rice 0.0% 0.4% 0.0% 0.4% Vegetables 0.0% 2.3% 0.0% 2.3% Source: Own estimates. 252. The results simulated with scenario P1 show positive outcomes for the agriculture sector. The gross margins for modern farmers increase by 19.9 percent from $636 million to $764 million in 2030 (Table 107A). This corresponds to an additional NPV of $450 million, 7.3 percent higher compared to the baseline. The total gross margin for the five crops and total NPV increase by 11.1 percent and 4.1 percent, respectively (Table 50). The share of modern farms in total gross margin in 2030 reaches 60 percent. 253. Scenario P2 simulates an annual shift of 4 percent of traditional to modern farming, but without improved productivity. This growth rate leads to three-fourths of farmers adopting modern technologies by 2030. The additional NPV generated by the shift corresponds to $887 million by 2030, 8.6 percent higher than the 2012 NPV (Table 108A). The gross margins for the five crops increase by 25.8 percent (Table 50), including for modern farms by 94.8 percent. This is the direct effect of more traditional farmers adopting modern technologies. Since the assumption is to keep the total cultivated land unchanged, the projected gains in gross margin and NPV exceed losses from potential agricultural land conversion into forest land simulated in the previous section. Moreover, the gain in gross margins for modern farms outweighs the loss in gross margin for traditional farms. 254. Scenario P3 presents more efficient farm production in combination with an annual 4 percent shift in the use of traditional versus modern technologies. The overall gross margins and NPV for the five crops increase by 47.4 percent and 14.9 percent, respectively (Table 50 and Table 109A), with 88 percent of the increased NPV provided by modern farms, which cultivate 77 percent of the total area. 102 Table 50: Simulation results for Productivity Growth scenarios P1-P3 Simulation Change NPV Returns Change in Change in Likely in Gross ($ to Ag. Demand Demand Effect on Margin, million) Labor for Ag. for Competitiv $ million ($/day) Labor Fertilizers, eness (%) million tons/year (DRC) days (%) (%) Baseline 1,023 10,220 6.41 166.5 464,861 P1 More efficient Improved/ modern farms 11.1% 10,640 7.14 -19.9% 26.6% DRC decreased P2 Shift from Improved/ traditional to 25.8% 11,107 10.10 0.0% 0.0% DRC modern farms decreased P3 Combined Improved/ effects of shift 47.4% 11,749 11.84 -19.9% 26.6% DRC and efficiency decreased Source: Own estimates. 255. The results of three scenarios show that the best economic outcome is achieved by a simultaneous increase in the number of farmers adopting modern technologies and more efficient use of production factors. Scenario P3, combining both the technology shift and higher efficiency, leads to the highest gross margins, the largest NPV, and the highest returns to agricultural labor, which is the proxy for farm incomes (Table 50 and Table 110A). The returns to labor are projected to increase to $11.84/day, 86 percent above the 2012 baseline. This high return, however, remains below the long-term Vision’s target as stated in Table 45 ($9/day in 2020 and $18/day in 2030), requiring high attention to additional sources of income growth (e.g., irrigation and diversification). 256. Modern farmers are associated with higher gross margins and higher costs because of the use of more modern inputs. The demand for chemical fertilizers peaks at 598,600 tons in 2030; i.e., an additional 125,000 tons or 26.6 percent higher compared to the baseline (Table 113A). The average application rate for cassava increases by 32 percent, from 14 kg/ha to 19 kg/ha. Chemical fertilizer use on maize increases by 89 percent, from 43 kg/ha to 82 kg/ha. Rice growers apply 158 kg/ha for wet season rice and 317 kg/ha for dry season rice, up by 25 percent and 23 percent, respectively, compared with the baseline. The increase chemical fertilizer use for vegetables is 52 percent, from 303 kg/ha to 460 kg/ha. 257. But modern farms will use less labor, freeing it for other nonfarm uses. With P2 and P3 scenarios, the demand for labor decreases by 19.9 percent in 2030, corresponding to 33 million fewer days (Table 112A) compared to the 2012 baseline. This overall decrease from 166 million to 133 million days, keeping area cultivated the same, comes mainly from reduced demand for rice production. Yet rice producers remain main users of labor, even though their share goes down from 80 percent in 2013 to 72 percent in 2030. Labor demand for other crops increases (7 percent for cassava, 4 percent for maize, and 28 percent for vegetables) but these higher needs are offset by the decreased demand from rice production. Post-harvest activities related to higher productivity generate higher demand for labor for cassava and maize production while for vegetables the increase in labor demand is linked to the overall vegetable production system. 103 Improved irrigation 258. A critical element for increased productivity is the availability and quality of irrigation water. Irrigation increases the likelihood of success when using modern inputs such as improved seeds and fertilizers. It also allows increasing crop intensity through multiple crop cycles per year and provides a system of water control and disaster risk reduction. In this simulation, the assumption is that irrigated land is cultivated twice a year and produces a higher yield (the dry season rice yield observed in the 2013 survey). Foregone revenues and costs are those related to the wet season rice production system, based on data from the 2013 survey. It is assumed that the land used to grow wet season rice receives irrigation and thus reaches the performance of dry season rice productivity. And lastly, all costs are held the same, with the exception of irrigation costs for dry season rice. 259. Scenario P4 focuses on the investment in irrigation. The assumption is that every year about 2 percent of nonirrigated land receives water irrigation infrastructure. If so, land productivity and profitability become similar to dry season rice production. If the rice field is irrigated, the expected yield is 4.34 tons/ha versus 2.78 tons/ha for rainfed rice. From the 2013 farm budgets, the unit price for paddy is $228/ton during the dry season and $262/ton during the wet season. Lastly, the gross margin for dry season rice is $370/ha (excluding irrigation costs) and $245/ha for wet season rice. 260. The NPV from investment in irrigation amounts to $11.98 billion, about 17 percent above the 2012 baseline. This increase comes from about 450,000 hectares of new land with irrigation, with the possibility of two crops per year. Table 51 presents the simulation results from irrigation investment. Table 51: Simulation results for Productivity Growth scenario P4 Simulation Change NPV Returns Change in Change in Likely in Gross ($ to Ag. Demand Demand Effect on Margin, million) Labor for Ag. for Competitiv $ million ($/day) Labor Fertilizers, eness (%) million tons/year (DRC) days (%) (%) Baseline 1,023 10,220 6.41 166.5 464,861 P4 Expansion of Improved/ irrigation 32.4% 11,984 8.52 3.1% 66.7% DRC coverage decreased Source: Own estimates. 261. Irrigation requires a large amount of upfront investment. Table 111A presents some costs observed in South and Eastern Asia that range from $1,400/ha to $4,000/ha. With these levels of prices, the total amount required for scenario P4 range between $636 million and $1.8 billion. The average cost for this scenario is assumed at $2,500/ha, which implies a need for $1.13 billion over the next 10 years to provide irrigation schemes for more than 450,000 ha. With a cost of irrigation of $4,000/ha, the computed internal rate of return (IRR) until 2030 is as low as 7 percent; at $2,500/ha, the IRR is 17 percent. However, if Cambodia manages to get reasonable irrigation investment costs of $1,400, comparable to costs in South Asia, then irrigation is a good candidate for agricultural development, with an IRR reaching 34 percent. 104 9.4. Scenarios for Improved Competitiveness 262. Over the past decade, the overall cost of production for important crops such as rice and maize increased more than the overall revenues from production, implying a decline in profitability per unit of land for farmers engaged in those activities. The concern is then whether the cost of production will continue to increase for key agricultural commodities and/or if profitability (as measured by gross margins per hectare) will decrease. Better linkages between farmers and enterprises and other value chain actors might be a solution to improving farm profitability. Improved linkages could result in higher output prices and lower prices of inputs due to lower transaction and transportation costs from better supply chain management and sales and marketing organization. 263. The DRC analysis showed that agricultural production in Cambodia has a high comparative advantage, with most DRCs less than 1. However, the DRC ratio is closely linked with the level of prices and thus any changes in prices will affect the advantage of the country producing the goods. Price changes for both outputs and inputs are simulated in scenarios C1-C6. 264. The scenarios’ assumptions are summarized in Table 52. Scenarios C1 and C2 hypothesize an increase in output prices by quarter of their past growth rate, and simulate the effect of changes on input prices (scenario C1: identical to the percentage change in output prices; scenario C2: double the change in output prices). In both scenarios C3 and C4, output prices remain stable, but in scenario 3 the wage rate increases by half of its past growth rate, while in scenario C4 the cost of services increases by a quarter of its growth rate. Scenario C5 assesses the impact of lower output prices (15 percent less than current prices in 2030 in nominal terms); in scenario C6, the opposite occurs, such that output prices have a growth rate of 15 percent. Table 52: Assumptions about Competitiveness scenarios C1-C6 Increase/decrease in Baseline C1 C2 C3 C4 C5 C6 Price of outputs 0.0% 25.0% 25.0% 0.0% 0.0% -15.0% 15% Price of seeds 0.0% 25.0% 50.0% 0.0% 0.0% 0.0% 0.0% Price of fertilizers 0.0% 25.0% 50.0% 0.0% 0.0% 0.0% 0.0% Price of other inputs 0.0% 25.0% 50.0% 0.0% 0.0% 0.0% 0.0% Wage rate 0.0% 25.0% 50.0% 50.0% 0.0% 0.0% 0.0% Service costs 0.0% 25.0% 50.0% 0.0% 25.0% 0.0% 0.0% Source: Own estimates. 265. Scenario C1 assumes equal increases in output and input prices. From 2005 to 2013, the annual increases in output prices were spectacular in nominal terms: 17 percent for cassava, 13 percent for vegetables, 7-9 percent for rice, and 4 percent for maize. For a more conservative assumption, the simulation uses one-quarter of the past trends for the calculation. All production costs are also assumed to increase at a quarter of their past growth rate. 266. Scenario C2 focuses on the effect of input prices increasing at twice the rate of output prices. The simulation assumes a growth rate increase of one-quarter of the past rate for outputs, and double the rate for the cost of inputs. Indeed, a comparison of the 2005 and 2013 farm budgets shows a decrease in the gross margin in real terms, partly due to the rapid increase of input costs relative to output prices. This scenario tests the strength of the gross margins with respect to different price trends. 105 267. Scenario C3 simulates the effect of status quo output prices and a 50 percent increase in the wage rate. The analysis simulates the impact of increased wages on key profitability indicators. Cultivated land and labor requirements are expected to remain at the 2013 level to isolate the effect of an increased wage rate. For the simulation, the target of $18/day of returns to labor is used to estimate the annual wage rate growth. An additional assumption is that labor use is inelastic to the wage rate (i.e., it does not change as the wage rate increases). 268. Scenario C4 is similar to C3 but simulates a moderate increase in the cost of services compared to its past growth rate. The cost of services, mainly mechanization, increases by 25 percent, with prices of other inputs assumed constant. 269. Scenario C5 looks at the effect of a reduction of output prices of 15 percent of the past growth rate by 2030. Scenario C6 assesses the impact of an increase in output prices of the same magnitude of 15 percent. Other factors’ costs remain constant. 270. The economic results of the scenarios are summarized in Table 53. A balanced increase of prices of outputs and inputs (scenario C1) results in an NPV gain of $766 million (7.5 percent increase). Price of outputs may increase due to an increase in world market prices or lower marketing costs along with a stronger domestic agroprocessing industry offering higher prices. The nominal yearly gross margin increases by 16.5 percent, from $1.02 billion in 2013 to $1.24 billion in 2030 (Table 116A). The overall growth, however, hides huge discrepancies across crops. Cassava, wet season rice, and vegetables benefit most from these balanced price increases. The NPVs for dry season rice and maize are 9 percent and 27 percent below the 2012 baseline, respectively. 271. Scenario C2 assumes that the prices of inputs (seeds, fertilizers, and other inputs), labor, and services increase at half the rate of their past trends and that output prices increase at one-fourth their past rate. This scenario’s effects harm the profitability of agriculture in Cambodia. The overall NPV is $6.5 billion less than that of the scenario without price changes ($3.76 billion versus $10.2 billion). The NPVs for all crops are lower, with proportional changes ranging from an 18 percent loss for vegetables to a 94 percent loss for maize. 272. If output prices remain stable but the daily wage rate grows at half of its past trend until 2030, Cambodia’s agricultural sector will collapse. The NPV for scenario C3 is only $3.76 billion by 2030, corresponding to a loss of $7.3 billion, the highest among the five competitiveness scenarios (Table 53). This $7.3 billion shortfall is the result of the wage rate increasing at an annual rate of 8.8 percent (half of the past growth rate), from $4.56/day in 2013 to $19.1/day in 2030, comparable to a long-term Vision’s target in Table 45. The largest drop in NPV is for wet season rice (82 percent reduction) due to the relatively low gross margin per hectare ($245/ha) combined with relatively higher labor use (48 days/ha); the smallest reduction is for maize (a 22 percent reduction). 106 Table 53: Simulation results for the Competitiveness scenarios C1-C6 Simulation Change NPV Returns Change in Change in Likely in Gross ($ to Ag. Demand Demand Effect on Margin, million) Labor for Ag. for Competitiv $ million ($/day) Labor Fertilizers, eness (%) million tons/year (DRC) days (%) (%) Baseline 1,023 10,220 6.41 166.5 464,861 C1 Equal increase Improved/ in output and 16.5% 10,986 7.09 0.0% 0.0% DRC input prices decreased (25%) C2 Input prices Worsened/ increase twice -247.7% 3,761 -10.92 0.0% 0.0% DRC that of output increased prices C3 Wages Worsened/ increase by -226.2% 3,655 -7.84 0.0% 0.0% DRC 50% increased C4 Cost of Worsened/ services -25.2% 9,290 4.85 0.0% 0.0% DRC increases by increased 25% C5 Output prices Worsened/ decrease by -66.5% 7,499 2.36 0.0% 0.0% DRC 15% increased C6 Output prices Improved/ increase by 182.8% 16,739 19.48 0.0% 0.0% DRC 15% decreased Source: Own estimates. 273. In scenario C4, a moderate increase in the cost of services (a quarter of the past growth rate, at 1.9 percent per year) affects the overall gross margin less than the wage rate increase does. The NPV is only by $930 million less (9 percent lower) compared to the baseline scenario. The nominal gross margins in 2030 decline by 25.2 percent (Table 53). Vegetables production remains stable with an NPV of $717 million; other crops’ NPV losses range from 7 percent (maize) to 13 percent (dry season rice). The scenario of a moderate increase in the cost of services is more plausible because more competition across small enterprises providing agricultural services may happen in the near future, and would keep the increase in the cost of services at a lower growth rate. 274. Scenario C5 considers the impact of reduced output prices on overall agricultural productivity. By 2030, the gross margins from the five crops decline by $3.5 billion and the annual gross margins are 66 percent lower compared to the baseline, decreasing from $1.02 billion to $357 million. 107 275. An increase in output prices as illustrated in scenario C6 results in higher gross margins and makes agricultural production profitable for farmers in Cambodia. The NPV increases by 64 percent, from $10.2 billion to $16.7 billion. The gains are between 64-80 percent, with the exception of vegetables (26 percent). The nominal gross margin in 2030 is 182 percent higher than the margin in 2012 ($3.02 billion versus 1.02 billion). 276. Ranking the impact of the five scenarios on gross margins with unchanged or reduced output prices, moderate increases in the cost of services result in the least negative impact. This is either compared to the changes due to a higher wage rate (half the past growth rate) or to unequal increases in the cost of inputs versus output prices with a ratio 2 to 1. The changes in gross margins and the level of the demand for labor both affect returns to labor. If the wage rate increases by 8.8 percent a year, which corresponds to a wage rate of $19.1/day in 2030 (scenario C3), and output prices remain the same as in 2013, returns to labor are negative by 2024 (Table 117A). By 2018, agricultural activities under that scenario do not provide incentives to farm given an average return of $4.28/day, below the prevailing 2013 wage rate of $4.56/day. The simulations show very similar results for scenarios C2 and C4 in terms of returns to labor. In these three scenarios, agriculture is not competitive since the simulated returns to labor tend to go below the prevailing wages and even plummet below zero. Farmers would not be willing to stay in agricultural production, and would instead seek other opportunities to earn income. Any increase in the costs of inputs, keeping the price of output constant, is a significant threat for the agriculture sector because in a competitive economy, farmers would choose to leave farming and become wage earners. 277. Scenarios C1 and C6 result in highest returns to labor. In scenario C1, returns to labor increase but only by 7.14 percent, from $6.41/day to $7.09/day. On the other hand, returns to labor increase quite substantially in scenario C6. The average returns to labor reach the expected $19.5/day in 2030. In scenario C6, returns to labor for cassava, maize and dry season rice, ranging between $30/day and $33/day, are highly attractive (Table 117A). Because the wage rate is assumed to not affect the demand for labor, the labor needs remain at 166.5 million days per year. The demand for fertilizer also remains stable, at 474,000 tons because of the zero price-elasticity assumption (Table 118A). 9.5. Scenarios for Income Growth 278. The convergence of farm and nonfarm incomes is the ultimate objective of agricultural transformation. In the past decade, the high growth in agricultural incomes led to a significant reduction in the gap between farm and nonfarm incomes. The ratio of non-agricultural GDP per nonfarm worker to agricultural GDP per farm worker fell from 3.2 in 2002 to 2.1 in 2012. The wages of agricultural workers grew much faster than the wages of unskilled workers employed in the construction and garment industries (recall As a result of increasing returns to labor, agricultural wages in general rose, converging with nonfarm wages. Agricultural wages grew by a factor of almost three between the 2005 and 2013 surveys (Error! Not a valid bookmark self-reference.). Over the same period, nonfarm wages increased by only 60 percent. As a result, the gap between farm and nonfarm sector wages declined. Thus agriculture appears to have provided excellent income-earning opportunities in Cambodia over the last decade. 279. Figure 34), causing the farm wage gap to fall from 0.35 in 2005 to 0.75 in 2013. 108 280. Growth in future farm incomes depends on many factors, but it is imperative for agriculture to grow faster than the rest of the economy on a per capita basis to continue the farm income convergence. The productivity growth analyzed in Chapter 9.3 is a powerful driver of future growth. A declining agricultural labor force will also help increase per capita income. Additional sources of income growth, simulated in this chapter, are: (i) a change in the variety of cultivated crops (for example, to fragrant rice); (ii) the addition of processing to increase the value added of crop production (for example, cassava); and (iii) a change in land distribution by farm size, which affects farmers’ per-capita income. 281. Hardest to support is the income convergence of the smallest farms. During the past decade, small farms became smaller while large farmers grew larger. Farms with less than 1 ha producing annual crops, except vegetables, have smaller incomes on average. They are less likely to adopt modern technologies and integrate into modern value chains. Special attention is needed for this group of farm households, as they will continue to account for the largest share of farm households in Cambodia. Fragrant rice production 282. The DRC analysis showed that producing fragrant rice instead of the IRRI rice would give Cambodia a higher comparative advantage relative to neighboring countries. One consequence of switching to fragrant rice production is the reduction in production volumes (fragrant rice has a lower yield than IRRI rice), but this is compensated by the higher price for fragrant paddy. 283. Two scenarios assess the impact of farmers switching varieties of rice from high- yielding IRRI to fragrant. Scenario I1 assumes a 4.2 percent annual shift from IRRI to fragrant rice, combined with a 75 percent price difference in the output (difference based on the 2013 survey). Scenario I2 looks at a further annual shift (9.9 percent), keeping the same price difference between fragrant rice and high-yielding IRRI rice varieties (Table 54). The growth rate of areas cultivated with fragrant rice is based on a hypothetical 10 percent and 25 percent of all rice area in Cambodia adopting fragrant rice by 2030. To reach these targets, the annual growth rates must be 4.2 percent and 9.9 percent, respectively, with initial land area producing fragrant rice estimated at 5 percent. Table 54: Assumptions about rice variety changes under Income Growth scenarios I1-I2 Scenario Baseline I1 I2 Land area under fragrant rice, annual change 0.0% 4.2% 9.9% Land area for other crops, annual change 0.0% -4.2% -9.9% Source: Own estimate. 284. Unser scenario I1, the areas cultivated with fragrant rice doubles by 2030. Gross margins for rice show an annual 6.2 percent increase, from $762 million in 2013 to $808 million in 2030 (Table 121A). These gains are the effect of a higher price for fragrant rice that compensates for the lower yields. It is assumed that the wet season rice yield gap is 16 percent (2.89 tons/ha for other varieties versus 2.5 tons/ha for fragrant rice), and the dry season rice gap is 37 percent (4.80 tons/ha versus 3.50 tons/ha for fragrant rice). Labor and fertilizer demand are held constant. The NPV gains 2.1 percent, from $7.29 billion to $7.44 billion (Table 55). The returns to agricultural labor from producing rice increase by 6.2 percent. 109 Table 55: Simulation results for Income Growth scenarios I1-I2 Simulation Change NPV Returns Change in Change in Likely in Gross ($ to Ag. Demand Demand Effect on Margin, $ million) Labor for Ag. for Competitiv million ($/day) Labor Fertilizers, eness (%) million tons/year (DRC) days (%) (%) Baseline 762.04 7,293 5.68 134.08 430,134 I1 Fragrant rice area Improved/ increases to 6.2% 7,443 6.03 0.0% 0.0% DRC 10% of total decreased rice area I2 Fragrant rice area Improved/ increases to 24.2% 7,744 7.06 0.0% 0.0% DRC 25% of total decreased rice area Source: Own estimate. 285. If the 2030 target is expanded to 25 percent of total rice land cultivated with fragrant rice, the gains on gross margins rise by 24.2 percent (scenario I2). This corresponds to an increase of the NPV to $481 million over the next 17 years. Returns to labor increase by 24 percent, reaching $7.06/day in 2030 (Table 55). Processing cassava into dry chips 286. The second group of scenarios assesses the effects of domestic value addition on farm incomes. Data on cassava from the 2013 survey are used to compare the gains from selling fresh cassava to those from selling dry chips processed from cassava. The loss in product weight is about 50 percent after drying but the gain in price could reach 50 percent or higher (Table 126). The assumed annual growth rates are a hypothetical 10 percent (scenario I3) and 25 percent (scenario I4) of total cassava area used for production of cassava for processing and selling as dried chips by 2030. Scenario I3 looks at the likely changes due to an annual 4.2 percent increase in the area processing cassava before sales (Table 56), corresponding to the hypothetical 10 percent of total area with processing by 2030. Scenario I2 assesses the changes resulting from a further shift from fresh to processed, with annual growth rate of 9.9 percent, aiming to have 25 percent of land selling processed cassava by 2030. Table 56: Assumption about cassava processing under Income Growth scenarios I3-I4 Scenarios Baseline I3 I4 Land area for cassava processing, annual change 0.0% 4.2% 9.9% Other cassava land area, annual change 0.0% -4.2% -9.9% Source: Own estimate. 287. Processing cassava into dry chips incurs additional costs, but the cost increase is offset by the gains from better output prices. Under scenario I3, the 2030 gross margin shows a slight increase of 2.2 percent, from $155.5 million to $158.9 million (Table 125). The changes are quite small because of the relatively lower starting point (5 percent of production processed). Scenario 110 I4 further simulates the shift to be around 9.9 percent a year, aiming at 25 percent production processed in 2030. With that hypothesis, cassava gross margins increase by 8.6 percent, from $155.5 million in 2013 to $168.9 million in 2030. As a consequence, the returns to labor increase by 8 percent, from $10.61/day to $11.49/day in nominal terms (Table 127A). As per the initial assumption, the demand for fertilizers is not affected by processing but the demand for labor is slightly altered because of the additional manual tasks required for cassava drying. For scenario I4, total demand for labor increases from 14.66 million to 15.01 million days per year by 2030. Farm size distribution 288. Agricultural land markets are already very active in Cambodia, and are likely to become more so in the future as the result of improved property rights and agricultural transformation. The latter might imply a number of situations whereby small and marginal farmers sell their land to larger farmers, commercial farmers consolidate their landholdings in contiguous plots, and new lands are opened up as a result of improved connectivity. The transformation of land size distribution is likely to be in the direction of larger farm size. 289. The 2013 survey indicates that farm size distribution is becoming more unequal. Large farms are becoming larger; small farms are becoming smaller; and the number of landless or marginally landless rural households is increasing. Sophal (2008) reported that about 21 percent of rural households are involuntarily landless, while a further 45 percent are land-poor owning no more than 1 ha per household. This process is fueled by an active land market that farmers use to either sell or buy land. Generally, young married couples are most affected. 290. In the current analysis, “large” is a relative term. For rice and maize production, large refers to farms with cultivated rice land of more than 3 ha. For cassava, the benchmark is 5 ha; and for vegetables, large farms have more than 1 ha of land. Therefore, even though the trend is towards larger farms, the agricultural farmland distribution over the next decade will continue to be dominated by farmers who have farms less than 5 ha. 291. The farm budget analysis in Chapter 4 showed that in general there are no economies of scale in Cambodian crop agriculture. The situation is very complex and varies from crop to crop and with different farm sizes and technologies. For some crops and some types of crop cultivation (e.g., vegetables production and modern wet season rice) there were indications of decreasing economies to scale; for maize, there was an indication of increasing economies of scale; and for cassava and dry season rice, there was no indication of economies of scale (either increasing or decreasing). 292. Given these findings, the current trend towards increasing inequality in farm size distribution might have either positive or negative implications in terms of overall profitability of the sector. The outcome is an empirical one that is not preordained a priori. Small farmers may be negatively affected by the trends toward agricultural transformation, becoming landless or near-landless. The initial distribution of farm size is not known with precision, so information collected during FGDS is used to estimate the proportion of farm by size. Though very rough, these data allow simulating the impact of the move toward larger farm sizes. On average, small farms account for 47-75 percent of total farms; medium-size farms for about 21-41 percent; and large farms for less than 10 percent, with the exception of maize (12 percent). Details of the farm size distribution used in the analysis are presented in Table 128A. 111 293. Two scenarios simulate the effect of a change in land distribution by farm size on gross margins and returns to labor. If the land holding per household increases, overall income per capita shall also increase. The initial data about unit margin by crop and by farm size allow derivation of the distribution of overall margins (reported in Table 128A). To compare the simulated effect of farm size distribution on key indicators, overall cultivated land is held constant at its 2013 level of 3.58 million ha for the five crops. 294. Scenario I5 assumes an annual decrease of 1 percent of land under small farms, holding the total cultivated land and average landholding by farm constant . This scenario assumes that half of the land area from small farms goes to medium-size and half goes to large farms (Table 57). It implies a decreasing number of small farms and inversely, an increasing number of medium-size and large farms, and an overall lower number of agricultural farms but with higher average farm size. 295. Scenario I6 assumes the same farm distribution as in scenario I5, but more efficient large and medium-size farms such that their gross margins are higher than in the 2012 baseline. For the simulation, the increase in efficiency is set at an annual 5 percent. This efficiency may, for instance, come from better farm management. Table 57: Assumptions about Income Growth under scenarios I5-I6 Scenarios Baseline I5 I6 % decrease in small farms 0.0% 1.0% 1.0% % increase in medium farms 0.0% 0.5% 0.5% % increase in large farms 0.0% 0.5% 0.5% Effectiveness of large farms 0.0% 0.0% 5.0% Source: Own estimate. 296. Using the parameters for scenario I5 over the 2030 horizon, total land cultivated by small farms decreases by 15.7 percent, from 2.10 million ha to 1.78 million ha. Land for medium-size farms increases by 14.6 percent, from 1.14 million ha to 1.31 million ha; and large farms cultivate 52 percent more land, from 319,000 ha to 486,000 ha. The resulting farm distributions are the same for both scenarios. 297. Changes in farm size distribution do not lead to any increase in the profitability per capita and per hectare farm production. Focusing only on land distribution, scenario I5 results in a tiny increase of 0.1 percent of the NPV, from $9,799 to $9,811 million (Table 132A). However, there are changes within each crop related to the area cultivated by farm size. For example, increases in gross margins for medium-size farms range from 9 percent for maize to 28 percent for vegetables under scenario I5. For large farms, the increases are more significant, from 30.8 percent (maize) to 147 percent (vegetables). These increases are proportional to per hectare gross margins and offset the decreases in gross margins for small farms (an average 16 percent reduction). Because the data on farm budgets do not show clear economies of scales, the impact on the overall gross margins is very marginal. Under scenario I5, simulating only shifts from small to medium-size and large farms, the gross margins from small farms still dominate the industry, capturing 49 percent of the overall gross margins, albeit down from the 58 percent 2013 baseline. 298. A more detailed analysis of the results of the changes in farm size distribution is presented in Table 130A. It shows that small farms cannot rely solely on agricultural revenue for 112 their livelihood. By 2030, the average total landholding for small farms is around 1 ha, which provides $0.14 gross margin per capita per day, assuming an average household size of 4.5. This per capita gross margin is 16 percent lower than the 2013 baseline. Even for medium-size farms, the daily agricultural gross margins are around $0.48, making these households vulnerable, yet they get 16 percent higher gross margin per capita gains. Only large farms with an average landholding of 15 ha benefit from agricultural production, with a daily per capita gross margin of $2.31, 51 percent higher than the 2013 baseline. This simulation demonstrates the challenge of small farms relying only on agriculture for their livelihoods. 299. To ensure that the agriculture sector economy benefits from increasing farm size, larger farms have to become more efficient, the main hypothesis in scenario I6. Higher efficiency may be a result of better farm management, lower prices of inputs through consolidation of purchases, or higher prices of outputs through better coordination with traders. To keep costs constant, there is no assumption on increased use of inputs, and the farm size distribution remains similar to scenario I5. With a 5 percent increase in gross margin for large and medium-size farms, gross margins for medium-size farm increase by 162 percent, from a total $342 million to $899 million (Table 130A), and gross margins for large farms by 250 percent, from $88 million to $308 million (Table 131A). On the other hand, the share of gross margins for small farms drops to 29 percent. Medium-size farms provide 53 percent and large farms 18 percent of the overall gross margins. 300. Demand for labor and fertilizers are barely affected by the changes in farm size distribution. Labor demand slightly decreases by 4 percent because large and medium-size farms use less labor for their agricultural activities. Similarly, demand for inorganic fertilizers stay practically, decreasing slightly from 438,000 tons to 429,000 tons per year. Overall, there is however a gain in efficiency because with lower labor and fertilizers demand, and the same amount of land, scenario I6 results in slightly higher gross margins. 9.5 Summary of Simulations’ Results 301. Table 58 presents the summary of the main results of all scenarios. A rise in agricultural productivity has the largest positive effect on farm incomes, especially if the shift from traditional to modern technologies is accompanied by higher efficiency of modern input use. Farmers producing fragrant rice, processing cassava into dry chips, and undertaking other value addition activities can further increase their incomes. Lowering production costs through better use of existing resources (e.g., productivity increase) and minimizing drops in farm output prices through lower logistics costs are the keys to maintaining farm competitiveness. The continuation of land expansion also provides additional income but sustainability considerations limit large expansions in the future. This constraint makes agricultural productivity, commercialization, and diversification even more critical for ensuring future agricultural growth that reduces poverty and boosts shared prosperity in Cambodia. 113 Table 58: Summary of the simulations’ results Simulations Effect on NPV ($ Effect on Effect on Demand Effect on Demand for Likely Effect on DRC Gross million) Returns to for Labor Fertilizers (tons/year) (Competitiveness) Margin (%) Labor ($/day) (mill. days) Sustainability Baseline 1,023 10,220 6.41 166.51 464,861 L1 Rice area increased by 25% 11.2% 10,646 6.42 11.1% 17.1% Neutral L2 Equal increase of cultivated land by 10% 18.2% 10,882 6.70 13.1% 8.2% Neutral L3 Upland area expansion by 14% 21.6% 10,958 6.86 13.7% 2.6% Neutral L4 Loss of soil fertility -5.0% 10,007 6.09 0.0% 0.0% Higher: Disadvantage L5 Decrease of upland areas by 2.4% a year -7.5% 9,901 6.23 -4.7% -1.2% Neutral L6 Put into fallow one third of upland areas -0.2% 10,211 6.72 -4.7% -1.2% Lower: Increased advantage Productivity Baseline 1,023 10,220 6.41 166.51 464,861 P1 More efficient modern farms 11.1% 10,640 7.14 -19.9% 26.6% Lower: Increased advantage P2 Shift from traditional to modern farms 25.8% 11,107 10.10 0.0% 0.0% Unknown: depends on costs P3 Combination of P1 and P2 scenarios 47.4% 11,749 11.84 -19.9% 26.6% Lower: Increased advantage P4 Irrigation 32.4% 11,984 8.52 3.1% 66.7% Lower: Increased advantage Competitiveness Baseline 1,023 10,220 6.41 166.51 464,861 C1 Equal increase of all prices by 25% 16.5% 10,986 7.09 0.0% 0.0% Lower: Increased advantage C2 Input prices increase twice output prices -247.7% 3,761 -10.92 0.0% 0.0% Higher: Disadvantage C3 50% increase of wages -226.2% 3,665 -7.84 0.0% 0.0% Higher: Disadvantage C4 25% increase of costs of machine services -25.2% 9,290 4.85 0.0% 0.0% Higher: Disadvantage C5 15% decrease of output prices -66.5% 7,449 2.36 0.0% 0.0% Higher: Disadvantage C6 15% increase of output prices 182.8% 16,379 19.48 0.0% 0.0% Lower: Increased advantage Income Growth Baseline 7,293 5.68 134.08 430,130 I1 Fragrant rice 10% of rice area in 2030 6.2% 7,443 6.03 0.0% 0.0% Lower: Increased advantage I2 Fragrant rice 25% of rice area in 2030 24.2% 7,744 7.06 0.0% 0.0% Lower: Increased advantage Baseline 1,488 10.54 14.66 4,577 I3 Cassava chips 10% of total in 2030 2.2% 1,499 10.78 0.6% 0.0% Lower: Increased advantage I4 Cassava chips 25% of total in 2030 8.6% 1,523 11.45 2.4% 0.0% Lower: Increased advantage Baseline 9,799 6.20 166.00 464,861 I5 Small farms shift land to medium and large 0.3% 9,811 6.40 -4.0% 0.0% Unknown: depends on costs I6 Farm distr. shift + large farms more efficient 6.8% 11,793 10.70 -4.0% -2.1% Lower: Increased advantage 114 10. POLICY AGENDA TO SUPPORT A LONG-TERM VISION FOR CAMBODIAN AGRICULTURE 302. Agriculture can and should continue to make a large contribution to economic growth and poverty reduction in Cambodia. Realization of a Long-Term Vision for Cambodian Agriculture can build on the transformative changes of the past decade and the strength of Cambodian farmers to respond quickly should market opportunities arise. However, it should be noted that despite the significant reduction in poverty in the past, vulnerability of the poor to shocks is still high, and any slowdown in agricultural and non-agricultural growth could push many people back into poverty. Further reforms are required to support future growth and to get closer to the final outcome of structural transformation where agriculture as an economic activity has no distinguishing characteristics from other sectors (in terms of the productivity of labor and capital, or the location of poverty). The main challenge is to move from extensive agricultural growth to intensive one. The past agricultural growth in Cambodia was largely achieved by farmland expansion and higher use of inputs in light of rising agricultural prices. With the land frontier closing and inputs getting more expensive relative to outputs, future growth depends on other driving forces and policy actions. 303. The medium-term policy agenda to support realization of a Long-Term Vision for Cambodian Agriculture needs to include “old” policies that helped drive past growth and “new” policies that were previously either ineffectively implemented or less relevant. Based on the analysis carried out in the previous chapters, the priority areas include: (i) maintaining a private sector friendly agricultural policy environment; (ii) strengthening the environmental sustainability of agricultural production; (iii) increasing the quality of and budgets for agricultural public programs, without undermining the fiscal sustainability; and (iv) helping develop the agribusiness and agroprocessing industry. These priority areas are important in the next five-year horizon period, and each broad priority area includes three to four critical actions presented below. 10.1. Maintaining a Private Sector Friendly Agricultural Policy Environment 304. Cambodia’s market-oriented agricultural and trade policy helped achieve the high past rates of agricultural growth. The private sector has benefited from minimal, if any, interventions in farm output and input pricing, from the strong commitment to open trade, including across the border, and from the reduction of export costs and time for export processing. This has created conditions for the private sector to invest and generate profits. These open trade and market-oriented agricultural policies remain critical to future growth, and need to be protected by all means. Some proposals being discussed in the country may undermine the past achievements, however, such as: a. Redirecting paddy and other raw commodities from exports to domestic processing or storage through the use of export restrictions would be a mistake. Restricting paddy exports in the face of high processing and storage costs in Cambodia would result in lower farm prices and thereby lower farm incomes. Farmers need to remain free in their production and trade choices. b. Promotion of agricultural diversification requires better rural infrastructure, investments in agricultural innovation systems (agricultural research, extension, education, training), and programs related to integrated pest management, food safety, 115 wholesale markets, and input market development. On the other hand, supply-driven promotion of diversification including through land zoning by allocating farmland to specific crops such as vegetables or subsidies to specific crops would not lead to sustainable, market-driven diversification. c. Slower agricultural growth and worsening agricultural TOT may lead to the calls to introduce direct farm support measures. This happens in countries where farm incomes do not converge quickly enough with nonfarm incomes (China and Thailand, for example), and this may well happen in Cambodia. Introducing direct farm protection measures can slow down agricultural growth and hamper structural transformation as is the case in other countries with high farm protection (see Chapter 2.4). Cambodia can use agriculture for poverty reduction for many years to come, if it avoids the farm subsidy trap seen in other countries. 305. In regard to the new policies, attention needs to be paid to improving efficiency in farm input industries, especially seeds. The removal of burdensome regulations and the state monopoly in the seed sector would provide further incentives for intensification, commercialization, and diversification, all needed to maintain high agricultural growth in the future. Access to high-quality seeds is essential to achieve the growth targets as discussed in Chapter 9. Without quality seeds, farmers’ investments in fertilizers and machinery bring lower rates of return. It is estimated that the current supply of improved rice seed varieties meets only 22 percent of farmers’ demand (IFC 2014). The main reason for this undersupply of rice seeds is the monopoly of the Cambodia Agricultural Research and Development Institute (CARDI) in production of the foundation seeds required for further multiplication to certified seeds. CARDI also produces and sells certified seeds, competing with seed companies, which is unusual for a research institution. The trial and release of imported varieties are not clearly defined and import of seeds for trials, production, and commercialization has proved to be difficult and expensive. Overall, the legislative and institutional environment for development of the seed sector is absent. 306. The seed industry requires urgent reform. The reform agenda is challenging but the costs of doing nothing are very high. Table 59 summarizes the key issues and recommendations for the seed sector reform developed by the International Finance Corporation, the World Bank Group (IFC 2014). 307. Regulatory improvements are also required for other inputs such as fertilizers to reduce costs to farmers. Regulatory burden is mainly related to simplification of importation procedures. If they are simplified and streamlined, farmers will be able to access cheaper fertilizers. Potential measures to improve regulations include the approval of import licenses on the basis of product suitability only and allowing all importers to import any quantity of registered fertilizers, without the need to receive license/approvals for each new import lot. 116 Table 59: Key issues and options to reform the seed sector in Cambodia Issues Options/Recommendations Lack of a framework for long-term  Finalize and approve the National Seed Policy development of the seed sector, which  Begin work on a separate Prakas to implement Chapter reduces confidence of investors 2 of the 2008 Seed Law which relates to Plant Breeders Rights Lack of clear procedures on variety  Approve the Prakas to implement the seed management testing/registration and seed quality part of the 2008 Seed Law and prepare two manuals for control technical procedures  Define and simplify the procedures for variety testing and registration, taking into account different needs of different crop groups, esp. rice, maize, and vegetables Lack of transparency in seed sector  Prepare and update regularly the National List of regulations varieties that recognizes new varieties within the country so they can be subject to proper regulations  Publish the scale of fees for all government services, maximum periods for processing requests, and clear mandatory requirements Inadequate supply of seeds  For rice varieties bred in Cambodia, increase the supply of foundation seed and explore alternative schemes to complement government resources  For varieties bred elsewhere, a company registered in Cambodia could be designated by the breeder as the maintainer and be responsible for foundation seed production, provided they can demonstrate the necessary competence  Facilitate imports of seed by streamlining and setting transparent and clear procedures, especially seed intended for trial purposes  Coordinate and strengthen the many small local seed producers by improving the quality of their product and facilitating market access, for example by linking them to rice mills Poor systems to ensure quality of seeds  Establish a single office to handle all official matters relating to seeds and varieties; this is consistent with the proposal now under consideration by the General Directorate of Agriculture of MAF to establish a ‘Seed Secretariat’  Establish the key elements of official quality control including a certification scheme and a Seed Testing Laboratory, based within the National Agricultural Laboratory of General Directorate of Agriculture Source: IFC 2014. 10.2. Strengthening the Environmental Sustainability 308. Sustainability would need to be more strongly embedded in agricultural policy. Past agricultural growth in upland areas was largely driven by rapid land expansion. Especially for 117 cassava, the FGDs and expert interviews revealed that some cultivated land expansion comes at the cost of forest land. Moreover, as soil nutrients are mined and not replenished due to low use of fertilizers for cassava production, this land expansion resulted in some land degradation. In other areas of the country, cultivated land expansion puts pressure on water resources and might also encroach on pasture land. This consideration, in line with any biodiversity conservation program that Cambodia may have, would argue against continuing a development strategy based on land expansion. And to reach the 60 percent target for forest land, such an expansion is not a feasible option in Cambodia. 309. The main policy implication is then the formulation of land use planning and its enforcement at the local level to prevent loss of forest land. This needs to be accompanied by other measures such as a program to raise awareness about loss of soil fertility and incentives to use good practices in soil nutrient management. Extension workers can work with farmers and the cassava processing industry to ensure sustainable supplies of raw material. The program could lead to the emergence of a domestic cassava processing industry rather than the current situation whereby Cambodian farmers sell raw material primarily to Thai cassava plants, and by doing so, they mine Cambodian soil nutrients. 310. In addition to protecting forest land, strengthening land tenure security would make agricultural growth in Cambodia more sustainable. Farm intensification and diversification require investments; more will be invested when land is secure. Yet 30 percent of farmers in Cambodia still operate on land without legal titles, according to ADB (2014a). This reduces investments in agriculture and creates risks of land degradation. 311. Promotion of sustainable land management practices is also needed. The slower expansion in farmland considered in Chapters 8 and 9 would reduce some farmers’ income but is necessary to avoid further soil degradation and loss of valuable forest resources. Short-term income losses of farmers can be compensated by longer-term benefits from higher soil fertility and more efficient management of available resources. Promotion of putting land on fallow to regenerate soil fertility (recall scenario L6 from Chapter 9), the use of crop mixes such as alternating pulses with maize to enhance nitrogen rates, and other sustainable land management practices can further enhance environmental sustainability (FAO 2014). The international experience shows that the most effective way to promote such practices is to combine the enforcement of natural resource protection with agricultural extension and other public programs that increase land users’ awareness of the longer-term costs and benefits and promote adoption of modern technologies. 312. With more farmers using chemicals for vegetable and other crop production, the government is recommended to strengthen public health regulations related to their safe use and promote alternative programs that ensure safe, effective and environmentally sound pest management. These alternative programs include integrated pest management, where pest population is controlled through biological controls, cultural practices and the development and use of crop varieties that are resistant or tolerant to the pest. 10.3. Improving the Agricultural “Public Goods” Investments 313. The quality of future agricultural growth will depend on the RGC’s success in spending more and better on agriculture with the ultimate objective of helping farmers raise their incomes. The higher spending on agriculture, therefore, would need to be considered within 118 the feasibility of total government budget so not to undermine fiscal sustainability (Box 9). The rising land and labor constraints call for closing yield gaps (intensification), more effective use of modern technologies (commercialization), better use of labor-saving technologies (mechanization), and faster response to the evolving market opportunities (diversification). The analysis in the previous chapters shows that these processes have already begun and that farms using modern technologies are more capable of overcoming these constraints than traditional farms. Converting more traditional farms into commercial ones offers a very powerful source of agricultural productivity growth in the upcoming decade in Cambodia. This in turn requires more and better public spending on productive infrastructure and human capital. Box 9: Importance of Fiscal Sustainability for Pro-Poor Agricultural Growth Sustainable macroeconomic environment is a necessary pre-condition for translating public expenditures into pro-poor agricultural growth. Public expenditure policy is a form of direct economic intervention. Like other interventions, public spending on agriculture should be part of a market-friendly approach to economic policy, and supportive of development and adjustment goals. Excessive agricultural public spending can lead to high or rising budget deficits that can result in different types of macroeconomic imbalances (e.g., higher inflation and misaligned exchange rate), causing lower economic growth and weaker demand for farm products. By investigating the underlying factors in agricultural growth Gardner (2005) found macroeconomic stability and real income growth in the non-agriculture economy among the most important factor explaining agricultural growth in 85 developing countries during 1960- 2001. That’s why public spending on agriculture should remain consistent with the aggregate fiscal discipline. 314. It is important to have a view of productivity growth broader than a narrow emphasis on yield growth. Yield growth could be accomplished without an increase in profitability; in fact, this is exactly what seems to have happened over the past decade for most crops in Cambodia. Farmers use more inputs but fail to translate this into much higher profits. The emphasis on profitability forces researchers, extension agents, farmers, and the private sector to look at the combination of inputs and outputs, prices, and technologies that make a certain production activity more profitable over time. If profitability could be increased, given the same natural resources and same amount of labor, then a real gain in productivity is achieved. 315. Future public programs for agriculture need to take into account the impacts of climate change. Cambodia is significantly exposed to weather shocks, mainly shorter rainy seasons and longer and drier dry seasons (World Bank and AusAid 2011; Thomas et al. 2013). High dependence on rainfall for agriculture makes Cambodia, and particularly the poorest farmers in Cambodia, vulnerable. Farmers already report the negatives impacts of more frequent drought spells, and to lesser degree, floods (see Chapter 3.9, Figure 27). Climate change leads to more variable growing seasons and water deficits, which require investments in water storage, development of new crop varieties and farming techniques that are more resilient to unpredictable growing seasons, and better weather forecasting. In other words, a move to the climate-smart 119 agriculture is required.48 As programs related to water management, agricultural research, and advisory services offer a triple win (by simultaneously helping Cambodian farmers raise productivity, mitigate the impacts of climate change, and support adaptation to climate change), they would need to be prioritized. 316. All these have implications for the selection and design of agricultural programs (e.g., agricultural public expenditures). Agricultural public spending does not seem to have played much of a role in past agricultural growth. Public investments in connectivity (i.e., roads, ports, and ICT), and private investments in rice mills, played the larger role as discussed in Chapter 6. One reason for the low impact in the past is the underfunding of agricultural programs. During 2007-2009, the total budget of three agriculture-related ministries, e.g., MAFF (agricultural innovation systems), MOWRAM (irrigation), and Ministry of Rural Development (rural roads), was only 1.4 percent of GDP (World Bank and AusAid 2011). In 2014, the agricultural budget has hardly changed in relation to GDP, except for higher spending on irrigation. Another reason is the low quality of agricultural programs. Going forward, both the level and quality of funding need to be addressed. 317. Investments in irrigation are critical to support future agricultural growth. But spending more on irrigation alone does not guarantee successful outcomes in Cambodia. Past irrigation investments have so far failed to improve farmers’ access to water in the dry season, due to the excessive focus on primary irrigation structures, small operation and maintenance (O&M) budgets, disempowered farmer water user communities (FWUCs), and the lack of river basin management institutions. These problems were described in the World Bank and AusAid report in 2011, and according to the ADB (2014b), they still prevail in 2014. In spite of the relatively large irrigation investments, the actual irrigated area in the country is only about 350,000 ha, or 8 percent of arable land, the lowest irrigation coverage in Asia (Table 5). 318. During 2014-2018, MOWRAM is projected to invest $250-300 million annually (Figure 42). This will be a significant increase compared to past years: in 2009, MOWRAM’s budget was about $60 million. Only 12 percent of the project future budget is national funding, with the remainder provided by development partners (Figure 43). China provides about half, Korea about a quarter, and the remainder from a number of other partners. This sizable investment means that substantial areas of new and rehabilitated irrigation areas will come on line in the coming years, highlighting the need to address these problems: a. To increase the effectiveness of irrigation investments, public funds need to be extended to secondary canals and associated infrastructure. Rehabilitation of tertiary canals using public funds is also justified, provided that it includes arrangements for O&M costs to be recovered from end users. In a 2011 survey carried out for the World Bank and AusAid, 85 percent of farmers reported the lack of water, mainly due to the lack of distribution canals. In the 2013 survey conducted for this report, most farms using irrigation reported using pumps rather than gravity irrigation. Farmers spend 11-20 percent of total variable costs on pump irrigation for dry season rice. Future irrigation investments, therefore, need to focus on improving farmers’ 48 Climate-smart agriculture seeks to increase productivity in an environmentally and socially sustainable way, strengthen farmers’ resilience to climate change, and reduce agriculture’s contribution to climate change by reducing greenhouse gas emissions and increasing carbon storage in farmland. 120 access to gravity irrigation to reduce farm production costs and maintain their competitiveness rather than investing in new dams and primary irrigation structures. b. FWUCs need to be empowered. They still play a very nominal role as they are not authorized to collect irrigation service fees and have little say in system management decisions. The FWCU subdecree that would assign appropriate responsibilities and rights for irrigation infrastructure to respective FWUCs and allow them to collect irrigation service fees remains unapproved. The CDTA/ADB provides technical assistance to MOWRAM to find flexible solutions for fee collection, and the subdecree needs to be approved soon to underpin the sustainability of large future irrigation investments (ADB 2014b). c. Capacity for O&M of irrigation schemes needs to be strengthened. In addition to a lack of financial resources49, fundamental problems exist with organizational systems, planning processes, assessment of O&M needs and priority, and capacity of staff. A 2014 study on the economics of O&M in irrigation schemes demonstrated that the lack of O&M results not only in suboptimal return on investment but also negative returns (CDTA 2014). Until recently, MOWRAM allocated almost nothing for O&M, and the Provincial Departments of Water Resources and Meteorology have limited capacity to maintain their assets, too. Figure 42: MOWRAM’s projected Figure 43: Structure of MOWRAM’s investments, $ million, 2014-2018 projected investments by source Source: ADB 2014b. 319. Investments in irrigation need to be accompanied by improved water use efficiency. This will allow achievement of one of the sustainability targets from Table 45. However, water efficiency may be low and irrigation investments may not be profitable if yields are not boosted or the production costs are high. Using the data from the 2013 survey it is estimated that at a unit 49 In 2009, only $1 million, or 14 percent of total recurrent budget of MOWRAM, was allocated to O&M versus the international best practice of allocating 33 percent of total investments on O&M (equivalent to $17 million in Cambodia). 121 cost of investment of $2,500/ha, an increase of 1.5 percent per year of dry season rice yield would lead to a 4.2 percentage point increase in the irrigation IRR. 320. Agricultural innovation systems that include agricultural research, extension, education, and training have an important role to play to raise yields and more importantly productivity. Research needs to produce new technologies and knowledge relevant to farmers. Extension is tasked with bringing these technologies and knowledge to farmers and helping them with adoption and improvement of their management skills. Education creates knowledge, while training helps youth and adults improve their vocational skills. Working together for these four elements is the key to bringing innovation to agriculture and accelerating long-term agricultural productivity. 321. Cambodian agricultural research has underperformed in the past decade. The main reasons include overall underfinancing, limited operational budgets, and project-based financing. Problems to resolve to increase public agricultural research’s contribution to future agricultural growth include: a. Too little is spent on agricultural research in Cambodia. In 2009, agricultural public research spending was only 0.1 percent of agricultural GDP, or $0.10 per $100 of agricultural GDP (Table 60). This is a very small amount compared to the countries with competitive agriculture sectors. Thailand, for example, spends 0.5 percent of agricultural GDP on public agricultural research (Nippon 2013). Other agricultural exporters spend even more, with more successful ones spending 1-2 percent of agricultural GDP. b. Especially little is allocated to operational expenses. Researchers are constrained in going to the field to talk to farmers and extension officers. The demand responsiveness of agricultural research programs is therefore very low. This also results in the weak link to agricultural extension services. c. Allocation of the majority of the agricultural research budget is largely short-term project-based. There are few large long-term research programs, which makes it difficult to ensure synergies between various small research projects as researchers are not able to work together under the same program to attain overarching strategic research objectives which would lead to knowledge accumulation and innovation. It also limits the ability to hire better quality research staff. Funds are provided mainly by donors. d. Project-based research funding should be complemented by core funding. The international experience shows that project-based funding approaches, often based on contestable funding mechanisms, while showing good results in some large research systems with strong research capacity, could have severe limitations as they tend to undermine long-term strategic planning. The longer-term nature of agricultural research (i.e., development of new crop varieties or a system to manage animal disease) raises the question of the ability of project-based research funding system to respond to the needs of agriculture as opposed to allocating resources through longer-term core funding commitments relative to the priority needs of the agriculture sector. 122 Table 60: Agricultural research intensity (public research spending in $ per $100 in agricultural GDP) Location 2000 2008 Developed countries 2.40 3.07 Developing countries 0.53 0.54 Asia 0.41 0.42 Cambodia n/a 0.10 Source: IFPRI 2012 and World Bank and AusAid 2011 for Cambodian data. 322. Public agricultural extension is very important in countries with many small farms, such as Cambodia. Agricultural extension encompasses several dimensions: the economic dimension focuses on increasing farm income, agricultural productivity, farm financial management, and food preservation and nutrition; the social dimension includes improvement in the health of family members, and leadership. Despite the recent increase in coverage (recall Figure 23.1), many farmers do not receive extension services of high quality. They value more highly services received from the public sector and NGOs than those from private companies (recall Figure 23.5). 323. Numerous problems weaken public agricultural extension in Cambodia. These include low funding, lack of trained manpower, poor communication skills of extension workers, unavailability of appropriate technologies for smallholder farmers, little attention to market linkages, and poor collaboration between research and extension. These problems are well recognized in the country, which led to the preparation of the Agricultural Extension Policy, to be completed early 2015. A number of strategic policy decisions are needed to operationalize this policy. Among many, these include sustainable financing, farmers’ outreach, service delivery models, and the evolving role of public extension: a. Similar to agricultural research, public agricultural extension is underfinanced . In 2009, the budget for agricultural extension was $6.7 million, 0.2 percent of agricultural GDP. Most funds come from donors. While continued funds from donors can be expected, more resources should also come from the RGC given the high importance of agricultural extension to future agricultural growth. Commercialization of some services may bring additional revenue, but it is probably unrealistic to expect large income flows in the near term. The 2013 survey showed farmers’ low willingness to pay for extension services. b. Many farmers are not reached out by extension services. At present, there are 1,103 public agricultural extension officers. An estimated additional 696 commune extension workers, 8,794 village extension workers, and 14,607 village animal health workers work through donor-funded projects, NGOs, and the private sector (MAFF 2014a). Considering public officers only, one officer has to reach out to 1,700 farm households and cover 4,100 ha of agricultural land to serve all farm households.50 This is a very low outreach rate. But with all extension workers included, except animal health workers, the outreach rate falls to 177 farm households and 425 ha, a reasonably good ratio. To maintain this outreach rate and even increase it, will MAFF seek to increase 50 According to the 2013 Agricultural Census, the total number of farm households in Cambodia is 1,875,712. Total agricultural area consisting of arable land and land under permanent crops is 4.5 million ha. 123 the number of public officers to replace the private agents? Or will it try to utilize the existing network of extension agents pursuing public private partnerships? The latter may be a preferred option from the perspectives of demand responsiveness and fiscal sustainability. c. How to deliver extension services? Many different extension service providers operate in Cambodia, and the pluralism is well recognized in the upcoming Agricultural Extension Policy. The key question is how to further promote it in a coordinated manner. The international experience suggests that matching grants to the private sector on a competitive basis or vouchers given to farmers to procure services of private extension providers are effective ways to deliver services to farmers. Recipients of such grants must be able to demonstrate improved service delivery based on expressed demand and the specific local needs of farmers for training and technology transfer. The public sector would play an important role in setting up the funding mechanisms, criteria to access funds, monitoring, and quality controls. ICT needs to be used to monitor delivery of services by the private sector and increase extension outreach at low costs. d. The focus of extension services would need to shift. It is necessary for extension to move away from supply-driven advice to support of technology adoption, market orientation, competitiveness, and sustainability. Extension services have a large role to play in achieving the targets presented in Chapter 8 and the successful outcomes presented in Chapter 9. It can do more to link farmers, especially smaller-scale ones, into value chains, through promotion of input purchase and marketing farm groups, contract farming, and productive partnerships. The trend in Cambodia is for small farms to become smaller, and these farms often face the biggest challenges integrating into food value chains. Extension can also do more to promote export of agricultural products. In collaboration with other departments, it can help increase buyers’ trust in the quality and safety of Cambodian products through better surveillance of pests, monitoring the use of agricultural inputs in production areas, and promoting good agricultural practices. Along with food safety actions at other segments of the value chains, these public programs are the key to promoting agricultural exports, including to China, the largest market in the world. 324. Investing in human development though education and vocational training would also help raise agricultural productivity. Farming is a knowledge-intensive industry with changing requirements, not least because of climate change. The recent movement towards more mechanized agriculture, for example, requires many workers with skills to operate tractors and other farm machines. This in turn would require more spending on vocational training for youth and continuous education programs for adult farmers. Yet public investment in vocational training and agricultural education remains very low. This needs to change to allow farmers to translate the use of machinery into higher profits, not just higher production volumes as has happened so far. 325. The public sector has other roles to play and services to deliver to raise agricultural productivity. The full list of such services cannot be made in this report. But the critical point for public programs is to identify and address the underlying roots of the problems, without crowding out the private sector. Examples for fertilizers and agricultural finance follow. 124 326. It is a fact that losses in yields from climate change can be compensated by increases availability of nitrogen in the soil. The research on the impact of climate change on Cambodian agriculture conducted by Thomas et al. (2013) shows that applying more fertilizers can help maintain crop yields (Table 61). Table 61: Yield response for supplementing nitrogen to soil, Cambodia Nitrogen Used for % Change in Yield for Each Additional High-Fertilizer 10 kg N/ha Scenario (kg N/ha) 2000 2050 Rainfed wet season rice 90 7.2 5.5 Irrigated dry season rice 90 6.7 5.5 Rainfed maize 90 4.3 3.4 Irrigated maize 90 4.8 3.7 Rainfed taro 90 3.7 3.1 Irrigated taro 90 5.4 4.5 Source: Thomas et al. 2013. 327. In recent years fertilizers use in Cambodia has increased. The FGDs in the 2013 survey showed a 30 percent increase in the number of urea users for rice production. According to ADB (2014a), 86 percent of rice farmers in Battambang, Kampong Thom, and Takeo use inorganic fertilizers. In contrast, fertilizer use for cassava and maize production remains limited. As a result, average fertilizer nutrient consumption remains low, at 15 kg/ha in 2011, according to FAOSTAT. Fertilizers are increasingly available on local markets, but their low consumption is explained by high prices, low quality, and poor knowledge of soil-specific requirements. The public sector has an important role to play in correcting these market failures: a. Fertilizer prices in Cambodia can be reduced . For example, import licenses can be approved on the basis of product suitability only, and all importers be allowed to import any quantity of registered fertilizer products. b. No reliable soil maps in Cambodia exist to allow the development of soil- specific fertilizer recommendations. Extension workers give blunt recommendations. As a result, many farmers apply too much, too little, or even the wrong fertilizer to trigger a profitable supply response. Production of such soil maps and fertilizer recommendations are core functions of public research and extension. c. The existence of fake fertilizers reduces their use by farmers. While the true extent of fraudulence is probably lower than perceived, this perception is strong enough to affect fertilizer adoption. MAFF estimated that in 2008 about 30 percent of fertilizers were fraudulent or of low quality. The rapid assessment in 2010 found a much smaller share of fraudulent fertilizers (IFDC 2010). To reduce this negative perception, it is essential that the Department of Agricultural Legislation in Cambodia and the field inspectors have access to reliable and fast fertilizer analytical services as visual inspections cannot detect product adulteration or below-guarantee analysis. In the field, inspectors can use a simple pill cutter to cut DAP and APS granules in half to check their consistency. It was recommended to MAFF to ask for independent quality assurance certificates of analysis from accredited organizations for all imported fertilizers, particularly those with multi-nutrient content, either NP or NPK. 125 d. Applying nitrogen fertilizers is not the only way to get more nitrogen into the soil for plant use. Another approach is to improve soil fertility management by using animal manure, cover crops, crop rotation, and crop or agroforestry residue. This is another important role for agricultural research and extension to play in helping farmers gain knowledge about these technologies and adopt them. 328. Cambodia does not need subsidy programs to promote fertilizer use. There are many examples of huge failures of such programs in Asia, ranging from Indonesia and India to Sri Lanka, and around the world. The example from Sri Lanka is presented in Box 10. Fertilizers are available on the market in Cambodia in all locations, even for the smallest farmers. The public role is to increase their sustainable use through extension, quality and safety controls, and investments in infrastructure, not through subsidies. 329. Another government role is to improve farmers’ access to affordable and suitable agrifinance products. As described in Chapter 3.7, financial services from several sources comprising commercial banks, MFIs, community saving groups, and money lenders are increasingly available to farmers in Cambodia. Access to finance can be further improved. 330. Subsidizing interest rates or providing concessional credits to agriculture are not recommended, due to their low sustainability and high economic distortions, including crowding out rural MFIs. There are innovative ways to improve access to finance in a sustainable manner at low cost. Cambodia’s financial sector needs institutions, markets, and products tailored to farmers and agribusinesses’ needs. One example of such products is agricultural loans tailored to farming practices, seasonality, and cash flow projections. Box 10: Fertilizer Subsidy in Sri Lanka Sri Lanka has subsidized fertilizer costs for many years, spending a large part of its agricultural budget on these subsidies. Its total agricultural budget share is large, accounting for 12 percent of agricultural GDP over 2008-2012. Yet these high subsidies have failed to induce strong growth. Since 2000, average growth in agricultural GDP has been 3.2 percent. A recent study analyzing the effectiveness of the fertilizer subsidy found that although fertilizer use positively and significantly explains paddy productivity in Sri Lanka, the marginal returns to fertilizer progressively declines with higher application rates (Gautam and Kar 2014). A majority of farms in irrigated areas, while behaving rationally in terms of profitability, tend to use fertilizer way above the level recommended by the Department of Agriculture. The lack of proper education in efficient fertilizer practices (like balanced use of nutrients) and existing inefficient practices result in lower private and social economic returns from fertilizer use. The study also found that despite the higher intensity of fertilizer use in Sri Lanka than in most South Asian countries, yields are modest and have not grown significantly since 2000. Improvements in the composition of public spending for agriculture could have greater impact on the sector’s productivity by targeting fertilizer subsidy benefits and investing in other production inputs. Investments on agricultural public goods such as research and extension services would also result in more efficient fertilizer use. 126 331. Amret, with the support of the World Bank/AgriFin project, helped develop and introduce such a loan product in Cambodia. Agricultural loans are provided for both agricultural activities (agricultural loans) and agriculture combined with other businesses (hybrid loans). In contrast to Amret’s existing credit products, agricultural loans’ repayment terms are flexible. They are aligned with individual clients’ seasonality of income and cash flow projections. The interest rates for agricultural loans are on par with Amret’s existing individual loan products and there are no penalties for early repayment. 332. Agricultural loans are reportedly popular with farmers. They also turned out to be more profitable for Amret compared to traditional loans in rural areas, which do not take into account farm business characteristics. An agricultural credit officer reaches break-even costs in month five with these loans (compared with month eight for traditional loans), and on average all costs of agricultural loans’ set up and provision are covered in eight months. As a result, the share of agricultural loans in Amret’s total loan portfolio increased from 0.6 percent in December 2013 to 2.8 percent in June 2014. 333. The RGC can increase awareness about such products and promote their wider adoption by other banks and MFIs. This would be a much more cost-efficient and sustainable way to improve access to agricultural finance in Cambodia. 10.4. Helping Develop the Agroprocessing and Agribusiness Industry 334. The growth of the agribusiness and agroprocessing industry is critical to future agricultural growth, job creation, farm income convergence, and structural transformation in Cambodia. Yet the country seriously lags behind in this respect, as presented in Chapter 2.2, in spite of the growing raw material base. The main trigger for a stronger industry is an improvement in the business environment, addressing the following main constraints (World Bank 2014e): a. The high cost of electricity (Cambodia has the second highest price per kilowatt in the Southeast Asia after Singapore). b. Informal payments and burdensome regulations. Cambodia’s ranking in the overall ease of doing business in the Doing Business 2014 report has fallen over the past ten years; it is now 137 out of 189 economies. It also ranks 162/189 for contract enforcement because of an unreliable conflict resolution system. c. Weak capacity and inefficiencies in transport systems resulting in high transport costs, long transport durations, and unreliable delivery. d. The undeveloped financial system, high informality, and predominantly fixed asset collateral-based lending limit private sector lending from banks and increase financing costs. 335. There are no substitutes for reforms and investments to improve a business climate that would facilitate private sector investments. However, particular policy actions that would have the strongest impact on agribusiness development in Cambodia in the upcoming decade include improvements related to: (i) food safety and sanitary and phytosanitary (SPS) measures; (ii) rice trade logistics; and (iii) access to finance. 336. Difficulty complying with food safety and SPS measures of importing countries is among the major constraints for agribusinesses in Cambodia. While there are positive changes 127 in rice industry in this respect,51 the majority of rice mills and other firms do not have good facilities and personnel to meet demanding market segment requirements and to supply high-end markets. The country does not have a good diagnostic capability for food safety/SPS testing, and the SPS implementation capacity in terms of human skills, technical capacity and operational funding is weak (van der Meer and Samrith 2014). Transaction costs for the private sector are high and awareness of the high impact of transaction costs on trade among SPS agencies is very low. The situation in Cambodia contrasts with the situation in Thailand, for example, where food safety standards are not a major constraint for agri-food exports (Box 11 and van der Meer 2014). Box 11: Food Safety Investments in Thailand Thailand was among the first group of emerging countries to invest in food safety infrastructure and has developed the capability to export safe food since the mid-1990s. For example, when Thai exports of shrimp and chicken faced stricter SPS measures in the EU in 1996 and the early 2000s, the Thai private sector quickly responded (Nidhiprabha and Chamchan 2005). It introduced bio-safety farms that use fewer chemicals and lobbied the government to invest in food-safety infrastructure and to legislate new food safety laws. Some large exporters have established brand names for hygienic products of high food safety standards. Consequently these actions make Thai exporters more competitive vis-a-vis their competitors in the region, though many small- scale exporters still have problems complying with the strict standards, and foods sold in domestic markets are not as safe as exported foods. 337. Even for rice, Cambodia needs new markets outside the EU.52 To fine them the RGC would need to: (i) assist rice mills with complying with food safety standards of different countries; (ii) foster technical upgrades; (iii) promote rice specifications and a strong brand identity for Cambodian rice; and (iv) facilitate trade with more information, building for example on the IFC’s Rice Sector Support Project.53 China is a particularly important market as the largest importer of rice in the last few years, but it is very strict with respect to food safety requirements for imports. The RGC can do many things to help its exporters sell more rice and other products to China (Box 12). 51 The Cambodian private sector has made significant progress in this area, including with the support from the IFC Rice Sector Support Project, which helped seven rice mills certify for the Hazard Analysis and Critical Control Points during 2014. More mills are in pipeline for such a food safety certification. This certification is critical in terms of accessing export markets such as the EU and the United States. 52 The major export market for Cambodian rice is the EU under the EBA. Cambodia enjoys the zero import tariff, which gives it a $240/ton cost advantage compared to other exporters. An EU safeguard may be triggered when imports under EBA exceed by 25 percent the volume imported the previous year. In 2014, Cambodia was close to this 25 percent ceiling, calling for identification of new outlets for expanded exports. 53 The IFC project was launched in 2012 to support the development of the Cambodian rice industry, aiming to increase farmers' access to improved inbred rice seed and increase their yields, increase milling efficiency and product quality, develop and implement a strategy to increase Cambodian rice exports, and reduce the price differential between Thai and Cambodian rice. 128 Box 12: Requirements for an Increase in Cambodian Rice Exports to China A 2014 rice marketing study (Agland Investment Services 2014) identified key constraints for Cambodia to access the growing Chinese rice import market. A selection of the recommended actions includes:  Focus on improved packaging and labelling on consumer packs.  Establish contacts at the government-to-government level with Chinese trade authorities who control the level of imports and quotas; understand the quota system54 and ensure that exporters comply with it.  Support the role of China Certification and Inspection Co. Ltd in Cambodia for inspections.  Improve storage and logistics to lower marketing costs. Cool storage may improve quality especially for the preferred new crop rice.  Establish the “White Gold” emblem as a quality design factor of genuine Cambodian rice.  Ensure compliance with buyer requirements. 338. Improving rice export logistics is another priority area to develop agribusiness. Currently, Cambodian rice export’s competitiveness is significantly affected by high logistics costs and unreliable timelines. Average logistics costs from rice mills to the Sihanoukville port in Cambodia are twice as high as similar costs in Vietnam and Thailand (Table 62). It also costs more in Cambodia to move paddy from farm gate to rice mills. Moreover, Cambodia’s logistics performance indicators on timelines have worsened since 2007, unlike in neighboring countries, damaging Cambodia’s image as a reliable partner for buyers (World Bank 2014d). This reduces the profitability not only of farmers but also of rice mills and traders, thereby hindering investment in the agroprocessing and post-harvest handling. 339. An important reason for the high logistics costs in Cambodia is cumbersome and expensive export procedures. Cambodia’s export procedures have improved recently both in terms of processing time and associated costs, but still largely fall behind its neighbors. The SOWS-REF was set up in the compound of the Council for Development of Cambodia. While all services related to rice export have yet to be included in the SOWS-REF,55 rice exporters note that the time and cost to obtain all necessary approvals have reduced remarkably. On average in 2013, export procedures cost $14/ton and took four days to process; i.e., the costs declined by 30 percent compared to 2012 and the processing time fell by one-half (Table 63). Yet these procedures are reported to be highly bureaucratic and loaded with informal costs. Despite the reduced export procedure cost per ton, it is still high compared to Thailand ($0.10/ton), Vietnam ($0.05/ton), and even Myanmar ($8.50/ton) (World Bank 2014f). Further reduction is therefore needed. 54 The tariff rate quota for import of rice to China is divided into 2.26 million tons each for long grain rice and medium grain rice. While the tariff rate quota for long grade rice was largely exhausted in each of 2012 and 2013, very little - if any - of the medium grain rice quota was utilized. This may provide an opening up for Cambodia (see Slayton and Muniroth 2012b). 55 The Certificate of Origin and Phyto-Sanitary Certificate are in the SOWS-REF but not the Quality Certificate (CamControl) and Custom Certificate. 129 Table 62: Prices and logistics costs of rice in Cambodia, Vietnam, and Thailand Farm-gate Costs from farm Rice mill Costs from rice FOB price to rice mill price mill to port price Cambodia 247 163 410 50 460 Vietnam 253 122 375 23 398 Thailand 339 126 465 25 490 Note: Prices of white rice per ton in 2013. Source: World Bank 2014d. 340. In addition to export procedures, container-based exports increase costs. In Cambodia, most aromatic and non-aromatic rice is shipped in containers. But in the rest of the world, most rice is transported as break bulk cargo. Break bulk is cheaper and easier to handle, but it has hardly been tried in Cambodia. The Sihanoukville and Phnom Penh Port drafts limit break bulk to ocean destinations only, not Asian destinations. Possibilities exist to export rice in bulk for intra-Asian trade from Sihanoukville Port if key constraints are tackled (World Bank 2014d), including: (i) lack of a suitable hygienic rice warehouse; (ii) lack of professional stevedoring; (iii) a combination of high port and National Shipping Agency and Brokerage charges; and (iv) the missing link between railway station and port quay. Improving logistics requires actions across several ministries, agencies, and the private sector. Table 63: Time and cost in rice export procedures, Cambodia, 2012-2013 Services Processing Costs Processing Time 2012 2013 2012 2013 SPS Certificate $150/case $35/case 2 days 1 day Fumigation Certificate $20/container $35/container 1 day 1 day Certificate of Origin $250/case $141/case 1 day 1 day Custom Certificate $25/container $6.50/container 2 days ½ day CamControl Certificate $25/container $52.50/ container 2 days ½ day GMO Certificate $150/sample $80/sample 3-4 days 2-3 days Average total* $20/ton $14/ton 8 days 4 days Note: *The average services fee per ton is lower for the larger export size. Source: World Bank team’s estimates based on information from key rice exporters and SOWS-REF. 341. Improving access to finance is necessary to support development of a strong agroprocessing industry in Cambodia. There are opportunities to introduce a warehouse receipt system (WRS), starting with rice. Currently, there is too little working capital in the rice milling sector (similar constraints have been reported in other agricultural processing sectors, such as cassava). Most mills use their land, buildings, and equipment (i.e., long-term fixed assets) to secure loans that they use for short-term working capital needs. These loans tend to fall far short of their actual working capital needs, and furthermore, using long-term assets for working capital needs reduces millers’ ability to upgrade their storage and production capacity (and thus to invest in the needed expansion of rice export capacity). 342. In Cambodia, paddy or rice are not yet used as collateral for working capital loans . Yet many of the conditions for such loans are present (Rutten 2014). Rice is a seasonal crop that has to be stored over a prolonged period so that it can be processed throughout the year – thus, 130 during many months large physical stocks exist in the country. The rice produced in Cambodia is far from uniform, but rice millers and traders categorize it into three varieties (fragrant, mixed and white), with the actual paddy and rice traded being priced at a discount or premium to a reference price for that variety – so for bankers who wish to finance against rice inventory, there is a reference price against which to determine its value. Rice production is fairly concentrated. Four of the country’s 24 provinces produce almost half of the total, permitting financiers to get access, at least in principle, to large paddy stocks without having to cover too large a geographical region. The potential market is attractive in size as one million tons of rice may require as much as $300 million in working capital finance, and rice for the domestic market also needs to be financed. 343. Rice mills are potential clients for banks. Rice mills maintain large inventories and have an incentive to continue doing so, not just to ensure that they have enough raw materials to keep their mills operating the longest time possible, but also to meet client demands for specific kinds of white rice. Expanding financing for mills would lead to improved financing for the whole rice sector: rice millers act as financiers both upstream (pre-financing traders and at times, even providing inputs on credit to farmers) and downstream (selling on credit to wholesalers and exporters, permitting the latter to sell on deferred payment terms to international buyers). 344. Even in the short run, the potential benefits of providing more working capital finance to the rice sector are large enough to promise profits to those who take the initiative to build its necessary institutional infrastructure. While many of the required activities are most logically undertaken by the private sector, support from the RGC, including in conceptualizing strategies and coordinating efforts, will help in realizing the possibilities outlined before. The RGC has an important role to play in creating a conducive environment for a WRS (Rutten 2014). The first important step is to remove discrepancies in the Secured Transaction Law and the Civil Code on the access to movable assets in case of default. Other steps include improving the collateral registry, strengthening lenders’ capacity and awareness, and providing insurance guarantees for a WRS. Pro-active support can be given to pilots before scaling up nation-wide. Avoiding credit subsidization and supply-driven investments in warehouse construction is also an important step for the RGC. Support for a WRS will create a financing system that can measure up to the government’s aspirations for its rice sector, and will lay the foundation for more sustainable, balanced growth. 131 11. CONCLUSIONS AND POLICY RECOMMENDATIONS 345. This report seeks to understand the successes, challenges and opportunities of Cambodia’s agricultural transformation over the past decade to derive lessons and insights on how to maintain future agricultural growth, and particularly on the government’s role in facilitating it. In 2013-2014, the agricultural growth slowed down to 1 percent from its average of 5.3 percent over 2004-2012. Is the country in transition to a slower agricultural growth? Cambodia can ill afford it because agricultural growth will be critical to continued poverty reduction in the country, given its large size in the economy. Market and private investment friendly policies and targeted public sector investments in irrigation, extension, and other “public good” agricultural services, as feasible within the government’s total budget, can help secure continued robust agricultural growth. Key Findings 346. In the last decade, the agricultural sector in Cambodia has gone through significant structural transformation. Although still playing a large role, the agriculture sector became relatively less “important” in the economy in terms of its share in GDP and labor force, but more “productive” in terms of land and labor productivity. This transformation was driven by high and pro-poor agricultural growth. 347. The agricultural growth in Cambodia was high. During 2004-2012, the annual growth in agricultural gross production was 8.7 percent. The agricultural value added grew by 5.3 during this period. This exceptional growth, among the highest in the world, was driven by crop production, mainly paddy rice (annual growth of 9 percent) but also spectacular growth in other crops, particularly maize (20 percent), cassava (51 percent), sugarcane (22 percent), and vegetables (10 percent). The growth in livestock and fisheries was modest at around? 348. The agricultural growth was also pro-poor. Cambodia’s poverty headcount declined from 50 percent in 2007 to 21 percent in 2011, with the number of poor declined from 7 to 3 million. Most poverty reduction took place in rural areas. More than 60 percent of the poverty reduction was attributed to the agriculture sector: higher rice prices stimulated the larger rice production that increased farm wages. Yet, in 2011, 91 percent of the poor still lived in rural areas. Thus, further poverty reduction will continue to depend on the success of agriculture for many years to come, due to its large role in labor force, value added, and exports. 349. Due to the high agricultural growth, agricultural wages have been converging with nonagricultural wages. Agricultural wages grew by 2.6 times between the surveys conducted in 2005 and 2013, while the nonfarm wages increased by 60 percent. As a result, the ratio between per worker nonagricultural to agricultural valued added in current prices fell to 2.1 in 2012 from 3.2 in 2004. Returns to farm labor have been increasing. 350. The crop diversification has started. Although crop production is still mainly focused on paddy, the crop mix has been changing driven by higher profitability of non-rice crops. In 2013, average farm gross margins (and returns to labor) were $506/ha ($9.4/day) for cassava, $303/ha ($8.8/day) for maize, and $1,393/ha ($7.2/day) for vegetable production, compared to $245/ha ($4.6/day) for wet season rice and $296/ha ($9.6/day) for dry season rice. Dry season rice competes with non-rice crops in terms of returns to labor but its expansion is constrained by limited irrigation. The share of total area planted under paddy declined from 86 percent in 2002 to 74 132 percent in 2011, while the share of planted area for maize and cassava production increased significantly, by 1.7 and 8 times, respectively. 351. Even the paddy sector has started to diversify. Triggered by the demand from modernized rice mills, more farmers grow more profitable aromatic paddy, estimated at 10 percent of rice cultivated area and 30 percent of total production. Thought aromatic yields tent to be lower? Further expansion of farmland under aromatic paddy is possible with improvements in quality seed supply, agricultural extension, and irrigation. 352. Yields increased for most crops. With the annual growth in cultivated land areas at 4.7 percent and agricultural gross production at 8.7 percent, the average growth of yields was 4 percent during 2004-2012. This growth in yields was triggered by a wide adoption of improved technologies, expanded (yet still limited) irrigation, more use of modern inputs, and better access to mechanized services, pointing to the advances in commercialization. Farmers also have better access to markets. 353. Cambodian farm products remain price competitive at farm gate. A domestic resource cost analysis of competitiveness shows that despite rising labor costs and prices of farm inputs, the value added generated by farmers exceeds the costs of domestic factors of production (land, labor, and capital). This high competitiveness explains the large increase in agricultural exports in the recent decade. However, the competitiveness of ordinary rice produced during the wet season has worsened in recent years due to the increase in labor costs and weak knowledge/skills of most farmers in using modern inputs. Many small farmers start losing a competitive edge. 354. The past agricultural growth was driven by several factors. Among the major was open trade and, in general, market-oriented agricultural policy. Cambodia was one of the few developing countries that did not overact to the 2008 global food price spike but actually saw higher food prices as an opportunity to induce agricultural growth. Other net-exporting countries such as India, Lao PDR, and Vietnam used export restrictions to limit transmission of the global food price spike into their markets. Higher agricultural prices in Cambodia made farmland expansion profitable. In addition, the agriculture sector benefited from: (i) improved access to overseas markets through the Everything but Arms Agreement with the European Union and open cross-border trade with neighbors; (ii) better availability and wider use of mechanization services triggered by the higher cost of rural labor; (iii) better farm access to finance; and (iv) private investments in rice mills. 355. But there have also been challenges. The large share of the past agricultural growth was driven by farmland expansion. The average contribution (weighted by crop areas) of land expansion to the change in farm gross margins in real terms between 2005 and 2013 was about 60 percent. Farmland expanded annually by 4.7 percent, with very large increases (128 percent) for cassava. This farmland expansion has contributed to the accelerated deforestation, especially in the upland areas. 356. The average increase in per hectare gross margin was good, at 3.4 percent per year. But it varied between 2.1 percent for dry season rice to 44.5 percent for vegetables. The per hectare maize’s margin even declined. In many cases, farmers expanding their land areas received higher incomes, but farmers with land areas unchanged or reduced were not able to substantially increase their incomes. The period of relatively high food prices was largely used to expand land areas rather than build a strong foundation to productivity increases. 133 357. While poverty was reduced significantly, the number of vulnerable people has increased. Most people who escaped poverty did so by a small margin. The loss of only 1,200 Riels per day (the cost of 2 small water bottles) would cause Cambodia’s poverty rate to double to 40 percent. This high rate of vulnerability is the sign of still modest productivity increases in the agricultural sector. 358. Vulnerability is largest among smallest farms. Small farmers with the land area less than a hectare, who still account for 48 percent of all farmers, reported to find it difficult to expand and get integrated into the emerging modern food supply chains. The larger farms are becoming larger and smaller farms are becoming smaller in Cambodia. This change in distribution of farmland occurs independently of economic land concessions. The average size of the interviewed farms with less 1 ha declined from 0.99 ha in 2008 to 0.88 ha in 2012, while the average size of medium farms (between 1 and 3 ha) increased from 1.55 ha to 2.38 ha and the larger farms (above 3 ha) from 3.61 ha to 7.03 ha. Productivity of small, traditional farms also hardly improved, as agricultural extension and other public services have not reached them at a large scale. It appears that the income increase for this group of farmers in the past decade largely came from higher production values, driven by high agricultural prices, and the sale of their labor to larger farms at higher wages. 359. Except for rice, the agroprocessing industry has played a limited role in agricultural growth. Almost all crops were exported to neighboring countries unprocessed. This indicates serious weakness in the value chain, particularly in the post-harvest system of supply chain management (collection of raw material, storage, finance, logistics, transportation, and information). 360. Addressing weaknesses of the past growth is a good strategy for maintaining a future agricultural growth at 5 percent. With the global food prices declining and land frontier closing down, the Cambodian agriculture has been losing its two major growth drivers. On top of that, agricultural labor continues to decline and become more expensive adding to production costs. Global agricultural prices are projected to continue their decline in the next decade, and selling low-quality ordinary rice on domestic and foreign markets will be less and less profitable. Relying on higher domestic demand driven by the increase in GDP and higher import demand alone, without more efficient use of resources, will not be enough to maintain the past growth. 361. The recent slowdown in agricultural growth in Cambodia is worrisome. As global food prices gradually declined, Thailand returned to the rice market after several years of quasi isolation due to its distortive rice pledging scheme and the re-entry of Myanmar into global rice market, the rice (especially dry season) production in Cambodia has stalled. And with it, total crop value added declined, from 4.9 percent in 2012 to 0.6 percent in 2013. The growth in agricultural value added slowed down to 1.6 percent in 2013 after the 5.3 percent average growth between 2004 and 2012. 362. Returning to a 5 percent growth is important. Agriculture maintains a large share in GDP, trade, and labor force, which means that it stays the key to further reduce poverty and vulnerability in the upcoming decade. If Cambodia’s structural transformation continues, with agricultural growth averaging 5 percent, by 2030 Cambodian agriculture will be less significant in the economy but more productive. Its share in GDP (and total labor force) is projected to go down to 17 percent (31 percent) in 2030, from 26 percent (51 percent) in 2012, while land productivity would increase from $1,300 to $2,700 and labor productivity rise from $1,200 to $3,700. 134 363. By contrast, the cost of slow agricultural growth would be huge. If average agricultural growth is only 3 percent between 2012 and 2030, the end agricultural value added would be 29 percent lower by 2030 compared with the 5 percent growth scenario, slicing overall GDP by 18 percent, keeping more people in agriculture, reducing agricultural labor productivity by 34 percent and stopping any income convergence between farmers and nonfarmers. A lower agricultural growth will also lead to a slower reduction in poverty. Policy Recommendations 364. A long-term vision for Cambodian agriculture includes elements of sustainability, productivity, competitiveness, and income growth, beyond a focus on production. Simulation of different scenarios for future agricultural development reveals that a rise in farm productivity would have the largest positive effect on farm incomes, especially if the shift from traditional to modern technologies is accompanied by higher efficiency of modern input use and irrigation. Farmers producing fragrant rice, processing cassava into dry chips, and undertaking other value addition activities can further increase their incomes, illustrating the importance of the agroprocessing industry. Lowering production costs through better use of existing resources (e.g., productivity increase) and minimizing drops in farm output prices through lower logistics costs are the keys to maintaining farm competitiveness. Continued land expansion also provides additional income but sustainability considerations will limit large expansions in the future. This constraint makes agricultural productivity, commercialization, and diversification even more critical for ensuring future agricultural growth that reduces poverty and boosts shared prosperity in Cambodia. 365. Continued rapid agricultural growth and further structural transformation in the agriculture sector are possible in Cambodia. Four sets of policies together will help support such continued agricultural growth during the next five years (short-to medium run). The first is maintaining a private sector friendly agricultural policy environment, with added attention to lower the regulatory burden in farm input sectors. The second is strengthening the environmental sustainability of agricultural production. The third is improving the quality of agricultural public programs and as feasible within total government budget, increasing allocations to more effective programs. And the fourth is helping develop the agribusiness and agroprocessing industry. Table 64 presents the key policy recommendations of the report by implementing agency. Maintaining a Private Sector Friendly Agricultural Policy Environment 366. Cambodia’s market-oriented agricultural and open trade policy helped achieve the high past rates of agricultural growth. The private sector has benefited from minimal, if any, interventions in farm output and input pricing, from the strong commitment to open trade, including across the border, and from the reduction of export costs and time for export processing. This has created conditions for farmers and other players to invest and generate profits. These open trade and market-oriented agricultural policies remain critical to future growth and need to be protected by all means. 367. In addition to maintaining a non-distortive agricultural policy, reforms in the seed sector can make an important contribution to agricultural growth. Insufficient supply of seeds and their poor quality in Cambodia have been among main reasons for the relatively slow productivity improvements, putting significant constraint on future agricultural growth mainly because investments in fertilizers and machinery pay off only when combined with the use of improved seeds. Many policy distortions exist in the seed sector, ranging from the state monopoly 135 on the supply of foundation seeds to the lack of clear and conducive regulations for trials, releases, production, and import of seeds. The regulatory and legislative bottlenecks in the seed sector need to be addressed in the upcoming years. 368. Regulations for inputs such as fertilizers can also be improved to reduce costs to farmers. Cambodia has in place the burdensome requirements to importers of fertilizers to register products and apply for import license for each imported lot. The procedures can be simplified through approval of import licenses on the basis of product suitability only, and once received all importers can be allowed to import any quantity of these registered inputs. Strengthening the Environmental Sustainability 369. Embedding environmental sustainability in agricultural policy will become increasingly important. This implies more sustainable expansion of cultivated land though better land use planning, strengthened land tenure security, and maintenance of soil fertility through better use of available land and promotion of sustainable land management practices. International experience shows that the most effective way to promote such practices is to combine the enforcement of natural resource protection with agricultural extension and other public programs that increase land users’ awareness of the longer-term costs and benefits and promote adoption of modern technologies. 370. With more farmers using agricultural chemicals for vegetable and other crop production, the government may wish to strengthen public health regulations related to their safe use and promote alternative programs that ensure safe, effective and environmentally sound pest management. These alternative programs include integrated pest management, where the pest population is controlled through biological controls, cultural practices and the development and use of crop varieties that are pest resistant or tolerant. Improving the Agricultural “Public Goods” Investments 371. Future growth will likely come from productivity improvements, commercialization, and diversification, all of which strongly depend on effective “public goods” investments in agriculture. The relatively small size of most farms in Cambodia and their large numbers require more resources for service provisions than in countries with large farms and few farmers. Public programs proven worldwide to contribute to long-term agricultural growth include those for irrigation, research, extension, education, vocational training, food safety, and provision of market information. While these programs are also critical to Cambodia’s future growth and continued structural transformation, they have little to show in terms of their contribution to past agricultural growth. The main reason is their low efficiency and effectiveness. Many programs have poor design and weak implementation record. Most expenditure goes to capital and wages, with little spent on operation and maintenance. Relatively low funding levels are another reason for the weak record of some of these programs in Cambodia. 372. Agricultural programs, especially extension, can become more gender sensitive. Female-headed households and women in male-headed households have different needs than men due to the different roles they play in agricultural production, and given women’s time constraints and limited mobility which affects their access to training. Hiring and training of more female extension workers, the use of varied information and communication technologies to improve extension outreach, and promotion of more active participation of women in farmer extension 136 groups, as well as nutrition and disaster risk management trainings are all measures that would empower women and bring economic benefits. 373. Irrigation investments deserve special attention. Without reliable access to affordable water in a tropical country such as Cambodia, it is hard to expect intensification and commercialization. Farmers’ access to water in the dry season, when it is mostly needed, has improved only marginally. Irrigation coverage in Cambodia is among the lowest in Asia, in spite of the fact that most resources allocated to irrigation investments have been spent on primary infrastructure. While irrigation coverage needs to be further expanded, this expansion needs to go hand-in-hand with the extension of secondary canals, rehabilitation and upgrade of the existing systems, investments in participatory management of irrigation and drainage infrastructure and better coordination of irrigation/flood management with the delivery of other agricultural public services. 374. Higher attention is also merited to the integration of small farms into modern food value chains. Public extension services are especially important for these farms because they are unlikely to receive sufficient extension services from the private sector, while without improved farm management skills and information they will stay poor and vulnerable, disconnected from economic transformation processes. Small farms, including households participating in social land concessions, can be supported through public services to facilitate stronger links with the agroprocessing and agribusiness industry through productive partnerships (e.g., contract farming and farmer organizations), training in business and other skills, and provision of market information. Helping Develop the Agribusiness and Agroprocessing Industry 375. Future agricultural growth in Cambodia depends on the extent of development of agribusiness and agroprocessing industry. A more developed industry would absorb the increased supply of raw materials, bring more stability to farm prices, create value added, and create nonfarm jobs in rural areas. Alternatively, with no changes to the current situation, urban Cambodians will increasingly eat imported processed foods while the country continues to export raw commodities. 376. Although development of the agribusiness and agroprocessing industry is to be considered part of the improvement in the overall business environment in Cambodia, several areas with a strong impact on agribusiness development will involve government action. They include investments in: (i) increased access to and reduced costs of electricity, to increase profit margins in the processing industry and stimulate private investments; (ii) better food safety and sanitary and phytosanitary capacity, to help Cambodian firms meet importers’ requirements and capture foreign markets and help them compete with the increasing food imports triggered by Cambodians’ rising incomes and demand for more diversified diets; (iii) export logistics, to reduce transaction costs; and (iv) improved access to finance, including through WRS, to create new investment opportunities for the processing and trade industry. 377. Future agricultural growth at its past speed and further structural transformation are possible in Cambodia. A slower growth cannot become a new normal. More attention to sustainability, the quality of public services, an enabling policy environment, and implementation would turn these possibilities into reality. It is worth well trying. 137 Table 64: Summary of the report’s key policy recommendations by implementing agency Policy recommendations Implementing agencies Maintaining a private sector friendly agricultural policy environment Continue the open trade policy and non-distortive Ministry of Commerce (MOC), Ministry of agricultural policy Economy and Finance (MEF), Ministry of Agriculture, Forestry and Fisheries (MAFF) Open up seed sector for private investments through MAFF, Cambodia Agricultural Research and deregulation and institutional strengthening, and reduce the Development Institute (CARDI) costs of importation of fertilizers Strengthening the environmental sustainability Improve land use planning and better enforce land MAFF, Ministry of Land Management, Urban laws/regulations Planning and Construction (MLMUC) Strengthen land tenure security MLMUC Promote sustainable land management practices MAFF (CARDI and Extension) Promote the safe use of agricultural chemicals MAFF Improving the agricultural “public goods” investments As feasible within total government budget, increase the MAFF, SNEC, MEF, Ministry of Labor and budget for core agricultural public goods, especially Vocational Training irrigation, applied research, extension, soil nutrient management, input quality control, food safety, vocational training, and rural roads In irrigation, pay more attention to upgrades and MOWRAM, MAFF rehabilitation of existing systems and participatory management of irrigation infrastructure Improve the quality of agricultural public programs MAFF, MOWRAM Facilitate integration of small farms into food value chains MAFF, MOC Helping develop the agribusiness and agroprocessing industry Improve access to and reduce electricity costs Ministry of Mines and Energy Invest in food safety public infrastructure CARD, Ministry of Health Improve rice trade logistics, including cost reduction Ministry of Public Works and Transport, MOC Remove barriers for introduction of warehouse receipts MEF, National Bank of Cambodia 138 12. REFERENCES Abdullah, A., H. Kobayashi, I. Matsumura and S. Ito. 2008. “World Rice Demand Towards 2050: Impact of Decreasing Demand of Per Capita Rice Consumption for China and India.” Japan and East Asian Regionalism. ADB. 2014a. “Improving Rice Production and Commercialization in Cambodia: Findings from the Farm Investment Climate Assessment.” Asian Development Bank, Manila. ADB. 2014b. “Water Resource Management Sector Development Program, Ninth Quarterly Progress Report (February 1-April 31, 2014).” Asian Development Bank, Phnom Penh. ADI. 2009. “Rice Sector Profile.” Agriculture Development International, Phnom Penh. Agland Investment Services. 2014. “Cambodia International Rice Marketing Study for Six Countries: China, Cote Divoire, France, Germany, Singapore and USA.” Report sponsored by IFC and Agence Francaise Development, Phnom Penh. Alexandratos, N., and J. Bruinsma. 2012. “World Agriculture: Towards 2030/2050.” UN Food and Agriculture Organization, ESA Working Paper, Rome. Bordey, F., P. Moya, J. Betran, C. Launio, A. Litonjua, R. Manalili, A. Mataia, J. Besenio, F. Macalinta, R. Malasa, E. Marciano, M. Milanes, S. Paran, G. Redondo, R. Relado, M. San Valentin, I. Tanzo, E. Tulay, S. Valencia, C. Viray, and C. Yusongco. 2014. “Benchmarking The Cost and Profitability of Paddy Production in Selected Asian Rice Bowls.” Philippine Rice Research Institute, International Rice Research Institute, Benguet State University, and Philippine Council for Agriculture and Fisheries. CDRI. 2012. “Impact Assessment of Farmer Organization on Food Security for Rural Poor.” Report commissioned by the World Bank and AusAID under the Cambodia Food Crisis Capacity Support Partnership Trust Fund funded by AusAID, Phnom Penh. CDTA. 2014. “Economics of O&M of Irrigation Systems in Cambodia.” Capacity Development Technical Assistance Project, Phnom Penh. FAO. 2003. “World Agriculture: Towards 2015/2030.” Food and Agriculture Organization of the United Nations, ESA Working Paper, Rome. FAO. 2014. “A Regional Rice Strategy for Sustainable Food Security in Asia and the Pacific.” RAP Publication 2014/05, Food and Agriculture Organization of the United Nations, Regional Office for Asia and Pacific, Bangkok. FAOSTAT. 2014. FAO Statistics: http://faostat.fao.org/default.aspx (accessed November 20, 2014). Fuglie, K. 2010. “Total Factor Productivity in the Global Agricultural Economy: Evidence from FAO Data.” USDA, Washington, DC. Gardner, B. 2005. “Causes of Rural Economic Development.” College of Agriculture and Natural Resources, University of Maryland, College Park, MD, USA. Gautam, M., and A. Kar. 2014. “Public Expenditure Priorities for Competitiveness. Farm Level Evidence on the Effectiveness of Subsidies in Sri Lanka.” World Bank, mimeo. 139 IFC. 2014. “Preliminary Assessment of the Rice Seed Sector in Cambodia: Policy Recommendations to Overcome Regulatory Constraints.” Investment Climate Business Line, Phnom Penh, Cambodia. IFDC. 2010. “Rapid Apparaisal of Fertilizer Quality in Cambodia.” International Fertilizer Development Center. Report commissioned by the World Bank and AusAID under the Cambodia Food Crisis Capacity Support Partnership Trust Fund funded by AusAID, Phnom Penh. IFPRI. 2012. ASTI Global Assessment of Agricultural R&D Spending: Developing Countries Accelerate Investments. Washington, DC. http://ifpri.org/sites/default/files/publications/astiglobalassessment.pdf. Isvilanonda, S., A. Ahmed, and M. Hossain. 2000. “Recent Changes in Thailand’s Rural Economy: Evidence from Six Villages.” Econ. Polit. Weekly 35: 4650-4656. Ivanich, M., and W. Martin. 2014. “Short and Long-Run Impacts of Food Price Changes on Poverty.” Policy Research Working Paper 7011, World Bank, Washington, DC. Jamora, N. 2014. “Food Demand Trends and Prospects in East Asia.” Report prepared for the World Bank, Washington, DC. Le, Q., E. Nkonya, and A. Mirzabaev. 2014. “Biomass Productivity-Based Mapping of Global Land Degradation Hotspots.” ZEF-Discussion Paper on Development Policy No. 193, Bonn. MAFF. 2014a. “Agricultural Extension Policy in Cambodia. Second Draft.” Ministry of Agriculture, Forestry and Fisheries, Phnom Penh, October 30, 2014. MAFF. 2014b. “Agricutural Statistics.” Ministry of Agriculture, Forestry and Fisheries, Phnom Penh. Nidhiprabha, B., and C. Chamchan. 2005. “SPS and Thailand’s Exports of Processed Food. Food Safety Regulation Concerns: The Developing Country Perspective." Edited by Rajesh Mehtha and J. George. Delhi: Macmillan India. Nippon. 2013. “Sustaining Competitiveness of the Thai Agriculture." Report prepared for the World Bank, Bangkok. NIS. 2014. Cambodian Statistics. The National Institue of Statistics. http://www.nis.gov.kh/index.php/en (accessed November 20, 2014). Rutten, L. 2014. “Review of Potential Constraints for Warehouse Receipts Financing in Cambodia with a Focus on Rice.” Report commissioned by the World Bank and AusAID under the Cambodia Food Crisis Capacity Support Partnership Trust Fund funded by AusAID, Phnom Penh. Sharma, R. 2014. “Outlook for Rice Demand, Supply and Trade.” In Dawe, D., S. Jaffe and N. Santos (eds.) Rice in the Shadow of Skyscrapers: Policy Choices in a Dynamic East and Southeast Asian Settings. Food and Agricultural Organization, International Rice Research Institute, and World Bank. 140 Slayton, T., and S. Muniroth. 2009. “A Road Map for Cambodian Rice Exports.” Report commissioned by the World Bank and AusAID under the Cambodia Food Crisis Capacity Support Partnership Trust Fund funded by AusAID, Phnom Penh. Slayton, T., and S. Muniroth. 2012a. “A More Detailed Roadmap for Cambodian Rice Exports.” Report commissioned by the World Bank and AusAID under the Cambodia Food Crisis Capacity Support Partnership Trust Fund funded by AusAID, Phnom Penh. Slayton, T., and S. Muniroth. 2012b. “Turning Rice into ‘White Gold.” Report commissioned by the World Bank and AusAID under the Cambodia Food Crisis Capacity Support Partnership Trust Fund funded by AusAID, Phnom Penh. Sophal, C. 2008. “Impact of High Food Prices in Cambodia.” CDRI Policy Brief No. 2, Phnom Pehn. Thomas, T., T. Ponlok, R. Bansok, T. De Lopez, C. Chiang, N. Phirun, and C. Chhun. 2013. “Cambodian Agriculture: Adaptation to Climate Change Impact.” IFPRI Discussion Paper 1285, Washington, DC. Timmer, P. and S. Akkus. 2008. “The Structural Transformation as a Pathway out of Poverty: Analytics, Empirics, and Politics.” Centre for Global Development Working Paper 150, www.cgdev.org. Tsakok, I. 1990. Agricultural Price Policy: Practitioner's Guide to Partial-Equilibrium Analysis. Cornell University Press: New York. van der Meer, K. 2014. “Implemeting SPS Measures to Facilitate Safe Trade. Principles and Practice in Cambodia, Lao PDR, Philippines, and Thailand.” Study conducted for the Standards and Trade Development Facility. van der Meer, K., and S. Samrith. 2014. “Implemeting SPS Measures to Facilitate Safe Trade in Cambodia.” Country study conducted for the Standards and Trade Development Facility. Wokker, C., P. Santos, and R. Bansok. 2014. “Irrigation Water Productivity in Cambodian Rice Systems.” Agricultural Economics 45: 1-10. World Bank. 2007. “Philippines: Agriculture Public Expenditure Review.” Technical Working Paper, World Bank, Manila and Washington, DC. World Bank. 2008. “Cambodia Agrarian Structure Study.” World Bank, Washington, DC. World Bank. 2011. “Cambodia Seed Sector Overview.” Report commissioned by the World Bank and AusAID under the Cambodia Food Crisis Capacity Support Partnership Trust Fund funded by AusAID, Phnom Penh. World Bank. 2013a. “Cambodia Vision 2030: Input Papers.” World Bank, Phnom Penh. World Bank. 2013b. “Where Have All the Poor Gone? Cambodia Poverty Assessment 2013.” World Bank, Washington, DC. World Bank. 2014a. “The Agricultural Transformation Matrix.” Presentation, Agriculture Global Practice, World Bank, Washington, DC. World Bank. 2014b. “Cambodia Rice Sector Monitoring Note 2013.” Technical Assistance Report ACS8955, World Bank, Washington, DC. 141 World Bank. 2014c. “Cambodia: National Accounts Data.” World Bank, Phnom Penh. World Bank. 2014d. “Improving Rice Trade Logistics to Help Reach 1 Million Tons Export Target.” Presentation on the Trade Seminar organized by the Ministry of Commerce on September 2, 2014, Phnom Penh. World Bank. 2014e. “Input Notes for Rectangular Strategy III and National Strategic Development Plan 2014-2018 based on the World Bank Investment Climate Assessment 2014.” World Bank, Phnom Penh. World Bank. 2014f. “Myanmar: Capitalizing on Rice Export Opportunities.” Economic and Sector Work Report No. 85804, World Bank, Washington, DC. World Bank. 2014g. “What Does Recent Evidence Tell Us About Poverty in Lao PDR?” Lao PDR Poverty Brief, December 2014, World Bank, Washington, DC. World Bank. 2015. “Cambodia Gender in Agriculture Assessment.” Draft Technical Assistance Note, Phnom Penh. World Bank and AusAid. 2011. “Agriculture, Irrigation, and Rural Roads Sectors: Integrated Fiduciary Assessment and Public Expenditure Review for Cambodia.” Report commissioned by the World Bank and AusAID under the Cambodia Food Crisis Capacity Support Partnership Trust Fund funded by AusAID, Phnom Penh. World Bank and AusAid. 2013. “Study on Access to Financial Services for Small and Medium Enterprises in Cambodia.” Technical Working Paper No. 82585-KH. Prepared by Agrifood Consulting International, Kensington, Maryland. Report commissioned by the World Bank and AusAID under the Cambodia Food Crisis Capacity Support Partnership Trust Fund funded by AusAID, Phnom Penh. WDI. 2014. World Development Indicators. World Bank: http://databank.worldbank.org/data (accessed November 20, 2014). 142 ANNEX 1: ANALYTICAL WORK FROM THE WORLD BANK AND AUSAID PARTNERSHIP 1. The World Bank and Australian Aid signed an agreement in 2009 to produce a series of analytical reports to underpin the efforts of the government and donors to improve food security and promote smallholder-driven agricultural growth in Cambodia. Eight reports were produced on topics ranging from input markets and rice sector development to access to finance, all of which feed into this large report on agricultural transformation, which is also supported through the partnership. 2. These reports are the following:  Rapid Appraisal of Fertilizer Quality in Cambodia prepared in 2010 by the International Fertilizer Development Center.  Seed Sector Overview prepared in 2011 by a team led by M. Chamroeun, the World Bank.  Impact Assessment of Farmer Organizations on Food Security for Rural Poor prepared in 2012 by CDRI.  Roadmap for Cambodian Rice Exports prepared in 2009 and updated in 2012 by T. Slayton and S. Muniroth.  Turning Rice into “White Gold” prepared in 2012 by T. Slayton and S. Muniroth.  The Agriculture, Irrigation, and Rural Roads Sectors: Public Expenditure Review, prepared in 2011 by a team led by P. Eliste, the World Bank.  Study on Access to Financial Services for Small and Medium Agribusiness Enterprises in Cambodia prepared in 2013 by a team led by P. Eliste, the World Bank.  Review of Potential and Constraints for Warehouse Receipts Financing in Cambodia with a Focus on Rice Sector, prepared in 2014 by L. Rutten. 143 ANNEX 2: METHODOLOGY FOR THE 2013 FARM SURVEY 1. This survey investigated the main changes in Cambodia’s agriculture sector over the past 10 years, the key drivers of those changes, and stakeholders’ medium- and long-term perspectives about the future of the agriculture sector. Based on the information from agricultural experts in Cambodia, supported by data gathered from farmers, the findings from this assessment will provide insights to help shape the RGC’s agriculture sector strategies. 2. The survey team used qualitative methods to collect data for this assignment, including key informant interviews and FGDs, complemented by quantitative data collection on farm budgets. The project team collected data at the national, provincial, district, commune, village, and individual farm level. Individual interviews are typically the preferred approach, but FGDs were determined to be more effective for talking with farmers. The survey team also used structured survey tools to gather data from individual farmers for the financial analysis of farm enterprises. Selection of Products 3. The commodities selected for this analysis were chosen because they best characterized the changes in Cambodia’s farming system since the mid-2000s. The team first selected four products that had significant changes in production growth over the past 10 years. In addition, each of these products had specific defining characteristics: (i) rice is the most important crop in terms of GDP contribution, employment, food security, and exports; (ii) maize is characterized by its rapid adoption of improved seeds and integration with the feed industry; (iii) cassava growth is exceptional but its development is subject to intense discussion related to sustainability and biodiversity conservation; and (iv) vegetables are related to nutrition and food safety, and are a good candidate for import substitution. 4. The survey team discussed the selection of these crops during a stakeholder meeting held in Phnom Penh on September 18, 2013. Table 65 summarizes the characteristics of the four selected commodities. Table 65: Product selection for the 2013 field survey Product Justification Related Policies Region and Province Wet season * Staple food * Land TONLE SAP: rice * Large domestic and international * Export promotion * Battambang demand * Food security * Bantey * RGC focus Meanchey * High economic impact * Contribution to food security COASTAL: * 2013 production: 7.14 million tons * Kampot MEKONG: * Takeo * Kampong Cham * Kandal 144 Dry season rice * Staple food * Land TONLE SAP: * Large domestic and international * Irrigation * Battambang demand * Food security * RGC focus * Export promotion MEKONG: * High economic impact * Takeo * Contribution to food security * Kampong * 2013 production: 2.15 million tons Cham Cassava * Subsidiary crop with relatively * Land TONLE SAP: reliable demand (starch, animal feed, * Environmental * Battambang ethanol) thus ensuring relatively sustainability * Bantey stable incomes for upland farms * Export promotion Meanchey * Drastic increase of production from * Agribusiness 2004 to 2013 Promotion MEKONG: * Valorization of upland * Kampong * 2013 production: 7.61 million tons Cham Maize * Production increased 20 percent a * Seed production TONLE SAP: year from 2003 to 2012 * Agribusiness * Battambang * Existence of demand (animal feed) promotion * Bantey thus ensuring relatively stable Meanchey incomes for upland farms * Farmers tend to adopt new MEKONG: technologies (e.g., hybrid seeds) * Kampong * 2013 production: 0.95 million tons Cham * Kandal Vegetables * Production increased 10 percent a * Food safety MEKONG: year from 2003 to 2012 * Pesticide management * Takeo * Import substitution * Seed production * Kandal * Contributes to improve nutrition * Kampong * Improved food safety Cham * From ADB 2009: vegetable production provides highest gross COASTAL: margin per ha * Kampot * 2013 production: 0.41 million tons TONLE SAP: * Battambang * Bantey Meanchey Selection of Sites 5. The team selected the provinces, districts, and communes for the field work based on their contribution to the production of the selected products. The team used data from MAFF and information from various experts in Phnom Penh. In addition, provinces, districts, and communes visited in the 2005 survey were included in the sample if they fit the purpose of the current research questions. In some cases, villages or communes had altered their farming systems, shifting to commodities not part of the survey list; in these cases, a new, more appropriate site selection for the four commodities was decided through discussion with the District of Agriculture Office Chief. 6. At the commune level, the team requested the commune council to provide a list of villages where many farmers produced the selected commodities. The team chose one village from the list proposed by commune leaders. 145 Table 66: Selected locations for 2013 field work Zone (3) Province (6) District (12) Crop (4) Commune (12) Visited in 2005 Mekong Kampong Memot Maize/Cassava Dar - Memot Yes Cham SreySnathor Rice/Maize Prey Poh - Yes PrekDamboke Kandal Sa Ang Vegetable/Maize PrekAmbel No KandalStoeng Rice/Vegetable Siem Reap No Takeo Batti Rice/Vegetable Champei No Tram Kok Rice/Vegetable Tram Kak No Coastal Kampot Chhouk Rice/Vegetable Meanchey - Yes TrapeangPhleang Kampong Bay Rice/Vegetable TraeyKoh - Andong Yes Khmer Tonle Sap Battambang Ek Phnom Rice/Vegetable PrekKhpop- Yes PrekNorin Banan Cassava/Maize Kanty 2 - Chheuteal Yes Bantey Malai Maize/Cassava OuSampou No Meanchey MongkolBorei Rice/Cassava RohatTouk No 7. In summary, the selection sequence was: (i) selection of provinces; (ii) selection of two districts within each province; (iii) selection of one commune within each district; and (iv) selection of one village within each commune. 8. The six sampled provinces accounted for 46 percent of the national production of rice in 56 2013 ; 83 percent of maize production, 57 percent of cassava production, and 60 percent of vegetable production. 56 Broken down by 44 percent of wet season rice production and 53 percent of dry season rice production. 146 Selection of Respondents A. Provinces, districts, and communes 9. At the national, provincial, district, and commune level, the project team conducted key informant interviews with experts in the field of agriculture, rural development, and land management. The list included senior officers at the Ministry of Economy and Finance; the Supreme National Economic Council; the Ministry of Agriculture, Forestry, and Fisheries; the Ministry of Land Management, Urban Planning, and Construction; and responsible staff members of development projects and NGOs working in Cambodia. During the fieldwork, the team interviewed provincial department and district office representatives of the two ministries (MAFF and MLMUPC), as well as the responsible person for agricultural development at the commune level. 10. The survey team used two different sampling methods for key informant interviews: (i) purposive sampling based on the information from initial meetings with MEF, SNEC, and MAFF and the initial identification of respondents by the project team; and (ii) snowball sampling, which consists of interviewing key informants identified by the respondents themselves. 11. The interview guide57 for key informants had two sections, aimed at eliciting information on: (i) changes that happened in the past 10 years in the agriculture sector, their driving forces, and stakeholders’ perspectives for the next five years; and (ii) prices, farmers’ use of inputs including land, and agricultural production. 12. For the first section, interviewers ensured that all possible changes and their causes were discussed during the interview. Topics included: a. Land use and land availability; b. Awareness and use of improved varieties and use of good (certified) seed; c. Awareness and use of other modern inputs such as fertilizer and pesticides; d. Access to agricultural services such as finance, extension, and information on prices; e. Extent of mechanization and types of farm machinery used; f. Use of family and hired labor and dynamics of hired labor costs across geographic locations, and the overall availability of labor; g. Role of women in agriculture; and h. Importance (or not) of farmers’ organizations. 13. In addition, interviewers asked questions about the key constraints to improved productivity and farmers’ perspectives about the agriculture sector over the next five years. B. Villages 14. At the village level, the team used FGDs to get information from farmers. For each village, one FGD was conducted separately for each selected crop for the sake of homogeneity of respondents. The FGD questionnaire had two sections: the first captured the same information collected from key informants but at the village and farm level, and the impact of changes on farm 57 The detailed field tools are listed in the Inception Report. 147 activities; the second section included more detailed questions on prices, yields, and use of inputs for agricultural production. 15. The team again used a purposive approach to sample FGD participants. Focus groups consisted of 10 to 12 farmers, and ideally included different farmers growing the selected crop in three categories of cultivated areas (by size) and two levels of use of modern technologies. The team asked village leaders to identify farmers who met the criteria for farm size (small, medium, and large cultivated areas) and level of use of modern technologies (use of improved seeds, fertilizers and pesticides, agricultural machineries, and irrigation). 16. To assess the level of modern technology use, the project team initially characterized farmers as traditional or modern based on the following: for each category of technology (improved seeds, fertilizers and pesticides, machineries, and irrigation), the farm was rated 2 in case of use and 1 for no use. The maximum attainable score was thus 8 for a farm using improved seeds, modern inputs (fertilizers or pesticides), agricultural machineries (e.g., harvester or power tiller for land preparation), and irrigation. At the other end of the spectrum, a completely traditional farmer was given a score of 4. Farmers’ scores thus ranged from 4 to 8. Farmers scoring less than 6 were qualified as traditional and those scoring 6 or more were categorized as modern. During the analysis, the project team used a statistical method (factor analysis) to distinguish between farmers adopting more modern technologies and farmers using more traditional practices for agricultural production. 17. In summary, a FGD included at least one farmer satisfying the criteria for each cell of the following matrix: Table 67: Matrix of size x technology Cultivated Area (farm size) Small Medium Large Level of use of Low Cell 1 Cell 2 Cell 3 modern technologies High Cell 4 Cell 5 Cell 6 C. Farms 18. At the farm level, the survey team collected data on farm budgets for the selected crops, qualitative information on changes observed by farmers in the past 10 years on their farms, and their perspectives for the next five years. 19. Data on farm budgets for the financial analysis of farm enterprises were collected from individual farmers. In every village, the survey team visited one to two households within each of the six cells in Table 67 (matrix of 3 sizes and 2 levels of technology use). However, not all cells existed in every village; for example, there were not always farmers growing vegetables on large areas (> 0.5 ha). Although information for each cell per village was not always available, the project team ensured that within each agro-ecological zone, each cell contained at least one sampled farmer. 20. The farm budget model was then defined by the four following parameters: 148 Model = M(c, s, l; z) where c is the commodity, s is farm size, l is the level of modernization, and z is the agro-ecological zone. In total, the project produced: (i) 29 budget models of crop production based on five crops (including wet season and dry season rice), three categories of farm size (small, medium, and large), and two levels of technology adoption (low and high), with one category missing (medium- size modern vegetable producers); and (ii) 23 models based on location. The project highlighted any sub-regional differences existing within each agro-ecological zone. 21. The individual farm survey contained three sections:  Section 1: Farm budget  Section 2: Information related to the past 10 years on: o Production trends o Trade trends o Changes in the use of inputs, machineries, and access to agricultural services o Changes in the use of labor and the impact of migration o Changes in marketing o Dynamics of landholdings over the past decade o Source of change in landholdings  Section 3: Data on farms’ and farmers’ characteristics D. Summary of the type and number of respondents 22. Table 68 presents a list of respondents for the survey. Process for Setting Up the Tools for Data Collection 23. The assignment required the use of five different data collection tools: a. Interview guide for provincial and district staff of MAFF b. Interview guide for provincial and district staff of MLMUPC c. Interview guide for key person responsible for agriculture in commune d. FGD guide for groups of farmers e. Questionnaire for individual farmer survey, including farm budget sheet 24. The survey team pretested the FGD guide and the individual farmer questionnaire with farmers in Takeo province. The objectives were to check for consistency and flow, to identify other relevant responses to each question in the guide, and to determine the timing of the interview and FGD. 25. The questionnaire was translated into Khmer before the field staff training. The training lasted for five days, including a second pretest of the revised questionnaires in the field, and another session to update the tools based on feedback from the field. 26. After the stakeholder consultations held on September 18, 2013, additional questions were added on gender, changes in the role of women in agriculture, the impact of migration on agricultural labor, and households’ consumption. Questions about who led the changes were also added. 149 Table 68: List of respondents to 2013 survey Categories of Respondents Objectives Location Executive Senior Staff at MEF, Information on the project Phnom Penh MAFF, SNEC members Guidance World Bank Guidance Phnom Penh Administrative Expectations Technical staff at CDRI, FAO, Statistical information Phnom Penh CAVAC, HARVEST/USAID, Documents IDE, LASED Experience Lessons learned Suggestions Provincial Department and Guidance In each of the 6 selected District Office of MLMPC Policy information provinces and 12 districts Data validation Provincial Department and Guidance In each of the selected 6 District Office of MAFF Statistical information provinces and 12 districts Find out other actors to talk to Commune Authority (Council) Guidance In each of the 12 selected Data validation communes Suggestions on names of villages Group of Farmers (FGD) Information In each of the 12 selected Data validation villages Suggestions on names of farmers by size, technology use Individual farmers representing Information 1 to 2 in each of the 12 selected the cell (crop x farm size x level Data validation villages (if possible), for each of technology use) Farm budget model selected crop, farm size, and level of technology use Quality Control 27. The survey team consolidated the information received from key informants and FGDs every day. Each completed interview was signed by the interviewer, and controlled and signed by the supervisor. For each questionnaire, the team: a. Checked the consistency of the gathered information; b. Checked its completeness; and c. Checked the reliability and accuracy of answers. 28. In addition, the supervisor of each field team communicated to the field survey manager on a daily basis the work progress, any encountered issues, and major findings worth sharing with the other teams. Data Analysis 29. The team used qualitative data gathered from key informants to explain why and how Cambodia has attained such a remarkable improvement in agricultural production in the past 10 150 years, and to identify opportunities and threats for the future of the agriculture sector. The ultimate goal was to recommend policies to ensure that the agriculture sector remains central to Cambodia’s overall economic development. 30. For each key question (such as land use and land availability; use of inputs and machinery; issues of labor; access to agricultural services; constraints to agricultural production; and perspectives about the sector), the project team produced tables summarizing the findings by agro- ecological zone and by farm size. 31. The team used farm-level data to compute financial indicators of farmers’ costs and profitability, including gross margins, net profit per hectare, and total production, as well as gross profit per day of family labor and net profit per day of total labor. The team also estimated the financial rates of return: gross margin over variable cost and net profit over total variable cost. 32. For the analysis of returns to labor, the project team compared the indicators of returns to labor from various farm enterprise models to prevailing nonfarm returns to unskilled labor in key economic sectors (manufacturing, construction, and services). 33. The team also conducted a trend analysis explaining trends in cost structures, profitability, and cash flows, and use of labor and machinery between the 2005 and 2013 farm enterprise models. Information from key informant interviews and the qualitative data were used to compare the factor dynamics between 2005 and 2013 in farming systems and to identify the causes of these changes. 34. The team benchmarked the cost of production and input uses against available farm enterprise data from other ASEAN countries and financial and returns to labor indicators from the above farm enterprise models. The team assessed the comparative advantage of Cambodia’s agricultural products relative to other countries in the region, using available data on comparative advantage indicators from recent analytical work and economic literature. To that end, the project team computed: a. Comparative advantage indicators using financial farm enterprise models as a starting point; economic prices (shadow prices) for internationally traded outputs and inputs; and other economic prices (such as for labor) that are adjusted for public transfers (subsidies and taxes); b. Indicators of social profitability, to measure the comparative advantage of various farm enterprise variations; and c. The DRC indicator to assess Cambodia’s comparative advantage in producing agricultural commodities relative to its neighbors. 151 ANNEX 3: DETAILED FARM BUDGETS BY CROP 1 Farm Budget for Wet Season Rice 1. From 2005 to 2013, the nominal gross margin on wet season rice production increased by 55 percent, from $158/ha to $245/ha. In real terms this increase was close to zero. Within the same period, gross revenue increased by 1.2 times and total variable costs by 1.7 times. In real terms, using the GDP deflator from 2005 to 2013 as an adjustment factor,58 the increase was much smaller. 2. The changes in gross revenues were mostly driven by prices. From 2005 to 2013, wet season paddy rice prices increased 98 percent in Cambodia, rising to $247 per ton from $125 per ton. For reference, FAOSTAT reported a similar trend in Thailand, with an increase of 144 percent over the same period, 50 percent higher than in Cambodia, but the Thai prices might have been altered by the policy of subsidizing rice farmers. On average, this variation in prices should lead to a doubling of the nominal values of production, thus significantly affecting nominal gross margins, assuming no or only slight changes in productivity. 3. Wet season rice producers used increasingly more modern inputs. A comparison of the cost structure between 2005 and 2013 shows that farmers used more fertilizers for wet season rice production. In 2013, traditional farmers spent on average $49/ha (75kg/ha) to purchase chemical fertilizers, up from almost no expenditures in 2005. For modern users, the amount spent increased two-fold, from $30-$40/ha (100-133 kg/ha) to over $80/ha (125kg/ha). However, between 2005 and 2013, the price of urea and DAP also increased almost two-fold, from $0.35/kg to $0.68/kg, an increase that may partially explain the increase in chemical fertilizer expenditures. Wet season rice producers also used more pesticides and herbicides as farm size increased. This expenditure amounted to $4/ha for small, $6/ha for medium, and $12/ha for large farms. Expenditures on herbicides, an alternative to manual weed control, constituted the main part of the costs of chemicals (other than fertilizers). 4. The relative share of costs allocated to hired labor decreased as farm size increased. For small farms, the relative share of labor in the partial farm budget represented more than half of total variable costs. The percentage decreased slightly below 50 percent for medium-size farms and dropped further to a quarter for large farms. Indeed, results from the FGDs and the interviews with MAFF and Ministry of Land in the regions and districts show that it is increasingly difficult to find agricultural workers in rural areas. In addition, the daily cost of hired labor increased four- fold, from $1.0-1.2/day to $4.5/day between 2006 and 2013, according to the survey data. These factors forced medium-size and large farms to find better alternatives and more cost-efficient approaches to substitute for the use of hired labor. 5. On the other hand, the cost structure shows the relative importance of expenditures on services for large farms. In 2013, 45 percent of the total production costs for large wet season rice producers were allocated to services. This proportion was 27 percent for medium-size farms and 22 percent for small farms. On a per hectare basis, large farms spent 35 percent and 25 percent more on services than small and medium-size farms, respectively. Most of these costs were related to harvest and post-harvest operations, as large farms sought services from businesses renting combines. According to the FGD results, farmers in Cambodia expected an increasing shift of 58 The cumulative change of the GDP deflator between 2005 and 2013 was 35 percent. 152 harvest and post-harvest operations from manual labor to the use of mechanized agricultural equipment. 6. Each province has its own specific cost composition for producing wet season rice depending on the cost of labor, technology, and connectivity to urban and industrial zones. Farmers in Takeo (65 percent) and Kampot (50 percent) allocated a larger percentage of their total variable costs to labor. In contrast, farmers in Battambang had the lowest share of labor costs (6 percent). Expenditures on services ranged between 17 percent (Kampot) and 47 percent (Battambang), and close to 40 percent in Kandal, Kampong Cham, and Bantey Meanchey. In most provinces, rice farmers spent an average of 16-22 percent of their costs on the purchase of fertilizers, with the exception of Kampong Cham, where the share was about 6 percent ($29/ha). These differences in cost composition illustrate the diversity of the techniques used by farmers to produce wet season rice. For example, direct seedlings are more commonly used in Battambang and Bantey Meanchey. They are also an indication of the cost and availability of agricultural inputs within each province. Farmers located farther away from urban centers and/or borders with neighboring countries might opt for a more labor-oriented approach, favoring the use of family and hired labor. Farmers closer to urban centers and economic industrial zones might adopt a more capital-oriented approach – e.g., greater use of herbicides, and use of combines for harvest and post-harvest operations. 7. The choice of technology is a critical component of profitability in wet season rice production: the gross margins of modern farmers were more than twice those of traditional farmers ($399/ha versus $178/ha). Modern technology users were characterized by higher expenditures on fertilizers ($99/ha) and lower expenditures on labor ($130/ha) compared to traditional farmers, who allocated about $5/ha for fertilizers and $259/ha for labor. These differences should allow modern farmers to reach higher yields, and indeed, modern farmers had a 25 percent higher yield compared to farmers using traditional practices. 8. According to MAFF statistics, average yields for wet season rice have slightly increased since 2006, to 2.9 tons/ha in 2012. In 2013, data collected by the project team for the purposively selected farmers showed an average yield of 2.6 tons/ha, varying from 2.4 tons/ha for farms using traditional technologies to 3.0 tons/ha for farms using modern technology. Although the current study did not intend to measure average yield, it is reassuring to find these data in the range of official figures. 2 Farm Budget for Dry Season Rice 9. The data show that on average, farmers earned higher gross margins per hectare in nominal terms in 2013 than in 2005, regardless of farm size. The magnitudes of the change were higher as farm size increased. In 2005, gross margins for small dry season rice producers ranged from $200/ha to $391/ha; in 2013, they ranged from $276/ha to $311/ha. For medium-size farms, the changes were higher, with an average $283/ha, almost 1.5 times the 2005 gross margin for this size category. Large farms showed the largest increase, with the 2013 gross margin on average three times the recorded 2005 gross margin due to higher use of inputs, which resulted in higher productivity. 10. On average from 2005 to 2013, dry season rice yields also increased, from an average 3.0 tons/ha to 4.8 tons/ha for the surveyed farms. Over the same period, statistics from MAFF show the yield increase of 500 kg, reaching 4.4 tons/ha in 2012 from 3.9 tons/ha in 2005. Data collected by the project team show the average 2013 farm gate price higher by 75 percent in nominal terms 153 compared to the 2005 situation; the price rose from $118.5/ton in 2005 to $206/ton in 2013. This increase was moderate compared to the paddy price changes in Thailand, which increased 144 percent from $172/ton in 2006, but was comparable to the farm gate prices in the Vietnam Mekong region at $240/ton. This difference in prices makes profitable the informal export of paddy from Cambodia to Vietnam and Thailand. 11. Dry season rice producers’ revenue increased less than their costs; the total increase in their margins, however, was still positive. Data collected by the project team in 2005 and 2013 show a 150 percent increase in gross returns, attributable to the combination of higher price and higher yield. However, the much higher increase in total variable costs (218 percent) resulted in an increase of only 79 percent in the gross margin (in nominal terms). This difference varied across type of farm, level of technology use, and location. 12. Use of modern technology is no guarantee of higher gross margins. In a few cases, dry season rice producers using modern technologies received lower gross margins compared to farmers using traditional practices. Besides the use of modern technologies, other factors such as soil quality, labor skills, and quality of seeds are important. In some cases, the expected returns to intensive use of chemicals, improved seeds, and irrigation are offset by gaps in these other factors. If all conditions were the same, modern technology would clearly be better than traditional methods. But all conditions are not the same, particularly when isolated observations are compared, as in this study. This finding seems to be counterintuitive and goes against the conventional wisdom about the advantage of modern technologies. Indeed, the distinction between modern and traditional is very qualitative, based on the level of use of factors such as irrigation, improved seeds, fertilizers, chemicals, and mechanization. Modern farms are expected to spend more on these inputs; e.g., modern small farms spent $228/ha on fertilizers versus $117/ha for traditional small farms; medium-size farms showed similar patterns ($168/ha versus $102/ha) and so did modern large farms ($203/ha versus $100/ha). The same disproportions were observed for herbicides and seed expenditures; this latter may reflect the use of high-quality improved seeds. Modern farms are also characterized by lower expenditures on hired labor regardless of their size: $113/ha for small traditional versus $32/ha for small modern farms; $98/ha for traditional medium- size versus $29/ha for medium-size modern farms; and $96/ha for large traditional versus $42/ha for large modern farms. The costs of harvest were very similar for both types of farms. The most striking difference was in expenditures on irrigation. Overall, the field interviews found that gross margins for modern farms ($255/ha) were lower than those of traditional farms ($336/ha). In summary, increasing yield is not enough if it comes at the cost of margins or sustainability. 13. Dry season rice producers adapt their production strategies to maximize their profit according to their assets: labor cost shares decrease and services cost shares increase with farm size. The analysis shows that regardless of farm size, gross margins remained similar, ranging from $283/ha for large farms to $306/ha for small farms. As with modern versus traditional farms, the structure of costs for different sizes of farms was quite different, especially in the use of labor- alternative technologies. For example, large farms used increasingly less manual labor (e.g., on weed control), using chemicals as substitute, and reapers and threshers or combines for harvest and postharvest. This resulted in lower costs for labor as farm size increased. In dry season rice farm budgets, expenditures on labor amounted to $53/ha, $140/ha, and $204/ha for large, medium, and small farms, respectively. On the other hand, larger farms reported higher expenditures for herbicides ($21/ha for large versus $9/ha for small farms). 154 14. The gross returns to dry rice production were similar for all provinces; however, the production costs showed huge disparities. Takeo, the province with the highest dry season rice production (460,000 tons in 2012, 21 percent of total production – Source: MAFF), also had the highest gross margin, $384/ha. The farm budget in Takeo was characterized by the absence of irrigation costs, which may denote the existence of good public irrigation infrastructure. The cost of labor in Takeo was the highest of all four provinces in this analysis, at $192/ha. The dry season rice farm budget in Battambang had a gross margin of $334/ha, slightly less than that in Takeo, but Battambang is “penalized” by high expenditures on irrigation ($132/ha). It also had high expenditures on chemical fertilizers ($186/ha). The combination of these investments brought about a higher yield, above 5.1 tons/ha, thus generating the highest gross return, at $1,047/ha. If farmers in Battambang did not pay as much for irrigation, their gross margin would be closer to farmers in Takeo. The total dry season rice production in Battambang was around 100,000 tons in 2013 (MAFF). Kandal was in third place with a gross margin of $288/ha. The province produced about 270,000 tons of dry season paddy (12 percent of total production) in 2012. It has the disadvantage of lower gross returns, a direct consequence of lower yield (4.3 tons/ha versus 4.9 tons/ha for Takeo for surveyed farmers). Bantey Meanchey’s gross margin was the lowest of the four provinces, at $175/ha. Its farmers’ expenditures on seeds were almost double those of Takeo’s farmers; they also had the highest expenditures on fertilizers ($207/ha) and the highest cost of irrigation ($150/ha). Their yield was similar to yields in Takeo and Battambang, but because of the relatively lower prices of products (gross return of $970/ha) and the high costs previously mentioned, their gross margin fell to $175/ha. 3 Farm Budget for Cassava 15. Overall, for all farm sizes and regardless of the level of technology use, the gross margins per hectare for cassava significantly increased in nominal terms between 2005 and 2013, varying from 86 percent for medium-size farms using traditional practices to 300 percent for small farms adopting modern technologies. In real terms, the increase in per hectare gross margins was more modest, varying from 51 percent to 265 percent for the two categories of cassava farmers previously mentioned. These changes were mostly driven by prices. From 2005 to 2013, fresh cassava prices increased 200 percent in Cambodia, rising to $59.4/ton from $19.8/ton. For reference, FAOSTAT reported a similar trend in Thailand, with an increase of 166 percent over the same period. The RGC policy on feeder roads rehabilitation in the past 10 years might have contributed to higher changes in farm gate prices. 16. Use of modern technologies is positively correlated to an increase in gross margin per hectare. Overall, cassava farmers using more modern technologies had a 65 percent higher gross margin per hectare compared to farmers using traditional practices. The difference was lower for large farms compared to medium-size and small farms. In the 2013 survey, modern technology adopters recorded a gross margin of $695/ha, about 64 percent higher than that of their traditional counterparts, at $422/ha. This difference is partly explained by the difference in yields: 21.9 tons/ha versus 17.2 tons/ha for modern and traditional technology users, respectively. For cassava, use of modern technologies seems to be limited to the adoption of high-productivity varieties, which may account for a productivity increase of 24 percent, and better crop management practices. Chemical and organic fertilizers are barely used by farmers, and represent only about 1 percent of total variable costs (slightly less than $2) for the surveyed cassava producers. The 2013 survey did not shows significant differences between the recorded average farm gate prices for the 155 two groups of producers. Traditional farmers sold fresh cassava at $71/ton, very similar to the $73/ton received by modern farmers. 17. According to MAFF statistics, average yield for cassava fluctuated between 2006 and 2012, with a reported yield of 22.5 tons/ha in 2012. Data collected by the project team show a lower average yield of 18.9 tons/ha, varying from 14 tons/ha for medium-size farms using traditional technologies to 27 tons/ha for farms using modern technology. Although the current study did not intend to measure average yield, it is reassuring to find the survey data in the range of official figures. Further analysis by category of farm size shows a slight inverse relationship between productivity and farm size: larger farms tended to have lower productivity per hectare compared to small farms. The difference was mostly between the small and other farm sizes, with a gap of close to 10 percent. 18. The most commonly observed changes in the cost structure were an increase in the relative cost of cassava cuttings and a decrease in the relative cost of services. The share of cutting costs out of total variable costs increased from 2-3 percent to 20-30 percent for farmers using traditional methods between the two periods. In 2013, the gap in the relative costs of cassava cuttings decreased between the levels of technology use. Overall, traditional farmers spent about 17 percent less than modern farmers on cassava cuttings. Indeed, use of improved varieties is one of the most cost-efficient technologies to improve productivity. Farmers expanding cassava plantation, in some cases to lower fertility land, would get higher productivity if they adopted more modern varieties of cassava. Modern farmers were also characterized by lower costs of services ($253/ha versus $314/ha), higher costs of labor ($257/ha versus $207/ha), and higher use of fertilizers and chemicals although in lower magnitude. 19. Low use of fertilizers, including organic fertilizers is a concern for the sustainability of cassava production. The cassava farm budgets showed that for any category of farm size, the amount spent on fertilizers remained very low, on average less than 1 percent of total variable costs. Out of the 26 observations, no farmers reported using manure or compost, one out of 5 used urea, and only one used DAP. The rates of use were around 50 kg/ha for urea (for users only) and 25 kg/ha for DAP (one case). 20. The analysis did not find significant differences across provinces. The magnitudes of revenue and total variable costs remained very similar across surveyed farm budgets in three of the six provinces surveyed. The ratios of gross margins to revenue per hectare were quite similar, ranging from 35-42 percent. In addition, the structures of total variable costs were not dissimilar. 21. About half of the interviewed cassava farmers dried their products before commercialization. Farmers who opted to sell chips received higher gross margins ($505/ha versus $452/ha). Their total variable costs were also higher by 4 percent, including a 12 percent increase on labor expenditures, but these costs were offset by a 19 percent higher final product price. The 12 percent gain on gross margin was mostly due to the difference in average yield (18.3 tons/ha versus 16.9 tons/ha) and the 10 percent value added to price from processing. 156 Table 69: Comparison of key indicators across type of commercial cassava (chips vs. fresh) Total Total Labor Gross Variable Yield Unit Price Processing Revenue Costs Margin Costs (tons/ha) (Riel/ton) ($/ha) ($/ha) ($/ha) ($/ha) No 1,224 772 197 452 16.9 72.3 Yes 1,306 801 220 505 18.3 79.3 % Change 7% 4% 12% 12% 8% 10% 4 Farm Budget for Maize 22. The trend between 2005 and 2013 for maize indicates a reduction in the gross margin per hectare received by farmers in real terms. In general, the gross margin per hectare is affected by the prices of outputs and inputs and crop yields. Over this period, recorded prices for maize grain increased from an average $140/ton to $230/ton (i.e., 64 percent, according to data recorded by the project team). For reference, between 2006 and 2011, farm gate prices for maize in Thailand increased 110 percent, from $119/ton to $250/ton (FAOSTAT). If the costs of production follow the same increase as the price of products, then the change in yields will determine the direction of the nominal change in gross margins. Maize yield over the period increased 25 percent according to MAFF statistics. On the other hand, input prices increased at a higher rate; e.g., the price of nitrogen fertilizers in 2012 was about three times higher than in 2006 (FAOSTAT). Based on these facts, it can be assumed that there was a decrease in real terms in the level of gross margins received by maize farmers. Information collected in 2005 and the results of the survey conducted by the project team in 2013 support this assumption. In 2005, maize farms received an average gross margin of $574/ha with a yield of 6.0 tons; this margin is 90 percent more than the same measure calculated from a similar farm budget in 2013 ($304/tons). The trend is even more negative in real terms. 23. Maize production is responsive to the use of modern inputs such as fertilizers, chemicals, and improved seeds. Farmers using modern inputs to produce maize are more likely to have a higher gross margin, and the gain increases as farm size decreases. In 2013, the overall gross margin for modern farmers was $396/ha and $211/ha for traditional farmers. There were large disparities across farm size: small farms using modern technologies had twice the margin of farmers practicing traditional technologies ($282/ton versus $118/ton). The same advantages of modern farmers were observed, albeit at lower magnitude, for large farms ($522/ton versus $253/ton). The budgets provided by maize producers showed that users of modern technologies spent about $83/ha on improved seeds versus $71/ha for traditional farmers; their expenditures on fertilizers and chemical amounted to 20 percent of total variable costs versus 5 percent for users of traditional practices. Modern farmers had 15 percent higher yields across all size categories. 24. According to MAFF statistics, the average yield for maize fluctuated between 2006 and 2012, with a reported yield of 4.4 tons/ha in 2012. For the last four years, average yield moved up and down around 4 tons/ha. Data collected by the project team show lower average yield (3.94 tons/ha), varying from 0.8 tons/ha to 10 tons/ha, with a median of 3.5 tons/ha. Although the current study did not intend to measure average yield, it is reassuring to find these data in the range of official figures. 157 25. Analysis of the cost structures showed that the cost of improved seeds was the most important budget line item, regardless of farm size and level of technology adoption. This cost represented between 16 percent (for small modern farms) and 24 percent (large modern farms) of total variable costs. The enthusiasm for using improved maize seeds, particularly hybrid seeds, was often highlighted by respondents during the expert interviews and FGDs. However, this does not automatically mean that yield will be higher. Many farmers complained of the existence of fraudulent seed suppliers. The second most significant expenditures were for land preparation. Depending on the type of farm, this cost ranged between 16 percent (small traditional) to 34 percent (medium-size traditional) of total variable costs. Note that this partial budget did not take into account the cost of family labor or that of work by farmers for their own agricultural production. Modern farmers were also characterized by a high percentage of cost allocated to fertilizers and chemicals, ranging from 9 percent for medium-size farms to 21 percent for small and large farms. These values contrast with the lower percentages for traditional farmers, which ranged from 4-5 percent of total variable costs. 26. The analysis found significant differences across provinces in the gross margin, total variable costs, total revenue, and cost structure. The magnitudes of revenue and total variable costs were quite different. Farmers in Bantey Meanchey and Kandal had higher revenues ($839/ha and $1,133/ha, respectively), while revenues in Kampot and Battambang were quite low (around $370/ha to $402/ha). Kandal exhibited the highest total variable costs at $634/ha, comprising $85/ha for improved seeds, $105/ha for fertilizers and chemicals, and $177/ha for land preparation. This province also had the highest gross margin, at $500/ha. 5 Farm Budget for Vegetables 27. Data from 2005 and 2013 show that the nominal gross margins per hectare for small farms growing vegetables decreased by about one-fifth. Medium-size farms exhibited an increase of more than 50 percent and large farms doubled their gross margins compared to 2005. Gross margins for small farmers in 2005 ranged from $415/ha (cucumber production in Battambang) to $1,775/ha (watermelon production in Kampot); for medium-size farms, the highest gross margin ($1,100/ha) was observed in Pursat for cucumber production and the lowest ($77/ha) in Sihanouk region for watermelon production. In 2013, a similar spread of gross margins per hectare was observed, especially for small farms, with values ranging from $164/ha for watermelon production in Bantey Meanchey to $3,450/ha for eggplant production in Kampot. For medium-size farms, the range was quite narrow: from $635/ha for mustard production in Kampong Cham to $1,302/ha for watermelon production in Battambang. Data for large farms were scarce and thus very difficult to interpret. The only case reported in 2005 was a large cucumber producer in Pursat with gross margin of $134/ha. In 2013, one large farm was observed producing watermelon and cucumber in Battambang. From these pieces of information, nominal gross margins for small vegetable producing farms appeared to decrease; the changes were more favorable for medium-size and large producers. In real terms, the trend was negative for small farms, stagnant for medium-size farms, and positive for large farms. One complication is that these data related to numerous types of vegetables from different provinces, making comparisons difficult. For example, in 2013, the project team analyzed several farm budgets for cucumber, mustard, watermelon, and eggplant producers in Battambang, Kampot, Kampong Cham, Bantey Meanchey, and Kandal provinces. In 2005, the data were collected in Pursat, Battambang, and Sihanouk provinces for cucumber and watermelon production. 158 28. Prices of vegetables in nominal terms increased one- to three-fold between 2005 and 2013, depending on the type of vegetable. The price of watermelon tripled from 2005 to 2013, reaching $277/ton versus $84/ton in 2005; cucumber prices more than doubled, from $86/ton to $187/ton. The price of eggplant did not change much compared to other vegetables, rising to $200/ton from $125/ton. In real terms, the increases were more modest: double for watermelon and about one- third for cucumber, while eggplant prices remained the same. Comparable figures from Thailand increased within these ranges but exhibited different patterns, with watermelon prices the least increased. Thai cucumber prices had very similar increases to Cambodian farm gate prices. On the other hand, statistics from MAFF show a steady increase in average yield and production of vegetables in the past four years. Based on these two data sets, farmers’ total revenue increased on average but the change in gross margin ultimately depends on the increase in production costs. 29. From 2006 to 2013, average total revenue increased three-fold, from $836/ha to $2,842/ha. This is the result of a combination of increases in yields and prices. However, during the same time, total variable costs increased eight- to nine-fold, from $161/ha to $1,481/ha. The increase in costs was attributable to an increase in input prices, the higher rate of use of modern inputs, and the higher amount of hired labor. For example in 2005, vegetable farmers spent about $23/ha for seeds, the same for fertilizers, and $4/ha for chemicals. In 2013, the figures were $141/ha for seeds, $341/ha for fertilizers (half of it manure), and $130 for pesticides (two-thirds for insecticides). In addition to an increase in the price, the four-fold increase in expenditures on seeds could be due to the use of higher-quality varieties to better respond to market demand. Expenditures on labor skyrocketed to $802/ha, an increase of about 16 times compared to the $50/ha in 2005. Again, this is the combination of an increase in the cost of labor (1 to 4 based on the 2013 data) and the shift to more labor-intensive activities, requiring greater use of hired labor. For the future, this could become a constraint to further development of vegetable production. 30. In 2013, the cost structures show the dominance of labor costs (54 percent), followed by the cost of inputs (41 percent). The share of services is very low, especially for small and medium- size farms. Despite the constraints linked to hiring rural laborers, vegetable farmers still relied heavily on manual labor, which represented about two-thirds of total variable costs. The ratios were lowest in Kampot, Takeo, and Kampong Cham provinces but still remained around 50 percent of total variable costs. Indeed, these variations were directly related to the type of vegetables grown in these provinces. Cucumber production required the highest amount of labor (72 percent of total variable costs), followed by eggplant production (64 percent), and mustard (50 percent). The lowest share was for watermelon production (42 percent), which also had the lowest average total production costs of $592/ha. In general, the share of labor decreased and share of services increased as farm size increased. 31. As expected, vegetable farmers adopting modern technologies used a high amount of inputs including manure, chemical fertilizers, pesticides, and seeds. Input expenditures for modern farmers represented five times the expenditures of traditional farmers, but there was variability across the type of vegetables and farm size. For example, small modern cucumber producers spent $728/ha versus $155/ha for traditional farmers; the figures were $920/ha versus $378/ha for mustard; and $636/ha versus $250/ha for eggplant producers. The variations were more difficult to assess for medium-size and large farms since very few of them were characterized as modern in the 2013 survey. In general, modern farmers allocated about 45 percent of their costs to inputs versus 30 percent for traditional vegetable producers. 159 32. Vegetable production is also characterized by intensive use of organic fertilizers. The use of manure or compost represented on average 9 percent of the total variable costs for vegetables. The share was highest in Takeo (17 percent) and lowest in Battambang and Bantey Meanchey (3- 4 percent). Modern and small farms were also the most intensive users of organic fertilizers, with expenditures reaching $150/ha. 33. Intensification does not always mean higher gross margins for vegetable production, as already reported for the case of dry season rice production. Besides modern inputs, other factors such as irrigation system, commercialization, type of vegetable crop and varieties, and crop management may enter into consideration. For cucumber production, small farms allocating $728/ha for inputs reported a gross margin $650/ha, 3.7 times higher than the gross margin of small traditional farmers. However, for mustard and eggplant production, modern small farms received a lower gross margin compared to their traditional peers: $760/ha versus $983/ha for mustard and $689/ha versus $1,107/ha for eggplant, despite the two- to three-fold greater expenditure on modern inputs. 34. Access to and use of services remained insignificant for vegetable production. Only large and medium-size farms required substantial expenditures on services as part of their cost structures. In most cases, there were no reported expenditures. Either farmers supplied these services using family assets, which are very plausible because of the relatively small cultivated areas, or the services were not available to or known by farmers. For the future, the farm budget for vegetables should include a substantial amount for irrigation, a critical element for reaching high and sustained productivity. During the interviews and FGDs, very few respondents were aware of the possibility of using drip irrigation or greenhouses to improve vegetable production and quality or to lower dependency on uncontrolled factors such as rainfall. 160 Table 70: Wet season rice farm budget by technology and farm size Level of Technology Use and Farm Size Modern Traditional Modern Traditional Modern Traditional Small Small Medium Medium Large Large Average cultivated area 0.60 0.54 1.33 1.27 3.80 5.00 (ha) Yield (tons/ha) 3.38 2.48 3.25 2.81 3.33 2.77 Inputs ($/ha) Seeds 42.50 19.81 30.42 17.13 49.69 19.84 Manure 22.50 41.59 8.17 23.70 13.08 1.67 Chemical fertilizers 173.27 43.28 81.88 60.82 79.08 36.13 Herbicides - 0.47 6.27 1.31 7.03 12.33 Insecticides 2.68 0.67 4.17 0.43 1.21 0.83 Subtotal inputs 240.95 105.82 130.90 103.38 150.09 70.80 Labor ($/ha) Land preparation - - - - - - Plantation, transplantation 156.25 124.72 148.75 90.04 26.75 117.92 Weed control 23.04 30.61 6.92 26.50 14.05 2.50 Crop management 63.57 119.27 29.17 100.85 6.04 114.00 Harvest - 11.36 - 10.26 - - Post-harvest 5.08 12.86 1.67 14.54 3.77 5.00 Subtotal labor 247.93 298.82 186.50 242.19 50.61 239.42 Services ($/ha) Land preparation 85.00 98.65 118.33 92.00 144.80 158.33 Plantation, transplantation - - - 4.04 - - Harvest 25.00 - 100.00 7.50 81.25 - Post-harvest 37.14 32.53 48.96 31.74 19.85 56.25 Subtotal services 147.14 131.18 267.29 135.27 245.90 214.58 Total variable costs ($/ha) 636.02 535.82 584.69 480.85 446.60 524.80 Total revenue ($/ha) 1,157.86 652.68 1,000.00 717.05 787.80 595.83 Gross margin ($/ha) 521.84 116.86 415.31 236.20 341.20 71.04 Returns to labor ($/day) 8.73 1.62 9.24 4.05 27.98 1.23 161 Table 71: Wet season rice farm budget by farm size Size Small Medium Large Total Average cultivated area (ha) 0.56 1.28 4.25 1.80 Yield (tons/ha) 2.68 2.89 3.12 2.89 Inputs ($/ha) Seeds 24.85 19.62 38.50 25.62 Manure 37.35 20.79 8.80 22.40 Chemical fertilizers 72.17 64.77 62.98 66.35 Herbicides 0.36 2.24 9.02 3.37 Insecticides 1.12 1.13 1.06 1.11 Subtotal inputs 135.85 108.54 120.36 118.85 Labor ($/ha) Land preparation - - - - Plantation, transplantation 131.73 101.05 60.94 99.69 Weed control 28.93 22.83 9.72 21.31 Crop management 106.89 87.41 46.53 82.81 Harvest 8.83 8.33 - 6.45 Post-harvest 11.13 12.12 4.23 9.94 Subtotal labor 287.51 231.75 121.41 220.21 Services ($/ha) Land preparation 95.62 96.94 149.88 109.41 Plantation, transplantation - 3.28 - 1.59 Harvest 5.56 24.84 50.78 25.87 Post-harvest 33.56 34.97 33.50 34.23 Subtotal services 134.73 160.03 234.16 171.10 Total variable costs ($/ha) 558.09 500.32 475.93 510.16 Total revenue ($/ha) 764.94 770.10 715.82 755.53 Gross margin ($/ha) 206.85 269.79 239.89 245.37 Returns to labor ($/day) 2.99 4.83 8.20 4.62 162 Table 72: Wet season rice farm budget by technology use Level of Technology Use Modern Traditional Total Average cultivated area (ha) 2.42 1.53 1.80 Yield (tons/ha) 3.32 2.70 2.89 Inputs ($/ha) Seeds 42.47 18.30 25.62 Manure 13.49 26.27 22.40 Chemical fertilizers 98.76 52.26 66.35 Herbicides 5.40 2.49 3.37 Insecticides 2.39 0.56 1.11 Subtotal inputs 162.50 99.87 118.85 Labor ($/ha) Land preparation - - - Plantation, transplantation 89.25 104.23 99.69 Weed control 13.71 24.62 21.31 Crop management 24.49 108.17 82.81 Harvest - 9.25 6.45 Post-harvest 3.40 12.78 9.94 Subtotal labor 130.84 259.06 220.21 Services ($/ha) Land preparation 124.90 102.68 109.41 Plantation, transplantation - 2.28 1.59 Harvest 75.63 4.24 25.87 Post-harvest 32.04 35.18 34.23 Subtotal services 232.57 144.37 171.10 Total variable costs ($/ha) 525.91 503.31 510.16 Total revenue ($/ha) 925.47 681.65 755.53 Gross margin ($/ha) 399.56 178.34 245.37 Returns to labor ($/day) 12.67 2.86 4.62 163 Table 73: Wet season rice farm budget by province Province Kampong Kandal Takeo Kampot Battambang Bantey Cham Meanchey Average cultivated area 4.17 2.00 1.12 0.88 5 3 (ha) Yield (tons/ha) 3.12 3.23 3.08 2.54 3.30 2.65 Inputs ($/ha) Seeds 16.25 35.83 28.27 12.53 74.75 30.32 Manure 1.67 7.06 33.62 26.99 23.13 - Chemical fertilizers 13.47 73.96 91.84 55.21 64.56 60.18 Herbicides 13.58 6.94 1.00 0.16 8.91 6.39 Insecticides 0.77 1.33 2.56 - 0.42 0.46 Subtotal inputs 45.73 125.12 157.29 94.90 171.76 97.35 Labor ($/ha) Land preparation - - - - - - Plantation, transplantation 129.17 115.42 118.40 111.27 5.00 6.58 Weed control - 13.53 27.29 31.10 3.13 4.72 Crop management 114.17 30.28 80.47 108.45 4.69 70.67 Harvest - - - 17.68 - 6.13 Post-harvest - - 11.91 11.14 9.42 18.54 Subtotal labor 243.33 159.22 238.07 279.64 22.23 106.64 Services ($/ha) Land preparation 150.00 174.17 98.98 94.25 127.25 86.00 Plantation, transplantation - - - 4.77 - - Harvest - 100.00 23.86 8.86 45.50 34.25 Post-harvest 72.50 44.79 45.48 21.67 - 13.02 Subtotal services 222.50 318.96 168.32 129.55 172.75 133.27 Total variable costs ($/ha) 511.57 603.30 563.68 504.09 366.74 337.26 Total revenue ($/ha) 662.50 991.67 895.75 654.66 734.25 482.40 Gross margin ($/ha) 150.93 388.37 332.07 150.57 367.51 145.14 Returns to labor ($/day) 2.57 10.12 5.79 2.23 68.61 5.65 164 Table 74: Dry season rice farm budget by technology and farm size Level of Technology Use and Farm Size Modern Traditional Modern Traditional Modern Traditional Small Small Medium Medium Large Large Average cultivated area 1.00 0.67 2.67 2.04 9.13 3.50 (ha) Yield (tons/ha) 5.00 4.64 4.50 4.46 5.19 5.14 Inputs ($/ha) Seeds 115.00 107.20 134.29 82.43 119.91 61.71 Manure - 16.67 - - - - Chemical fertilizers 228.75 117.38 168.13 102.96 203.59 100.71 Herbicides 20.00 4.91 19.00 2.33 21.18 5.71 Insecticides 7.50 11.14 7.44 13.61 10.99 19.64 Subtotal inputs 371.25 257.30 328.85 201.33 355.66 187.79 Labor ($/ha) Land preparation - 62.50 - 36.50 - 92.50 Plantation, transplantation 12.50 16.55 9.58 14.15 20.97 14.29 Weed control - 29.84 - 19.00 - 30.00 Crop management - 27.07 - 24.20 - 38.57 Irrigation - - - - - - Harvest - 7.32 - 6.16 - 12.86 Post-harvest - 93.33 - 119.00 - 125.00 Subtotal labor 12.50 236.61 9.58 219.01 20.97 313.21 Services ($/ha) Land preparation 132.50 29.17 129.17 55.00 114.84 - Plantation, transplantation - - - - - - Harvest 100.00 98.33 105.83 96.50 105.16 112.50 Post-harvest 20.00 33.19 19.42 35.29 20.56 - Subtotal services 252.50 160.69 254.42 186.79 240.56 112.50 Cost of irrigation services 150.00 16.67 150.00 - 136.56 - Total variable costs ($/ha) 786.25 671.27 742.85 607.14 753.76 613.50 Total revenue ($/ha) 1,062.50 983.19 916.67 956.42 1,036.56 1,028.57 Gross margin ($/ha) 276.25 311.92 173.81 349.28 282.81 415.07 Returns to labor ($/day) 90.61 5.40 74.36 6.54 55.30 5.43 165 Table 75: Dry season rice farm budget by farm size Size Small Medium Large Total Average cultivated area (ha) 0.71 2.28 8.50 4.15 Yield (tons/ha) 4.70 4.48 5.18 4.80 Inputs ($/ha) Seeds 108.31 101.88 113.44 108.09 Manure 14.29 - - 4.17 Chemical fertilizers 133.29 127.40 192.16 153.40 Herbicides 7.07 8.58 19.46 12.22 Insecticides 10.62 11.29 11.95 11.34 Subtotal inputs 273.58 249.15 337.01 289.22 Labor ($/ha) Land preparation 53.57 22.81 10.28 27.08 Plantation, transplantation 15.97 12.44 20.22 16.39 Weed control 25.58 11.88 3.33 12.67 Crop management 23.21 15.13 4.29 13.42 Irrigation - - - - Harvest 6.27 3.85 1.43 3.65 Post-harvest 80.00 74.38 13.89 53.33 Subtotal labor 204.60 140.48 53.44 126.54 Services ($/ha) Land preparation 43.93 82.81 102.08 78.70 Plantation, transplantation - - - - Harvest 98.57 100.00 105.97 101.82 Post-harvest 31.31 29.34 18.28 25.77 Subtotal services 173.81 212.15 226.33 206.29 Cost of irrigation services 35.71 56.25 121.39 74.69 Total variable costs ($/ha) 687.70 658.03 738.17 696.74 Total revenue ($/ha) 994.52 941.51 1,035.68 992.28 Gross margin ($/ha) 306.83 283.48 297.50 295.55 Returns to labor ($/day) 6.15 8.27 22.83 9.58 166 Table 76: Dry season rice farm budget by technology use Level of Technology Use Modern Traditional Total Average cultivated area (ha) 6.83 1.48 4.15 Yield (tons/ha) 5.00 4.61 4.80 Inputs ($/ha) Seeds 123.09 93.09 108.09 Manure - 8.33 4.17 Chemical fertilizers 196.82 109.98 153.40 Herbicides 20.54 3.90 12.22 Insecticides 9.81 12.88 11.34 Subtotal inputs 350.26 228.19 289.22 Labor ($/ha) Land preparation - 54.17 27.08 Plantation, transplantation 17.41 15.36 16.39 Weed control - 25.34 12.67 Crop management - 26.83 13.42 Irrigation - - - Harvest - 7.30 3.65 Post-harvest - 106.67 53.33 Subtotal labor 17.41 235.66 126.54 Services ($/ha) Land preparation 119.90 37.50 78.70 Plantation, transplantation - - - Harvest 104.90 98.75 101.82 Post-harvest 20.23 31.30 25.77 Subtotal services 245.02 167.55 206.29 Cost of irrigation services 141.04 8.33 74.69 Total variable costs ($/ha) 753.74 639.73 696.74 Total revenue ($/ha) 1,008.75 975.82 992.28 Gross margin ($/ha) 255.01 336.08 295.55 Returns to labor ($/day) 60.04 5.85 9.58 167 Table 77: Dry season rice farm budget by province Province Bantey Meanchey Battambang Kandal Takeo Total Average cultivated area (ha) 7.83 5.83 1.25 1.70 4.154 Yield (tons/ha) 4.92 5.08 4.31 4.91 4.81 Inputs ($/ha) Seeds 144.02 102.17 111.47 74.71 108.09 Manure - - 16.67 - 4.17 Chemical fertilizers 207.68 185.96 96.53 123.43 153.40 Herbicides 16.71 24.37 - 7.81 12.22 Insecticides 8.98 10.64 10.23 15.53 11.34 Subtotal inputs 377.39 323.13 234.90 221.47 289.22 Labor ($/ha) Land preparation - - 14.58 93.75 27.08 Plantation, transplantation 9.54 25.29 18.46 12.26 16.39 Weed control - - 17.22 33.45 12.67 Crop management - - 15.93 37.74 13.42 Irrigation - - - - - Harvest - - - 14.59 3.65 Post-harvest - - 97.92 115.42 53.33 Subtotal labor 9.54 25.29 164.11 307.21 126.54 Services ($/ha) Land preparation 130.83 108.96 75.00 - 78.70 Plantation, transplantation - - - - - Harvest 106.04 103.75 110.42 87.08 101.82 Post-harvest 20.48 19.98 42.19 20.42 25.77 Subtotal services 257.35 232.69 227.60 107.50 206.29 Cost of irrigation services 150.00 132.08 16.67 - 74.69 Total variable costs ($/ha) 794.28 713.19 643.28 636.18 696.74 Total revenue ($/ha) 970.21 1,047.29 931.60 1,020.04 992.28 Gross margin ($/ha) 175.93 334.10 288.31 383.86 295.55 Returns to labor ($/day) 75.65 54.15 7.20 5.12 9.58 168 Table 78: Cassava farm budget by technology and farm size Level of Technology Use and Farm Size Modern Traditional Modern Traditional Modern Traditional Small Small Medium Medium Large Large Average cultivated area 3.25 2.50 6.25 6.70 19.50 15.38 (ha) Yield (tons/ha) 21.11 19.32 27.85 14.16 19.32 16.27 Inputs ($/ha) Seeds 200.63 244.51 317.50 153.17 250.31 235.00 Manure - - - - - - Chemical fertilizers 18.75 10.49 - - 16.34 - Herbicides 75.35 20.27 60.10 17.94 62.61 21.56 Insecticides - 5.93 - 3.53 - 2.81 Subtotal inputs 294.72 281.20 377.60 174.63 329.26 259.37 Labor ($/ha) Land preparation - - - - - - Plantation, transplantation 12.50 42.99 43.75 35.08 45.49 36.47 Weed control 45.00 37.96 11.79 38.28 45.57 30.39 Crop management 69.44 62.93 5.96 62.68 91.53 83.72 Harvest - 19.44 146.63 4.25 - - Post-harvest 106.50 42.47 172.88 80.36 25.00 42.19 Subtotal labor 233.44 205.79 381.03 220.66 207.59 192.77 Services ($/ha) Land preparation 87.50 110.53 118.75 166.16 75.94 136.63 Plantation, transplantation 21.25 10.00 - 10.00 10.63 9.38 Harvest 31.25 51.50 - 41.52 22.34 12.81 Post-harvest 77.78 181.14 - 52.45 229.08 122.25 Subtotal services 217.78 353.17 118.75 270.13 337.99 281.06 Total variable costs ($/ha) 745.94 840.17 877.37 665.42 874.83 733.20 Total revenue ($/ha) 1,519.44 1,333.33 1,665.06 963.15 1,483.81 1,148.67 Gross margin ($/ha) 773.50 493.17 787.69 297.73 608.97 415.48 Returns to labor ($/day) 13.75 9.95 8.58 5.60 12.17 8.94 169 Table 79: Cassava farm budget by farm size Size Small Medium Large Total Average cultivated area (ha) 2.64 6.57 17.44 8.25 Yield (tons/ha) 19.65 18.07 17.79 18.65 Inputs ($/ha) Seeds 236.53 200.12 242.66 228.61 Manure - - - - Chemical fertilizers 11.99 - 8.17 7.59 Herbicides 30.29 29.99 42.09 33.84 Insecticides 4.85 2.52 1.41 3.16 Subtotal inputs 283.66 232.62 294.32 273.20 Labor ($/ha) Land preparation - - - - Plantation, transplantation 37.45 37.56 40.98 38.56 Weed control 39.24 30.72 37.98 36.56 Crop management 64.11 46.48 87.63 66.60 Harvest 15.91 44.93 - 18.83 Post-harvest 54.11 106.80 33.59 61.98 Subtotal labor 210.82 266.48 200.18 222.53 Services ($/ha) Land preparation 106.35 152.61 106.28 118.78 Plantation, transplantation 12.05 7.14 10.00 10.10 Harvest 47.82 29.66 17.58 33.62 Post-harvest 162.35 37.46 175.66 132.82 Subtotal services 328.56 226.88 309.52 295.32 Total variable costs ($/ha) 823.03 725.98 804.01 791.05 Total revenue ($/ha) 1,367.17 1,163.69 1,316.24 1,296.72 Gross margin ($/ha) 544.14 437.72 512.23 505.67 Returns to labor ($/day) 10.71 6.82 10.62 9.43 170 Table 80: Cassava farm budget by technology use Level of Technology Use Modern Traditional Total Average cultivated area (ha) 12.13 6.53 8.25 Yield (tons/ha) 21.90 17.21 18.65 Inputs ($/ha) Seeds 254.69 217.03 228.61 Manure - - - Chemical fertilizers 12.86 5.24 7.59 Herbicides 65.17 19.91 33.84 Insecticides - 4.57 3.16 Subtotal inputs 332.71 246.75 273.20 Labor ($/ha) Land preparation - - - Plantation, transplantation 36.81 39.34 38.56 Weed control 36.98 36.37 36.56 Crop management 64.62 67.48 66.60 Harvest 36.66 10.90 18.83 Post-harvest 82.35 52.93 61.98 Subtotal labor 257.41 207.03 222.53 Services ($/ha) Land preparation 89.53 131.78 118.78 Plantation, transplantation 10.63 9.86 10.10 Harvest 18.98 40.13 33.62 Post-harvest 133.98 132.30 132.82 Subtotal services 253.12 314.08 295.32 Total variable costs ($/ha) 843.25 767.85 791.05 Total revenue ($/ha) 1,538.03 1,189.47 1,296.72 Gross margin ($/ha) 694.79 421.61 505.67 Returns to labor ($/day) 11.20 8.45 9.43 171 Table 81: Cassava farm budget by province Province Bantey Meanchey Battambang Kampong Cham Total Average cultivated area (ha) 10.67 8.42 6.19 8.25 Yield (tons/ha) 20.30 20.44 17.75 18.65 Inputs ($/ha) Seeds 225.45 228.85 230.63 228.61 Manure - - - - Chemical fertilizers - 14.11 3.48 7.59 Herbicides 23.57 33.30 42.34 33.84 Insecticides 2.44 5.10 0.79 3.16 Subtotal inputs 251.46 281.37 277.24 273.20 Labor ($/ha) Land preparation - - - - Plantation, transplantation 43.13 27.30 52.02 38.56 Weed control 19.72 17.22 78.18 36.56 Crop management 56.67 48.02 101.92 66.60 Harvest 3.54 39.02 - 18.83 Post-harvest 64.15 102.22 - 61.98 Subtotal labor 187.21 233.79 232.13 222.53 Services ($/ha) Land preparation 144.48 143.98 61.72 118.78 Plantation, transplantation - 17.71 6.25 10.10 Harvest 44.45 40.73 14.84 33.62 Post-harvest 163.33 60.04 219.11 132.82 Subtotal services 352.26 262.46 301.92 295.32 Total variable costs ($/ha) 790.93 777.63 811.28 791.05 Total revenue ($/ha) 1,219.56 1,352.28 1,271.24 1,296.72 Gross margin ($/ha) 428.64 574.65 459.96 505.67 Returns to labor ($/day) 9.50 10.20 8.22 9.43 172 Table 82: Maize farm budget by technology and farm size Level of Technology Use and Farm Size Modern Traditional Modern Traditional Modern Traditional Small Small Medium Medium Large Large Average cultivated area 0.67 1.00 2.10 2.54 9.75 11.40 (ha) Yield (tons/ha) 3.98 2.91 4.15 3.38 5.70 3.77 Inputs ($/ha) Seeds 79.17 50.58 88.28 68.20 87.75 76.41 Manure 12.17 - 12.25 0.03 9.66 - Chemical fertilizers 93.40 10.00 16.76 - 32.94 - Herbicides 29.88 10.78 16.78 16.37 26.41 14.93 Insecticides 5.12 - 8.06 0.38 8.44 2.16 Subtotal inputs 219.75 71.36 142.13 84.98 165.19 93.49 Labor ($/ha) Land preparation - - - - - - Plantation, transplantation 40.43 34.12 29.90 26.73 13.35 19.98 Weed control 26.35 22.09 16.75 12.92 11.56 15.71 Crop management 71.40 66.24 56.77 65.08 28.21 69.01 Harvest 41.91 6.25 68.82 19.21 3.75 29.81 Post-harvest 4.86 - 0.83 - - - Subtotal labor 184.96 128.70 173.07 123.93 56.88 134.51 Services ($/ha) Land preparation 150.90 65.83 112.08 132.05 60.03 94.56 Plantation, transplantation 4.00 - 8.94 6.43 12.82 14.66 Harvest 4.00 - - 9.46 37.50 - Post-harvest 52.69 24.23 13.03 35.42 38.61 26.51 Subtotal services 211.59 90.06 134.06 183.36 148.96 135.73 Total variable costs ($/ha) 616.30 290.11 449.26 392.27 371.03 363.73 Total revenue ($/ha) 897.77 408.74 953.31 584.07 893.13 616.83 Gross margin ($/ha) 281.47 118.62 504.06 191.80 522.10 253.10 Returns to labor ($/day) 6.32 3.83 12.09 6.42 38.10 7.81 173 Table 83: Maize farm budget by farm size Size Small Medium Large Total Average cultivated area (ha) 0.75 2.34 10.93 4.83 Yield (tons/ha) 3.73 3.73 4.32 3.94 Inputs ($/ha) Seeds 72.57 77.47 79.65 76.64 Manure 9.36 5.67 2.76 5.85 Chemical fertilizers 74.16 7.74 9.41 29.91 Herbicides 25.48 16.56 18.21 20.03 Insecticides 3.94 3.93 3.95 3.94 Subtotal inputs 185.50 111.36 113.98 136.37 Labor ($/ha) Land preparation - - - - Plantation, transplantation 38.97 28.19 18.09 28.16 Weed control 25.37 14.68 14.52 18.10 Crop management 70.21 61.25 57.36 62.80 Harvest 33.68 42.11 22.36 32.46 Post-harvest 3.74 0.38 - 1.34 Subtotal labor 171.98 146.61 112.33 142.85 Services ($/ha) Land preparation 131.27 122.83 84.70 112.23 Plantation, transplantation 3.08 7.59 14.13 8.41 Harvest 3.08 5.10 10.71 6.41 Post-harvest 46.12 25.09 29.97 33.63 Subtotal services 183.54 160.60 139.51 160.68 Total variable costs ($/ha) 541.03 418.57 365.82 439.90 Total revenue ($/ha) 784.91 754.49 695.77 743.83 Gross margin ($/ha) 243.89 335.92 329.95 303.92 Returns to labor ($/day) 5.89 9.51 12.19 8.83 174 Table 84: Maize farm budget by technology use Level of Technology Use Modern Traditional Total Average cultivated area (ha) 2.91 6.74 4.83 Yield (tons/ha) 4.38 3.50 3.94 Inputs ($/ha) Seeds 83.62 69.66 76.64 Manure 11.69 0.01 5.85 Chemical fertilizers 58.32 1.50 29.91 Herbicides 25.26 14.81 20.03 Insecticides 6.67 1.21 3.94 Subtotal inputs 185.55 87.19 136.37 Labor ($/ha) Land preparation - - - Plantation, transplantation 31.85 24.46 28.16 Weed control 20.51 15.69 18.10 Crop management 58.37 67.22 62.80 Harvest 42.35 22.56 32.46 Post-harvest 2.68 - 1.34 Subtotal labor 155.78 129.93 142.85 Services ($/ha) Land preparation 121.08 103.37 112.23 Plantation, transplantation 7.25 9.58 8.41 Harvest 9.50 3.31 6.41 Post-harvest 37.98 29.28 33.63 Subtotal services 175.80 145.55 160.68 Total variable costs ($/ha) 517.13 362.68 439.90 Total revenue ($/ha) 913.50 574.15 743.83 Gross margin ($/ha) 396.37 211.47 303.92 Returns to labor ($/day) 10.56 6.75 8.83 175 Table 85: Maize farm budget by province Province Bantey Battambang Kampong Kandal Total Meanchey Cham Average cultivated area 11.08 2.10 3.25 1.12 4.83 (ha) Yield (tons/ha) 5.31 2.72 2.56 4.35 3.94 Inputs ($/ha) Seeds 91.05 67.08 57.03 84.18 76.64 Manure 1.14 5.00 - 16.29 5.85 Chemical fertilizers 5.48 46.53 - 74.83 29.91 Herbicides 15.44 19.53 18.25 26.48 20.03 Insecticides 7.18 2.94 0.99 3.81 3.94 Subtotal inputs 120.29 141.07 76.28 205.58 136.37 Labor ($/ha) Land preparation - - - - 0 Plantation, transplantation 3.51 36.85 37.73 40.41 28.16 Weed control 7.50 14.50 21.47 27.11 18.10 Crop management 63.97 45.88 61.04 70.29 62.80 Harvest 41.83 25.75 5.86 50.26 32.46 Post-harvest - 1.00 - 4.05 1.34 Subtotal labor 116.82 123.98 126.10 192.12 142.85 Services ($/ha) Land preparation 88.72 75.50 83.59 177.29 112.23 Plantation, transplantation 24.71 - - 3.33 8.41 Harvest 12.50 - 6.02 3.33 6.41 Post-harvest 30.48 - 31.92 52.37 33.63 Subtotal services 156.40 75.50 121.53 236.33 160.68 Total variable costs ($/ha) 393.51 340.54 323.91 634.03 439.90 Total revenue ($/ha) 838.98 401.50 370.36 1,133.65 743.83 Gross margin ($/ha) 445.47 60.96 46.46 499.62 303.92 Returns to labor ($/day) 15.83 2.04 1.53 10.79 8.83 176 Table 86: Vegetable (mix) farm budget by technology and farm size Level of Technology Use and Farm Size Modern Traditional Modern Traditional Modern Traditional Small Small Medium Medium Large Large Average cultivated area 0.12 0.30 0.84 3.00 2.17 (ha) Yield (tons/ha) 12.32 6.00 6.00 1.67 8.33 Inputs ($/ha) Seeds 238.31 36.15 28.07 243.00 127.97 Manure 260.99 57.28 10.12 20.83 24.57 Chemical fertilizers 315.42 86.92 52.21 108.75 69.58 Herbicides 31.57 1.64 3.65 - 5.56 Insecticides 142.97 27.03 18.01 78.75 16.39 Subtotal inputs 989.26 209.02 112.07 451.33 244.06 Labor ($/ha) Land preparation 229.04 100.28 45.93 207.50 164.58 Plantation, transplantation 151.47 67.70 40.78 50.00 44.44 Weed control 151.29 49.47 55.11 50.00 21.11 Crop management 541.73 219.71 68.65 33.33 46.53 Harvest 158.57 31.32 30.42 - 55.56 Post-harvest - - - - - Subtotal labor 1,232.11 468.49 240.89 340.83 332.22 Services ($/ha) Land preparation 38.45 41.07 29.42 - 175.56 Plantation, transplantation - 0.76 5.33 - - Harvest - 4.55 17.06 - 6.67 Post-harvest 32.00 1.50 10.98 - - Subtotal services 70.45 47.88 62.79 - 182.22 Total variable costs ($/ha) 2,291.82 725.39 415.74 792.17 758.51 Total revenue ($/ha) 4,530.73 1,336.84 1,163.16 500.00 1,263.89 Gross margin ($/ha) 2,238.92 611.45 747.41 (292.17) 505.38 Returns to labor ($/day) 7.54 5.42 12.88 (3.56) 6.31 177 Table 87: Vegetable (mix) farm budget by farm size Size Small Medium Large Total Average cultivated area (ha) 0.19 0.84 2.38 0.41 Yield (tons/ha) 9.74 6.00 6.67 9.10 Inputs ($/ha) Seeds 155.80 28.07 156.73 140.96 Manure 177.85 10.12 23.63 148.00 Chemical fertilizers 222.15 52.21 79.38 192.81 Herbicides 19.36 3.65 4.17 16.51 Insecticides 95.65 18.01 31.98 82.34 Subtotal inputs 670.80 112.07 295.88 580.62 Labor ($/ha) Land preparation 176.49 45.93 175.31 161.18 Plantation, transplantation 117.28 40.78 45.83 103.59 Weed control 109.73 55.11 28.33 97.93 Crop management 410.29 68.65 43.23 345.96 Harvest 106.63 30.42 41.67 93.41 Post-harvest - - - - Subtotal labor 920.43 240.89 334.38 802.08 Services ($/ha) Land preparation 39.52 29.42 131.67 44.48 Plantation, transplantation 0.31 5.33 - 0.87 Harvest 1.86 17.06 5.00 3.84 Post-harvest 19.55 10.98 - 17.25 Subtotal services 61.24 62.79 136.67 66.45 Total variable costs ($/ha) 1,652.46 415.74 766.92 1,449.14 Total revenue ($/ha) 3,227.10 1,163.16 1,072.92 2,842.70 Gross margin ($/ha) 1,574.64 747.41 305.99 1,393.56 Returns to labor ($/day) 7.10 12.88 3.80 7.21 178 Table 88: Vegetable (mix) farm budget by technology use Level of Technology Use Modern Traditional Total Average cultivated area (ha) 0.21 0.61 0.41 Yield (tons/ha) 11.96 6.24 9.10 Inputs ($/ha) Seeds 238.47 43.45 140.96 Manure 252.99 43.01 148.00 Chemical fertilizers 308.53 77.09 192.81 Herbicides 30.52 2.50 16.51 Insecticides 140.83 23.86 82.34 Subtotal inputs 971.33 189.90 580.62 Labor ($/ha) Land preparation 228.33 94.03 161.18 Plantation, transplantation 148.09 59.09 103.59 Weed control 147.91 47.95 97.93 Crop management 524.78 167.14 345.96 Harvest 153.29 33.54 93.41 Post-harvest - - - Subtotal labor 1,202.40 401.76 802.08 Services ($/ha) Land preparation 37.17 51.80 44.48 Plantation, transplantation - 1.75 0.87 Harvest - 7.68 3.84 Post-harvest 30.93 3.56 17.25 Subtotal services 68.10 64.79 66.45 Total variable costs ($/ha) 2,241.83 656.45 1,449.14 Total revenue ($/ha) 4,396.37 1,289.02 2,842.70 Gross margin ($/ha) 2,154.55 632.57 1,393.56 Returns to labor ($/day) 7.44 6.53 7.21 179 Table 89: Vegetable (mix) farm budget by province Province Kampong Kandal Takeo Kampot Battambang Bantey Cham Meanchey Average cultivated area (ha) 0.33 0.18 0.20 0.22 1.833 0.511 Yield (tons/ha) 6.77 9.78 10.67 10.15 6.11 5.89 Inputs ($/ha) Seeds 125.50 460.13 153.24 72.87 114.94 30.68 Manure 49.90 234.18 262.81 123.95 20.34 15.25 Chemical fertilizers 176.66 320.41 211.27 217.95 67.42 75.92 Herbicides 70.41 32.06 5.90 13.10 2.78 2.87 Insecticides 25.85 87.70 165.05 43.34 29.65 55.05 Subtotal inputs 448.32 1,134.47 798.28 471.22 235.13 179.76 Labor ($/ha) Land preparation 63.65 229.17 179.07 210.66 129.38 20.40 Plantation, transplantation 161.98 283.40 91.07 85.07 43.06 19.07 Weed control 51.55 227.06 104.23 109.35 18.89 41.09 Crop management 219.80 637.71 313.30 485.26 66.32 140.09 Harvest 40.19 105.56 32.83 200.83 40.28 47.10 Post-harvest - - - - - - Subtotal labor 537.17 1,482.89 720.51 1,091.18 297.92 267.75 Services ($/ha) Land preparation 57.29 78.33 17.78 24.31 96.22 86.76 Plantation, transplantation - - - - - 8.74 Harvest - - - - 3.33 35.05 Post-harvest 7.81 125.69 11.32 1.67 - - Subtotal services 65.10 204.03 29.10 25.97 99.55 130.55 Total variable costs ($/ha) 1,050.59 2,821.39 1,547.88 1,588.37 632.59 578.06 Total revenue ($/ha) 1,878.19 7,080.85 3,579.70 2,169.37 1,090.28 1,130.44 Gross margin ($/ha) 827.60 4,259.46 2,031.81 581.00 457.68 552.37 Returns to labor ($/day) 6.39 11.92 11.70 2.21 6.38 8.56 Detailed farm budgets by crop 180 Table 90: Changes in prices of commodities, Thailand and Cambodia Thailand 2005 2011 % Change from (Farm gate price $/ton) ($/ton) ($/ton) 2005 to 2011 Cassava 33.00 87.90 166 Rice 172.20 419.80 144 Maize 118.90 250.20 110 Eggplant 211.00 413.20 96 Cucumber 160.40 367.00 129 Watermelon 138.40 243.60 76 Cambodia 2005 2013 % Change from (Farm gate price $/ton) ($/ton) ($/ton) 2005 to 2013 Cassava 19.79 59.38 200 Paddy (wet season) 124.79 237.50 90 Paddy (dry season) 118.50 209.98 77 Maize 140.43 230.39 64 Mustard N/A 297.50 Eggplant 125.00 200.00 60 Cucumber 86.16 187.50 118 Watermelon 83.44 277.50 233 Source: Thailand prices from FAOSTAT; Cambodia prices from 2005 and 2013 surveys. 181 Table 91: Change in the prices of inputs, services, and labor, Cambodia % Change from Operations and Inputs 2005 2013 2005 to 2013 Services Land preparation (tractor $/ha) 38.21 61.48 61 Land preparation (animal $/ha) 26.96 76.58 184 Combine harvester ($/ha) 95.67 Fertilizers Urea ($/kg) 0.30 0.56 86 DAP ($/kg) 0.35 0.68 96 Seeds Paddy (wet season $/kg) 0.17 0.66 293 Paddy (dry season $/kg) 0.18 0.74 323 Cassava ($/kg) 0.02 0.10 542 Maize ($/kg) 0.61 4.06 563 Eggplant ($/kg) 4.75 11.60 144 Cucumber ($/kg) 8.52 18.00 111 Watermelon ($/kg) 10.63 53.91 407 Labor Dry season ($/day) 1.20 4.50 275 Wet season ($/day) 1.25 4.56 265 Unskilled labor (garment $/month) 50.00 159.00 218 Source: 2005 and 2013 surveys. Table 92: Average use of inputs, Cambodia, 2013 Seeds Urea DAP Herbicides Insecticides (kg/ha) (kg/ha) (kg/ha) ($/ha) ($/ha) Cassava 9.5 0.8 33.8 3.16 Maize 22.58 27.91 16.36 20.03 3.94 Wet season rice 60.07 49.49 58.34 3.37 1.11 Dry season rice 111.28 139.12 12.22 11.34 Vegetables 158.2 137.25 20.65 49.28 Source: 2013 survey. 182 Table 93: Farm budget indicators by crop, farm size, and technology use, Cambodia, 2013 Level of Farm Size Technology Use Labor Use ($/ha) Small Medium Large Modern Traditional Overall Cassava 211 284 200 257 214 227 Maize 212 183 131 200 147 174 Wet season rice 288 232 121 131 259 220 Dry season rice 125 66 40 17 129 73 Vegetables 921 241 334 1,190 415 802 Services ($/ha) Cassava 329 209 310 253 307 290 Maize 144 125 121 129 130 130 Wet season rice 135 160 234 233 144 171 Dry season rice 174 212 226 245 168 206 Vegetables 61 63 137 77 56 66 Returns to Labor ($/day) Cassava 10.71 6.82 10.62 11.20 8.45 9.43 Maize 5.89 9.51 12.19 10.56 6.75 8.83 Wet season rice 2.99 4.83 8.20 12.67 2.86 4.62 Dry season rice 6.15 8.27 22.83 60.04 5.85 9.58 Vegetables 7.10 12.88 3.80 7.44 6.53 7.21 Source: 2013 survey. 183 ANNEX 4: PROJECTION OF SELECTED ECONOMIC INDICATORS, CAMBODIA Indicator Unit Future 2012 2015 2020 2030 Comments Growth (Baseline) (Short (Medium (Long Rate term) term) term) 1 GDP $ billion 7.2% 14.17 17.46 24.71 49.53 Baseline data is from the World Bank 2014c. The assumption of future growth is from the Cambodia Vision 2030. 2 Ag GDP $ billion 5.0% 4.75 5.50 7.00 11.40 Baseline data is from the World Bank 2014c. Future growth is assumed to be slightly below the past rate (2004-2012). 3 Share of % -2.1% 34 31 28 23 agriculture in Indicator #2 divided by Indicator #1 GDP 4 Agribusiness $ billion 9.7% 1.30 1.72 2.73 6.89 Baseline data is from the World Bank 2014c. The assumption GDP of future growth is based on the target to double the ratio of agribusiness GDP to agricultural GDP. 5 Population Million 1.1% 14.4 14.9 15.7 17.5 Baseline data is from the NIS Social Economic Surveys. The assumption of future growth is from the Cambodia Vision 2030. 6 GDP per capita $ 6.1% 986 1,180 1,600 2,900 Indicator #1 divided by Indicator #5 7 Urban Million 5.1% 3.1 3.6 4.7 7.7 Baseline data is from the NIS Social Economic Surveys. The population assumption of future growth is from the UN population projections. 8 Rural population Million -0.8% 11.2 11.2 11.0 9.8 Baseline data is from the NIS Social Economic Surveys. The assumption of future growth is from the UN population projections. 9 Urbanization % 4.0% 28 31 38 56 The assumption of future growth is from the Cambodia Vision 2030. 10 Total labor force Million 1.26% 7.7 8.0 8.5 9.7 Baseline data is from the NIS Social Economic Surveys. The assumption of future growth is from the UN population projections. 11 Agricultural Million -1.4% 3.9 3.8 3.5 3.0 Baseline data is from the NIS Social Economic Surveys. The labor force assumption of future growth is from the UN population projections. 184 12 Labor share in % -2.7% 51 47 41 31 Indicator #11 divided by Indicator #10 agriculture 13 GDP/labor $/capita 5.9% 1,840 2,190 2,920 5,200 Indicator #1 divided by Indicator #10 14 Ag GDP/Ag $/capita 6.4% 1,210 1,460 2,000 3,700 Indicator #2 divided by Indicator #11 labor 15 Non-Ag $/capita 4.2% 2,500 2,800 3,500 5,780 (Indicator #1-Indicator #2)/ (Indicator #10-Indicator #11) GDP/Non-Ag divided by Indicator #2/Indicator #11. labor 185 ANNEX 5: RESULTS OF POLICY SIMULATIONS Table 94: Results of simulations on land use (baseline) Baseline 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 Cultivated Areas (Ha) 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 Cassava 337,800 337,800 337,800 337,800 337,800 337,800 337,800 337,800 337,800 337,800 337,800 337,800 337,800 337,800 337,800 337,800 337,800 337,800 Maize 215,442 215,442 215,442 215,442 215,442 215,442 215,442 215,442 215,442 215,442 215,442 215,442 215,442 215,442 215,442 215,442 215,442 215,442 Wet Season Rice 2,484,832 2,484,832 2,484,832 2,484,832 2,484,832 2,484,832 2,484,832 2,484,832 2,484,832 2,484,832 2,484,832 2,484,832 2,484,832 2,484,832 2,484,832 2,484,832 2,484,832 2,484,832 Dry Season Rice 495,465 495,465 495,465 495,465 495,465 495,465 495,465 495,465 495,465 495,465 495,465 495,465 495,465 495,465 495,465 495,465 495,465 495,465 Vegetables 54,155 54,155 54,155 54,155 54,155 54,155 54,155 54,155 54,155 54,155 54,155 54,155 54,155 54,155 54,155 54,155 54,155 54,155 Total 5 products 3,587,694 3,587,694 3,587,694 3,587,694 3,587,694 3,587,694 3,587,694 3,587,694 3,587,694 3,587,694 3,587,694 3,587,694 3,587,694 3,587,694 3,587,694 3,587,694 3,587,694 3,587,694 - 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% Total Rice 2,980,297 2,980,297 2,980,297 2,980,297 2,980,297 2,980,297 2,980,297 2,980,297 2,980,297 2,980,297 2,980,297 2,980,297 2,980,297 2,980,297 2,980,297 2,980,297 2,980,297 2,980,297 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% PRODUCTION (tons) 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 Cassava 6,299,970 6,299,970 6,299,970 6,299,970 6,299,970 6,299,970 6,299,970 6,299,970 6,299,970 6,299,970 6,299,970 6,299,970 6,299,970 6,299,970 6,299,970 6,299,970 6,299,970 6,299,970 Maize 848,841 848,841 848,841 848,841 848,841 848,841 848,841 848,841 848,841 848,841 848,841 848,841 848,841 848,841 848,841 848,841 848,841 848,841 Wet Season Rice 7,181,164 7,181,164 7,181,164 7,181,164 7,181,164 7,181,164 7,181,164 7,181,164 7,181,164 7,181,164 7,181,164 7,181,164 7,181,164 7,181,164 7,181,164 7,181,164 7,181,164 7,181,164 Dry Season Rice 2,378,232 2,378,232 2,378,232 2,378,232 2,378,232 2,378,232 2,378,232 2,378,232 2,378,232 2,378,232 2,378,232 2,378,232 2,378,232 2,378,232 2,378,232 2,378,232 2,378,232 2,378,232 Vegetables 492,811 492,811 492,811 492,811 492,811 492,811 492,811 492,811 492,811 492,811 492,811 492,811 492,811 492,811 492,811 492,811 492,811 492,811 Total Rice 9,559,396 9,559,396 9,559,396 9,559,396 9,559,396 9,559,396 9,559,396 9,559,396 9,559,396 9,559,396 9,559,396 9,559,396 9,559,396 9,559,396 9,559,396 9,559,396 9,559,396 9,559,396 - 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% GROSS MARGINS (Million $) 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 Cassava 170.82 170.82 170.82 170.82 170.82 170.82 170.82 170.82 170.82 170.82 170.82 170.82 170.82 170.82 170.82 170.82 170.82 170.82 Maize 65.48 65.48 65.48 65.48 65.48 65.48 65.48 65.48 65.48 65.48 65.48 65.48 65.48 65.48 65.48 65.48 65.48 65.48 Wet Season Rice 609.70 609.70 609.70 609.70 609.70 609.70 609.70 609.70 609.70 609.70 609.70 609.70 609.70 609.70 609.70 609.70 609.70 609.70 Dry Season Rice 146.43 146.43 146.43 146.43 146.43 146.43 146.43 146.43 146.43 146.43 146.43 146.43 146.43 146.43 146.43 146.43 146.43 146.43 Vegetables 75.47 75.47 75.47 75.47 75.47 75.47 75.47 75.47 75.47 75.47 75.47 75.47 75.47 75.47 75.47 75.47 75.47 75.47 Total 5 products 1,067.90 1,067.90 1,067.90 1,067.90 1,067.90 1,067.90 1,067.90 1,067.90 1,067.90 1,067.90 1,067.90 1,067.90 1,067.90 1,067.90 1,067.90 1,067.90 1,067.90 1,067.90 - 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% Total Rice 756.13 756.13 756.13 756.13 756.13 756.13 756.13 756.13 756.13 756.13 756.13 756.13 756.13 756.13 756.13 756.13 756.13 756.13 Share of Rice 0.71 0.71 0.71 0.71 0.71 0.71 0.71 0.71 0.71 0.71 0.71 0.71 0.71 0.71 0.71 0.71 0.71 0.71 186 Table 95: Results of simulations on land use (scenario L1) Scenario L1 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 Cultivated Areas (Ha) 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 Cassava 337,800 337,800 337,800 337,800 337,800 337,800 337,800 337,800 337,800 337,800 337,800 337,800 337,800 337,800 337,800 337,800 337,800 337,800 Maize 215,442 215,442 215,442 215,442 215,442 215,442 215,442 215,442 215,442 215,442 215,442 215,442 215,442 215,442 215,442 215,442 215,442 215,442 Wet Season Rice 2,484,832 2,500,983 2,517,240 2,533,602 2,550,070 2,566,646 2,583,329 2,600,121 2,617,021 2,634,032 2,651,153 2,668,386 2,685,730 2,703,187 2,720,758 2,738,443 2,756,243 2,774,159 Dry Season Rice 495,465 503,888 512,454 521,166 530,026 539,036 548,200 557,519 566,997 576,636 586,439 596,408 606,547 616,858 627,345 638,010 648,856 659,886 Vegetables 54,155 54,155 54,155 54,155 54,155 54,155 54,155 54,155 54,155 54,155 54,155 54,155 54,155 54,155 54,155 54,155 54,155 54,155 3,587,694 3,612,268 3,637,091 3,662,165 3,687,493 3,713,079 3,738,926 3,765,037 3,791,415 3,818,065 3,844,989 3,872,191 3,899,674 3,927,443 3,955,500 3,983,850 4,012,496 4,041,442 - 0.7% 1.4% 2.1% 2.8% 3.5% 4.2% 4.9% 5.7% 6.4% 7.2% 7.9% 8.7% 9.5% 10.3% 11.0% 11.8% 12.6% Total Rice 2,980,297 3,004,871 3,029,694 3,054,768 3,080,096 3,105,682 3,131,529 3,157,640 3,184,018 3,210,668 3,237,592 3,264,794 3,292,277 3,320,046 3,348,103 3,376,453 3,405,099 3,434,045 0.8% 1.7% 2.5% 3.3% 4.2% 5.1% 6.0% 6.8% 7.7% 8.6% 9.5% 10.5% 11.4% 12.3% 13.3% 14.3% 15.2% PRODUCTION (tons) 0 1 2 3 4 5 6 7 8 9 Cassava 6,299,970 6,299,970 6,299,970 6,299,970 6,299,970 6,299,970 6,299,970 6,299,970 6,299,970 6,299,970 6,299,970 6,299,970 6,299,970 6,299,970 6,299,970 6,299,970 6,299,970 6,299,970 Maize 848,841 848,841 848,841 848,841 848,841 848,841 848,841 848,841 848,841 848,841 848,841 848,841 848,841 848,841 848,841 848,841 848,841 848,841 Wet Season Rice 7,181,164 7,227,842 7,274,823 7,322,109 7,369,703 7,417,606 7,465,821 7,514,348 7,563,192 7,612,352 7,661,833 7,711,635 7,761,760 7,812,212 7,862,991 7,914,101 7,965,542 8,017,318 Dry Season Rice 2,378,232 2,418,662 2,459,779 2,501,595 2,544,123 2,587,373 2,631,358 2,676,091 2,721,585 2,767,852 2,814,905 2,862,758 2,911,425 2,960,920 3,011,255 3,062,447 3,114,508 3,167,455 Vegetables 492,811 492,811 492,811 492,811 492,811 492,811 492,811 492,811 492,811 492,811 492,811 492,811 492,811 492,811 492,811 492,811 492,811 492,811 Total Rice 9,559,396 9,646,504 9,734,602 9,823,705 9,913,826 10,004,979 10,097,179 10,190,439 10,284,776 10,380,204 10,476,738 10,574,393 10,673,186 10,773,131 10,874,246 10,976,547 11,080,050 11,184,773 - 0.9% 1.8% 2.8% 3.7% 4.7% 5.6% 6.6% 7.6% 8.6% 9.6% 10.6% 11.7% 12.7% 13.8% 14.8% 15.9% 17.0% GROSS MARGINS (Million $) 0 1 2 3 4 5 6 7 8 9 Cassava 170.82 170.82 170.82 170.82 170.82 170.82 170.82 170.82 170.82 170.82 170.82 170.82 170.82 170.82 170.82 170.82 170.82 170.82 Maize 65.48 65.48 65.48 65.48 65.48 65.48 65.48 65.48 65.48 65.48 65.48 65.48 65.48 65.48 65.48 65.48 65.48 65.48 Wet Season Rice 609.70 613.67 617.66 621.67 625.71 629.78 633.87 637.99 642.14 646.31 650.51 654.74 659.00 663.28 667.59 671.93 676.30 680.70 Dry Season Rice 146.43 148.92 151.45 154.03 156.64 159.31 162.01 164.77 167.57 170.42 173.32 176.26 179.26 182.31 185.41 188.56 191.76 195.02 Vegetables 75.47 75.47 75.47 75.47 75.47 75.47 75.47 75.47 75.47 75.47 75.47 75.47 75.47 75.47 75.47 75.47 75.47 75.47 Total 5 Products 1,067.90 1,074.35 1,080.87 1,087.46 1,094.12 1,100.85 1,107.65 1,114.52 1,121.47 1,128.49 1,135.59 1,142.77 1,150.02 1,157.35 1,164.76 1,172.25 1,179.82 1,187.48 - 0.6% 1.2% 1.8% 2.5% 3.1% 3.7% 4.4% 5.0% 5.7% 6.3% 7.0% 7.7% 8.4% 9.1% 9.8% 10.5% 11.2% Total Rice 756.13 762.59 769.11 775.70 782.35 789.08 795.89 802.76 809.71 816.73 823.83 831.00 838.26 845.59 853.00 860.49 868.06 875.72 - 0.9% 1.7% 2.6% 3.5% 4.4% 5.3% 6.2% 7.1% 8.0% 9.0% 9.9% 10.9% 11.8% 12.8% 13.8% 14.8% 15.8% Share of Rice 0.71 0.71 0.71 0.71 0.72 0.72 0.72 0.72 0.72 0.72 0.73 0.73 0.73 0.73 0.73 0.73 0.74 0.74 187 Table 96: Results of simulations on land use (scenario L2) Scenario L2 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 Cultivated Areas (Ha) 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 Cassava 337,800 348,981 360,532 372,466 384,795 397,531 410,690 424,284 438,327 452,836 467,825 483,310 499,307 515,834 532,909 550,548 568,771 587,597 Maize 215,442 217,532 219,642 221,772 223,924 226,096 228,289 230,503 232,739 234,997 237,276 239,578 241,902 244,248 246,617 249,009 251,425 253,864 Wet Season Rice 2,484,832 2,491,293 2,497,770 2,504,264 2,510,775 2,517,303 2,523,848 2,530,410 2,536,989 2,543,585 2,550,199 2,556,829 2,563,477 2,570,142 2,576,824 2,583,524 2,590,241 2,596,976 Dry Season Rice 495,465 498,834 502,226 505,641 509,080 512,541 516,027 519,536 523,069 526,625 530,207 533,812 537,442 541,096 544,776 548,480 552,210 555,965 Vegetables 54,155 54,577 55,003 55,432 55,865 56,300 56,739 57,182 57,628 58,077 58,530 58,987 59,447 59,911 60,378 60,849 61,324 61,802 Total 5 Products 3,587,694 3,611,217 3,635,174 3,659,576 3,684,438 3,709,772 3,735,593 3,761,915 3,788,752 3,816,121 3,844,037 3,872,516 3,901,575 3,931,232 3,961,504 3,992,411 4,023,971 4,056,204 - 0.7% 1.3% 2.0% 2.7% 3.4% 4.1% 4.9% 5.6% 6.4% 7.1% 7.9% 8.7% 9.6% 10.4% 11.3% 12.2% 13.1% Total Rice 2,980,297 2,990,127 2,999,996 3,009,905 3,019,855 3,029,845 3,039,875 3,049,946 3,060,058 3,070,211 3,080,405 3,090,641 3,100,919 3,111,239 3,121,600 3,132,005 3,142,451 3,152,941 0.3% 0.7% 1.0% 1.3% 1.7% 2.0% 2.3% 2.7% 3.0% 3.4% 3.7% 4.0% 4.4% 4.7% 5.1% 5.4% 5.8% PRODUCTION (tons) 0 1 2 3 4 5 6 7 8 9 Cassava 6,299,970 6,508,499 6,723,930 6,946,492 7,176,421 7,413,961 7,659,363 7,912,888 8,174,804 8,445,390 8,724,933 9,013,728 9,312,083 9,620,313 9,938,745 10,267,717 10,607,579 10,958,690 Maize 848,841 857,075 865,389 873,783 882,259 890,817 899,458 908,182 916,992 925,887 934,868 943,936 953,092 962,337 971,672 981,097 990,614 1,000,223 Wet Season Rice 7,181,164 7,199,836 7,218,555 7,237,323 7,256,140 7,275,006 7,293,921 7,312,886 7,331,899 7,350,962 7,370,074 7,389,237 7,408,449 7,427,711 7,447,023 7,466,385 7,485,798 7,505,261 Dry Season Rice 2,378,232 2,394,404 2,410,686 2,427,079 2,443,583 2,460,199 2,476,928 2,493,772 2,510,729 2,527,802 2,544,991 2,562,297 2,579,721 2,597,263 2,614,924 2,632,706 2,650,608 2,668,632 Vegetables 492,811 496,654 500,528 504,432 508,367 512,332 516,328 520,356 524,415 528,505 532,627 536,782 540,969 545,188 549,441 553,726 558,046 562,398 Total Rice 9,559,396 9,594,239 9,629,241 9,664,402 9,699,723 9,735,205 9,770,850 9,806,657 9,842,628 9,878,764 9,915,066 9,951,534 9,988,169 10,024,974 10,061,947 10,099,091 10,136,406 10,173,893 - 0.4% 0.7% 1.1% 1.5% 1.8% 2.2% 2.6% 3.0% 3.3% 3.7% 4.1% 4.5% 4.9% 5.3% 5.6% 6.0% 6.4% GROSS MARGINS (Million $) 0 1 2 3 4 5 6 7 8 9 Cassava 170.82 176.47 182.31 188.34 194.58 201.02 207.67 214.55 221.65 228.99 236.56 244.40 252.48 260.84 269.48 278.40 287.61 297.13 Maize 65.48 66.11 66.76 67.40 68.06 68.72 69.38 70.06 70.74 71.42 72.12 72.81 73.52 74.23 74.95 75.68 76.42 77.16 Wet Season Rice 609.70 611.29 612.88 614.47 616.07 617.67 619.28 620.89 622.50 624.12 625.74 627.37 629.00 630.64 632.28 633.92 635.57 637.22 Dry Season Rice 146.43 147.43 148.43 149.44 150.45 151.48 152.51 153.54 154.59 155.64 156.70 157.76 158.84 159.92 161.00 162.10 163.20 164.31 Vegetables 75.47 76.06 76.65 77.25 77.85 78.46 79.07 79.69 80.31 80.93 81.57 82.20 82.84 83.49 84.14 84.80 85.46 86.12 Total 5 Products 1,067.90 1,077.35 1,087.02 1,096.90 1,107.01 1,117.34 1,127.91 1,138.72 1,149.78 1,161.10 1,172.68 1,184.54 1,196.68 1,209.12 1,221.85 1,234.89 1,248.25 1,261.94 - 0.9% 1.8% 2.7% 3.7% 4.6% 5.6% 6.6% 7.7% 8.7% 9.8% 10.9% 12.1% 13.2% 14.4% 15.6% 16.9% 18.2% Total Rice 756.13 758.71 761.31 763.91 766.52 769.15 771.78 774.43 777.09 779.76 782.44 785.13 787.84 790.55 793.28 796.02 798.77 801.53 - 0.3% 0.7% 1.0% 1.4% 1.7% 2.1% 2.4% 2.8% 3.1% 3.5% 3.8% 4.2% 4.6% 4.9% 5.3% 5.6% 6.0% Share of Rice 0.71 0.70 0.70 0.70 0.69 0.69 0.68 0.68 0.68 0.67 0.67 0.66 0.66 0.65 0.65 0.64 0.64 0.64 188 Table 97: Results of simulations on land use (scenario L3) Scenario L3 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 Cassava 0 1 2 3 4 5 6 7 8 9 Cassava 337,800 353,454 369,833 386,971 404,903 423,666 443,299 463,841 485,336 507,826 531,359 555,982 581,746 608,704 636,912 666,426 697,308 729,622 Maize 215,442 218,368 221,333 224,339 227,385 230,473 233,603 236,775 239,991 243,250 246,553 249,901 253,295 256,735 260,221 263,755 267,337 270,967 Wet Season Rice 2,484,832 2,484,832 2,484,832 2,484,832 2,484,832 2,484,832 2,484,832 2,484,832 2,484,832 2,484,832 2,484,832 2,484,832 2,484,832 2,484,832 2,484,832 2,484,832 2,484,832 2,484,832 Dry Season Rice 495,465 495,465 495,465 495,465 495,465 495,465 495,465 495,465 495,465 495,465 495,465 495,465 495,465 495,465 495,465 495,465 495,465 495,465 Vegetables 54,155 54,746 55,344 55,949 56,560 57,177 57,802 58,433 59,071 59,716 60,368 61,027 61,694 62,367 63,048 63,737 64,433 65,136 Total 5 Products 3,587,694 3,606,865 3,626,807 3,647,555 3,669,145 3,691,614 3,715,000 3,739,346 3,764,694 3,791,089 3,818,577 3,847,208 3,877,032 3,908,103 3,940,478 3,974,215 4,009,375 4,046,022 - 0.5% 1.1% 1.7% 2.3% 2.9% 3.5% 4.2% 4.9% 5.7% 6.4% 7.2% 8.1% 8.9% 9.8% 10.8% 11.8% 12.8% PRODUCTION (tons) 0 1 2 3 4 5 6 7 8 9 Cassava 6,299,970 6,591,911 6,897,380 7,217,004 7,551,440 7,901,374 8,267,524 8,650,641 9,051,511 9,470,959 9,909,843 10,369,065 10,849,567 11,352,336 11,878,404 12,428,849 13,004,802 13,607,444 Maize 848,841 860,369 872,053 883,895 895,898 908,065 920,396 932,895 945,564 958,405 971,420 984,612 997,983 1,011,535 1,025,272 1,039,195 1,053,307 1,067,611 Wet Season Rice 7,181,164 7,181,164 7,181,164 7,181,164 7,181,164 7,181,164 7,181,164 7,181,164 7,181,164 7,181,164 7,181,164 7,181,164 7,181,164 7,181,164 7,181,164 7,181,164 7,181,164 7,181,164 Dry Season Rice 2,378,232 2,378,232 2,378,232 2,378,232 2,378,232 2,378,232 2,378,232 2,378,232 2,378,232 2,378,232 2,378,232 2,378,232 2,378,232 2,378,232 2,378,232 2,378,232 2,378,232 2,378,232 Vegetables 492,811 498,192 503,632 509,132 514,692 520,312 525,994 531,738 537,544 543,414 549,348 555,347 561,412 567,542 573,740 580,005 586,339 592,742 Total Rice 9,559,396 9,559,396 9,559,396 9,559,396 9,559,396 9,559,396 9,559,396 9,559,396 9,559,396 9,559,396 9,559,396 9,559,396 9,559,396 9,559,396 9,559,396 9,559,396 9,559,396 9,559,396 - 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% GROSS MARGINS (Million $) 0 1 2 3 4 5 6 7 8 9 Cassava 170.82 178.73 187.01 195.68 204.75 214.24 224.16 234.55 245.42 256.79 268.69 281.14 294.17 307.80 322.07 336.99 352.61 368.95 Maize 65.48 66.37 67.27 68.18 69.11 70.05 71.00 71.96 72.94 73.93 74.93 75.95 76.98 78.03 79.09 80.16 81.25 82.36 Wet Season Rice 609.70 609.70 609.70 609.70 609.70 609.70 609.70 609.70 609.70 609.70 609.70 609.70 609.70 609.70 609.70 609.70 609.70 609.70 Dry Season Rice 146.43 146.43 146.43 146.43 146.43 146.43 146.43 146.43 146.43 146.43 146.43 146.43 146.43 146.43 146.43 146.43 146.43 146.43 Vegetables 75.47 76.29 77.12 77.97 78.82 79.68 80.55 81.43 82.32 83.22 84.13 85.04 85.97 86.91 87.86 88.82 89.79 90.77 Total 5 Products 1,067.90 1,077.52 1,087.54 1,097.96 1,108.81 1,120.10 1,131.84 1,144.08 1,156.81 1,170.07 1,183.89 1,198.27 1,213.26 1,228.88 1,245.15 1,262.11 1,279.78 1,298.21 - 0.9% 1.8% 2.8% 3.8% 4.9% 6.0% 7.1% 8.3% 9.6% 10.9% 12.2% 13.6% 15.1% 16.6% 18.2% 19.8% 21.6% Total Rice 756.13 756.13 756.13 756.13 756.13 756.13 756.13 756.13 756.13 756.13 756.13 756.13 756.13 756.13 756.13 756.13 756.13 756.13 - 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% Share of Rice 0.71 0.70 0.70 0.69 0.68 0.68 0.67 0.66 0.65 0.65 0.64 0.63 0.62 0.62 0.61 0.60 0.59 0.58 189 Table 98: Results of simulations on land use (scenario L4) Scenario L4 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 Cassava 0 1 2 3 4 5 6 7 8 9 Cassava 337,800 337,800 337,800 337,800 337,800 337,800 337,800 337,800 337,800 337,800 337,800 337,800 337,800 337,800 337,800 337,800 337,800 337,800 Maize 215,442 215,442 215,442 215,442 215,442 215,442 215,442 215,442 215,442 215,442 215,442 215,442 215,442 215,442 215,442 215,442 215,442 215,442 Wet Season Rice 2,484,832 2,484,832 2,484,832 2,484,832 2,484,832 2,484,832 2,484,832 2,484,832 2,484,832 2,484,832 2,484,832 2,484,832 2,484,832 2,484,832 2,484,832 2,484,832 2,484,832 2,484,832 Dry Season Rice 495,465 495,465 495,465 495,465 495,465 495,465 495,465 495,465 495,465 495,465 495,465 495,465 495,465 495,465 495,465 495,465 495,465 495,465 Vegetables 54,155 54,155 54,155 54,155 54,155 54,155 54,155 54,155 54,155 54,155 54,155 54,155 54,155 54,155 54,155 54,155 54,155 54,155 Total 5 Products 3,587,694 3,587,694 3,587,694 3,587,694 3,587,694 3,587,694 3,587,694 3,587,694 3,587,694 3,587,694 3,587,694 3,587,694 3,587,694 3,587,694 3,587,694 3,587,694 3,587,694 3,587,694 - 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% PRODUCTION (tons) 0 1 2 3 4 5 6 7 8 9 Cassava 6,299,970 6,161,873 6,026,803 5,894,694 5,765,481 5,639,100 5,515,490 5,394,589 5,276,338 5,160,679 5,047,556 4,936,912 4,828,694 4,722,848 4,619,322 4,518,065 4,419,028 4,322,161 Maize 848,841 848,841 848,841 848,841 848,841 848,841 848,841 848,841 848,841 848,841 848,841 848,841 848,841 848,841 848,841 848,841 848,841 848,841 Wet Season Rice 7,181,164 7,181,164 7,181,164 7,181,164 7,181,164 7,181,164 7,181,164 7,181,164 7,181,164 7,181,164 7,181,164 7,181,164 7,181,164 7,181,164 7,181,164 7,181,164 7,181,164 7,181,164 Dry Season Rice 2,378,232 2,378,232 2,378,232 2,378,232 2,378,232 2,378,232 2,378,232 2,378,232 2,378,232 2,378,232 2,378,232 2,378,232 2,378,232 2,378,232 2,378,232 2,378,232 2,378,232 2,378,232 Vegetables 492,811 492,811 492,811 492,811 492,811 492,811 492,811 492,811 492,811 492,811 492,811 492,811 492,811 492,811 492,811 492,811 492,811 492,811 Total Rice 9,559,396 9,559,396 9,559,396 9,559,396 9,559,396 9,559,396 9,559,396 9,559,396 9,559,396 9,559,396 9,559,396 9,559,396 9,559,396 9,559,396 9,559,396 9,559,396 9,559,396 9,559,396 - 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% GROSS MARGINS (Million $) 0 1 2 3 4 5 6 7 8 9 Cassava 170.82 167.07 163.41 159.83 156.32 152.90 149.55 146.27 143.06 139.92 136.86 133.86 130.92 128.05 125.25 122.50 119.82 117.19 Maize 65.48 65.48 65.48 65.48 65.48 65.48 65.48 65.48 65.48 65.48 65.48 65.48 65.48 65.48 65.48 65.48 65.48 65.48 Wet Season Rice 609.70 609.70 609.70 609.70 609.70 609.70 609.70 609.70 609.70 609.70 609.70 609.70 609.70 609.70 609.70 609.70 609.70 609.70 Dry Season Rice 146.43 146.43 146.43 146.43 146.43 146.43 146.43 146.43 146.43 146.43 146.43 146.43 146.43 146.43 146.43 146.43 146.43 146.43 Vegetables 75.47 75.47 75.47 75.47 75.47 75.47 75.47 75.47 75.47 75.47 75.47 75.47 75.47 75.47 75.47 75.47 75.47 75.47 Total 5 Products 1,067.90 1,064.15 1,060.49 1,056.91 1,053.40 1,049.98 1,046.63 1,043.35 1,040.14 1,037.00 1,033.94 1,030.94 1,028.00 1,025.13 1,022.33 1,019.58 1,016.90 1,014.27 - -0.4% -0.7% -1.0% -1.4% -1.7% -2.0% -2.3% -2.6% -2.9% -3.2% -3.5% -3.7% -4.0% -4.3% -4.5% -4.8% -5.0% Total Rice 756.13 756.13 756.13 756.13 756.13 756.13 756.13 756.13 756.13 756.13 756.13 756.13 756.13 756.13 756.13 756.13 756.13 756.13 - 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% Share of Rice 0.71 0.71 0.71 0.72 0.72 0.72 0.72 0.72 0.73 0.73 0.73 0.73 0.74 0.74 0.74 0.74 0.74 0.75 Cassava yield with L4 18.65 18.24 17.84 17.45 17.07 16.69 16.33 15.97 15.62 15.28 14.94 14.61 14.29 13.98 13.67 13.37 13.08 12.80 -2% -4% -6% -8% -10% -12% -14% -16% -18% -20% -22% -23% -25% -27% -28% -30% -31% 190 Table 99: Results of simulations on land use (scenario L5) Scenario L5 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 Cassava 0 1 2 3 4 5 6 7 8 9 Cassava 337,800 329,693 321,780 314,057 306,520 299,164 291,984 284,976 278,137 271,461 264,946 258,588 252,381 246,324 240,413 234,643 229,011 223,515 Maize 215,442 210,271 205,225 200,299 195,492 190,800 186,221 181,752 177,390 173,133 168,977 164,922 160,964 157,101 153,330 149,650 146,059 142,553 Wet Season Rice 2,484,832 2,484,832 2,484,832 2,484,832 2,484,832 2,484,832 2,484,832 2,484,832 2,484,832 2,484,832 2,484,832 2,484,832 2,484,832 2,484,832 2,484,832 2,484,832 2,484,832 2,484,832 Dry Season Rice 495,465 495,465 495,465 495,465 495,465 495,465 495,465 495,465 495,465 495,465 495,465 495,465 495,465 495,465 495,465 495,465 495,465 495,465 Vegetables 54,155 54,155 54,155 54,155 54,155 54,155 54,155 54,155 54,155 54,155 54,155 54,155 54,155 54,155 54,155 54,155 54,155 54,155 Total 5 Products 3,587,694 3,574,416 3,561,457 3,548,809 3,536,464 3,524,416 3,512,657 3,501,180 3,489,979 3,479,046 3,468,376 3,457,961 3,447,797 3,437,877 3,428,195 3,418,745 3,409,522 3,400,520 - -0.4% -0.7% -1.1% -1.4% -1.8% -2.1% -2.4% -2.7% -3.0% -3.3% -3.6% -3.9% -4.2% -4.4% -4.7% -5.0% -5.2% PRODUCTION (tons) 0 1 2 3 4 5 6 7 8 9 Cassava 6,299,970 6,148,771 6,001,200 5,857,171 5,716,599 5,579,401 5,445,495 5,314,803 5,187,248 5,062,754 4,941,248 4,822,658 4,706,914 4,593,948 4,483,694 4,376,085 4,271,059 4,168,554 Maize 848,841 828,469 808,586 789,180 770,240 751,754 733,712 716,103 698,916 682,142 665,771 649,792 634,197 618,977 604,121 589,622 575,471 561,660 Wet Season Rice 7,181,164 7,181,164 7,181,164 7,181,164 7,181,164 7,181,164 7,181,164 7,181,164 7,181,164 7,181,164 7,181,164 7,181,164 7,181,164 7,181,164 7,181,164 7,181,164 7,181,164 7,181,164 Dry Season Rice 2,378,232 2,378,232 2,378,232 2,378,232 2,378,232 2,378,232 2,378,232 2,378,232 2,378,232 2,378,232 2,378,232 2,378,232 2,378,232 2,378,232 2,378,232 2,378,232 2,378,232 2,378,232 Vegetables 492,811 492,811 492,811 492,811 492,811 492,811 492,811 492,811 492,811 492,811 492,811 492,811 492,811 492,811 492,811 492,811 492,811 492,811 Total Rice 9,559,396 9,559,396 9,559,396 9,559,396 9,559,396 9,559,396 9,559,396 9,559,396 9,559,396 9,559,396 9,559,396 9,559,396 9,559,396 9,559,396 9,559,396 9,559,396 9,559,396 9,559,396 - 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% GROSS MARGINS (Million $) 0 1 2 3 4 5 6 7 8 9 Cassava 170.82 166.72 162.71 158.81 155.00 151.28 147.65 144.10 140.65 137.27 133.98 130.76 127.62 124.56 121.57 118.65 115.80 113.02 Maize 65.48 63.91 62.37 60.88 59.42 57.99 56.60 55.24 53.91 52.62 51.36 50.12 48.92 47.75 46.60 45.48 44.39 43.33 Wet Season Rice 609.70 609.70 609.70 609.70 609.70 609.70 609.70 609.70 609.70 609.70 609.70 609.70 609.70 609.70 609.70 609.70 609.70 609.70 Dry Season Rice 146.43 146.43 146.43 146.43 146.43 146.43 146.43 146.43 146.43 146.43 146.43 146.43 146.43 146.43 146.43 146.43 146.43 146.43 Vegetables 75.47 75.47 75.47 75.47 75.47 75.47 75.47 75.47 75.47 75.47 75.47 75.47 75.47 75.47 75.47 75.47 75.47 75.47 Total 5 Products 1,067.90 1,062.22 1,056.69 1,051.29 1,046.01 1,040.87 1,035.85 1,030.94 1,026.16 1,021.49 1,016.93 1,012.49 1,008.14 1,003.91 999.77 995.74 991.80 987.95 - -0.5% -1.0% -1.6% -2.0% -2.5% -3.0% -3.5% -3.9% -4.3% -4.8% -5.2% -5.6% -6.0% -6.4% -6.8% -7.1% -7.5% Total Rice 756.13 756.13 756.13 756.13 756.13 756.13 756.13 756.13 756.13 756.13 756.13 756.13 756.13 756.13 756.13 756.13 756.13 756.13 - 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% Share of Rice 0.71 0.71 0.72 0.72 0.72 0.73 0.73 0.73 0.74 0.74 0.74 0.75 0.75 0.75 0.76 0.76 0.76 0.77 Cassava yield with L5 18.65 18.65 18.65 18.65 18.65 18.65 18.65 18.65 18.65 18.65 18.65 18.65 18.65 18.65 18.65 18.65 18.65 18.65 Maize Yield with L5 3.94 3.94 3.94 3.94 3.94 3.94 3.94 3.94 3.94 3.94 3.94 3.94 3.94 3.94 3.94 3.94 3.94 3.94 191 Table 100: Results of simulations on land use (scenario L6) Scenario L6 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 Cassava 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 Cassava 337,800 329,693 321,780 314,057 306,520 299,164 291,984 284,976 278,137 271,461 264,946 258,588 252,381 246,324 240,413 234,643 229,011 223,515 Maize 215,442 210,271 205,225 200,299 195,492 190,800 186,221 181,752 177,390 173,133 168,977 164,922 160,964 157,101 153,330 149,650 146,059 142,553 Wet Season Rice 2,484,832 2,484,832 2,484,832 2,484,832 2,484,832 2,484,832 2,484,832 2,484,832 2,484,832 2,484,832 2,484,832 2,484,832 2,484,832 2,484,832 2,484,832 2,484,832 2,484,832 2,484,832 Dry Season Rice 495,465 495,465 495,465 495,465 495,465 495,465 495,465 495,465 495,465 495,465 495,465 495,465 495,465 495,465 495,465 495,465 495,465 495,465 Vegetables 54,155 54,155 54,155 54,155 54,155 54,155 54,155 54,155 54,155 54,155 54,155 54,155 54,155 54,155 54,155 54,155 54,155 54,155 Total 5 Products 3,587,694 3,574,416 3,561,457 3,548,809 3,536,464 3,524,416 3,512,657 3,501,180 3,489,979 3,479,046 3,468,376 3,457,961 3,447,797 3,437,877 3,428,195 3,418,745 3,409,522 3,400,520 - -0.4% -0.7% -1.1% -1.4% -1.8% -2.1% -2.4% -2.7% -3.0% -3.3% -3.6% -3.9% -4.2% -4.4% -4.7% -5.0% -5.2% PRODUCTION (tons) 0 1 2 3 4 5 6 7 8 9 Cassava 6,299,970 6,296,341 6,292,715 6,289,090 6,285,467 6,281,847 6,278,229 6,274,612 6,270,998 6,267,386 6,263,776 6,260,168 6,256,562 6,252,959 6,249,357 6,245,757 6,242,160 6,238,564 Maize 848,841 848,353 847,864 847,376 846,887 846,400 845,912 845,425 844,938 844,451 843,965 843,479 842,993 842,507 842,022 841,537 841,052 840,568 Wet Season Rice 7,181,164 7,181,164 7,181,164 7,181,164 7,181,164 7,181,164 7,181,164 7,181,164 7,181,164 7,181,164 7,181,164 7,181,164 7,181,164 7,181,164 7,181,164 7,181,164 7,181,164 7,181,164 Dry Season Rice 2,378,232 2,378,232 2,378,232 2,378,232 2,378,232 2,378,232 2,378,232 2,378,232 2,378,232 2,378,232 2,378,232 2,378,232 2,378,232 2,378,232 2,378,232 2,378,232 2,378,232 2,378,232 Vegetables 492,811 492,811 492,811 492,811 492,811 492,811 492,811 492,811 492,811 492,811 492,811 492,811 492,811 492,811 492,811 492,811 492,811 492,811 Total Rice 9,559,396 9,559,396 9,559,396 9,559,396 9,559,396 9,559,396 9,559,396 9,559,396 9,559,396 9,559,396 9,559,396 9,559,396 9,559,396 9,559,396 9,559,396 9,559,396 9,559,396 9,559,396 - 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% GROSS MARGINS (Million $) 0 1 2 3 4 5 6 7 8 9 Cassava 170.82 170.72 170.62 170.52 170.42 170.32 170.23 170.13 170.03 169.93 169.83 169.74 169.64 169.54 169.44 169.35 169.25 169.15 Maize 65.48 65.44 65.40 65.37 65.33 65.29 65.25 65.22 65.18 65.14 65.10 65.07 65.03 64.99 64.95 64.92 64.88 64.84 Wet Season Rice 609.70 609.70 609.70 609.70 609.70 609.70 609.70 609.70 609.70 609.70 609.70 609.70 609.70 609.70 609.70 609.70 609.70 609.70 Dry Season Rice 146.43 146.43 146.43 146.43 146.43 146.43 146.43 146.43 146.43 146.43 146.43 146.43 146.43 146.43 146.43 146.43 146.43 146.43 Vegetables 75.47 75.47 75.47 75.47 75.47 75.47 75.47 75.47 75.47 75.47 75.47 75.47 75.47 75.47 75.47 75.47 75.47 75.47 Total 5 Products 1,067.90 1,067.76 1,067.62 1,067.49 1,067.35 1,067.22 1,067.08 1,066.94 1,066.81 1,066.67 1,066.54 1,066.40 1,066.27 1,066.13 1,066.00 1,065.86 1,065.73 1,065.59 - 0.0% 0.0% 0.0% -0.1% -0.1% -0.1% -0.1% -0.1% -0.1% -0.1% -0.1% -0.2% -0.2% -0.2% -0.2% -0.2% -0.2% Total Rice 756.13 756.13 756.13 756.13 756.13 756.13 756.13 756.13 756.13 756.13 756.13 756.13 756.13 756.13 756.13 756.13 756.13 756.13 - 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% Share of Rice 0.71 0.71 0.71 0.71 0.71 0.71 0.71 0.71 0.71 0.71 0.71 0.71 0.71 0.71 0.71 0.71 0.71 0.71 Cassava yield with L6 18.65 19.10 19.56 20.03 20.51 21.00 21.50 22.02 22.55 23.09 23.64 24.21 24.79 25.39 25.99 26.62 27.26 27.91 2% 5% 7% 10% 13% 15% 18% 21% 24% 27% 30% 33% 36% 39% 43% 46% 50% Maize Yield with L6 3.94 4.03 4.13 4.23 4.33 4.44 4.54 4.65 4.76 4.88 4.99 5.11 5.24 5.36 5.49 5.62 5.76 5.90 2% 5% 7% 10% 13% 15% 18% 21% 24% 27% 30% 33% 36% 39% 43% 46% 50% 192 Table 101: Demand for labor for environmental sustainability scenarios DEMAND FOR LABOR (Million DAYS) 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 Scenario L1 Cassava 16.48 16.48 16.48 16.48 16.48 16.48 16.48 16.48 16.48 16.48 16.48 16.48 16.48 16.48 16.48 16.48 16.48 16.48 Maize 6.75 6.75 6.75 6.75 6.75 6.75 6.75 6.75 6.75 6.75 6.75 6.75 6.75 6.75 6.75 6.75 6.75 6.75 Wet Season Rice 120.00 120.78 121.56 122.35 123.15 123.95 124.75 125.56 126.38 127.20 128.03 128.86 129.70 130.54 131.39 132.24 133.10 133.97 Dry Season Rice 13.75 13.98 14.22 14.46 14.71 14.96 15.21 15.47 15.73 16.00 16.27 16.55 16.83 17.12 17.41 17.70 18.01 18.31 Vegetables 9.53 9.53 9.53 9.53 9.53 9.53 9.53 9.53 9.53 9.53 9.53 9.53 9.53 9.53 9.53 9.53 9.53 9.53 166.51 167.52 168.54 169.57 170.61 171.67 172.73 173.79 174.87 175.96 177.06 178.17 179.29 180.42 181.56 182.71 183.87 185.04 - 0.6% 1.2% 1.8% 2.5% 3.1% 3.7% 4.4% 5.0% 5.7% 6.3% 7.0% 7.7% 8.4% 9.0% 9.7% 10.4% 11.1% Scenario L2 Cassava 16.48 17.03 17.59 18.18 18.78 19.40 20.04 20.71 21.39 22.10 22.83 23.59 24.37 25.17 26.01 26.87 27.76 28.68 Maize 6.75 6.81 6.88 6.95 7.01 7.08 7.15 7.22 7.29 7.36 7.43 7.51 7.58 7.65 7.73 7.80 7.88 7.95 Wet Season Rice 120.00 120.31 120.62 120.94 121.25 121.56 121.88 122.20 122.52 122.83 123.15 123.47 123.79 124.12 124.44 124.76 125.09 125.41 Dry Season Rice 13.75 13.84 13.94 14.03 14.13 14.22 14.32 14.42 14.52 14.61 14.71 14.81 14.91 15.02 15.12 15.22 15.32 15.43 Vegetables 9.53 9.60 9.67 9.75 9.83 9.90 9.98 10.06 10.14 10.22 10.30 10.38 10.46 10.54 10.62 10.70 10.79 10.87 166.51 167.60 168.71 169.84 171.00 172.17 173.37 174.60 175.85 177.12 178.42 179.75 181.11 182.49 183.91 185.35 186.83 188.34 - 0.7% 1.3% 2.0% 2.7% 3.4% 4.1% 4.9% 5.6% 6.4% 7.2% 8.0% 8.8% 9.6% 10.5% 11.3% 12.2% 13.1% Scenario L3 Cassava 16.48 17.25 18.05 18.88 19.76 20.68 21.63 22.64 23.68 24.78 25.93 27.13 28.39 29.71 31.08 32.52 34.03 35.61 Maize 6.75 6.84 6.93 7.03 7.12 7.22 7.32 7.42 7.52 7.62 7.72 7.83 7.93 8.04 8.15 8.26 8.37 8.49 Wet Season Rice 120.00 120.00 120.00 120.00 120.00 120.00 120.00 120.00 120.00 120.00 120.00 120.00 120.00 120.00 120.00 120.00 120.00 120.00 Dry Season Rice 13.75 13.75 13.75 13.75 13.75 13.75 13.75 13.75 13.75 13.75 13.75 13.75 13.75 13.75 13.75 13.75 13.75 13.75 Vegetables 9.53 9.63 9.73 9.84 9.95 10.06 10.17 10.28 10.39 10.50 10.62 10.73 10.85 10.97 11.09 11.21 11.33 11.46 166.51 167.46 168.46 169.50 170.58 171.70 172.86 174.08 175.34 176.65 178.02 179.44 180.92 182.46 184.07 185.74 187.48 189.30 - 0.6% 1.2% 1.8% 2.4% 3.1% 3.8% 4.5% 5.3% 6.1% 6.9% 7.8% 8.7% 9.6% 10.5% 11.6% 12.6% 13.7% Scenario L4 Cassava 16.48 16.48 16.48 16.48 16.48 16.48 16.48 16.48 16.48 16.48 16.48 16.48 16.48 16.48 16.48 16.48 16.48 16.48 Maize 6.75 6.75 6.75 6.75 6.75 6.75 6.75 6.75 6.75 6.75 6.75 6.75 6.75 6.75 6.75 6.75 6.75 6.75 Wet Season Rice 120.00 120.00 120.00 120.00 120.00 120.00 120.00 120.00 120.00 120.00 120.00 120.00 120.00 120.00 120.00 120.00 120.00 120.00 Dry Season Rice 13.75 13.75 13.75 13.75 13.75 13.75 13.75 13.75 13.75 13.75 13.75 13.75 13.75 13.75 13.75 13.75 13.75 13.75 Vegetables 9.53 9.53 9.53 9.53 9.53 9.53 9.53 9.53 9.53 9.53 9.53 9.53 9.53 9.53 9.53 9.53 9.53 9.53 166.51 166.51 166.51 166.51 166.51 166.51 166.51 166.51 166.51 166.51 166.51 166.51 166.51 166.51 166.51 166.51 166.51 166.51 - 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% Scenario L5 Cassava 16.48 16.09 15.70 15.33 14.96 14.60 14.25 13.91 13.57 13.25 12.93 12.62 12.32 12.02 11.73 11.45 11.18 10.91 Maize 6.75 6.59 6.43 6.27 6.12 5.98 5.83 5.69 5.56 5.42 5.29 5.17 5.04 4.92 4.80 4.69 4.58 4.47 Wet Season Rice 120.00 120.00 120.00 120.00 120.00 120.00 120.00 120.00 120.00 120.00 120.00 120.00 120.00 120.00 120.00 120.00 120.00 120.00 Dry Season Rice 13.75 13.75 13.75 13.75 13.75 13.75 13.75 13.75 13.75 13.75 13.75 13.75 13.75 13.75 13.75 13.75 13.75 13.75 Vegetables 9.53 9.53 9.53 9.53 9.53 9.53 9.53 9.53 9.53 9.53 9.53 9.53 9.53 9.53 9.53 9.53 9.53 9.53 166.51 165.95 165.40 164.87 164.35 163.85 163.35 162.87 162.40 161.94 161.49 161.06 160.63 160.21 159.81 159.41 159.02 158.64 - -0.3% -0.7% -1.0% -1.3% -1.6% -1.9% -2.2% -2.5% -2.7% -3.0% -3.3% -3.5% -3.8% -4.0% -4.3% -4.5% -4.7% Scenario L6 Cassava 16.48 16.09 15.70 15.33 14.96 14.60 14.25 13.91 13.57 13.25 12.93 12.62 12.32 12.02 11.73 11.45 11.18 10.91 Maize 6.75 6.59 6.43 6.27 6.12 5.98 5.83 5.69 5.56 5.42 5.29 5.17 5.04 4.92 4.80 4.69 4.58 4.47 Wet Season Rice 120.00 120.00 120.00 120.00 120.00 120.00 120.00 120.00 120.00 120.00 120.00 120.00 120.00 120.00 120.00 120.00 120.00 120.00 Dry Season Rice 13.75 13.75 13.75 13.75 13.75 13.75 13.75 13.75 13.75 13.75 13.75 13.75 13.75 13.75 13.75 13.75 13.75 13.75 Vegetables 9.53 9.53 9.53 9.53 9.53 9.53 9.53 9.53 9.53 9.53 9.53 9.53 9.53 9.53 9.53 9.53 9.53 9.53 166.51 165.95 165.40 164.87 164.35 163.85 163.35 162.87 162.40 161.94 161.49 161.06 160.63 160.21 159.81 159.41 159.02 158.64 - -0.3% -0.7% -1.0% -1.3% -1.6% -1.9% -2.2% -2.5% -2.7% -3.0% -3.3% -3.5% -3.8% -4.0% -4.3% -4.5% -4.7% 193 Table 102: Demand for fertilizers for environmental sustainability scenarios DEMAND FOR FERTILIZERS (TONS) 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 Scenario L1 Cassava 4,578 4,578 4,578 4,578 4,578 4,578 4,578 4,578 4,578 4,578 4,578 4,578 4,578 4,578 4,578 4,578 4,578 4,578 Maize 11,507 11,507 11,507 11,507 11,507 11,507 11,507 11,507 11,507 11,507 11,507 11,507 11,507 11,507 11,507 11,507 11,507 11,507 Wet Season Rice 294,408 296,322 298,248 300,187 302,138 304,102 306,078 308,068 310,070 312,086 314,114 316,156 318,211 320,279 322,361 324,457 326,566 328,688 Dry Season Rice 135,722 138,029 140,376 142,762 145,189 147,657 150,168 152,720 155,317 157,957 160,642 163,373 166,151 168,975 171,848 174,769 177,740 180,762 Vegetables 18,646 18,646 18,646 18,646 18,646 18,646 18,646 18,646 18,646 18,646 18,646 18,646 18,646 18,646 18,646 18,646 18,646 18,646 464,861 469,082 473,355 477,680 482,058 486,490 490,977 495,519 500,118 504,774 509,488 514,260 519,093 523,986 528,940 533,957 539,037 544,181 - 0.9% 1.8% 2.8% 3.7% 4.7% 5.6% 6.6% 7.6% 8.6% 9.6% 10.6% 11.7% 12.7% 13.8% 14.9% 16.0% 17.1% Scenario L2 Cassava 4,578 4,730 4,887 5,048 5,215 5,388 5,566 5,751 5,941 6,138 6,341 6,551 6,767 6,991 7,223 7,462 7,709 7,964 Maize 11,507 11,619 11,731 11,845 11,960 12,076 12,193 12,311 12,431 12,551 12,673 12,796 12,920 13,045 13,172 13,300 13,429 13,559 Wet Season Rice 294,408 295,174 295,941 296,711 297,482 298,255 299,031 299,808 300,588 301,369 302,153 302,939 303,726 304,516 305,308 306,101 306,897 307,695 Dry Season Rice 135,722 136,645 137,574 138,510 139,451 140,400 141,354 142,316 143,283 144,258 145,239 146,226 147,221 148,222 149,230 150,244 151,266 152,295 Vegetables 18,646 18,791 18,938 19,085 19,234 19,384 19,536 19,688 19,842 19,996 20,152 20,309 20,468 20,628 20,788 20,951 21,114 21,279 464,861 466,958 469,071 471,199 473,343 475,504 477,680 479,874 482,085 484,312 486,558 488,821 491,102 493,402 495,721 498,058 500,415 502,792 - 0.5% 0.9% 1.4% 1.8% 2.3% 2.8% 3.2% 3.7% 4.2% 4.7% 5.2% 5.6% 6.1% 6.6% 7.1% 7.6% 8.2% Scenario L3 Cassava 4,578 4,791 5,013 5,245 5,488 5,742 6,008 6,287 6,578 6,883 7,202 7,536 7,885 8,250 8,632 9,032 9,451 9,889 Maize 11,507 11,663 11,822 11,982 12,145 12,310 12,477 12,646 12,818 12,992 13,169 13,347 13,529 13,712 13,899 14,087 14,279 14,473 Wet Season Rice 294,408 294,408 294,408 294,408 294,408 294,408 294,408 294,408 294,408 294,408 294,408 294,408 294,408 294,408 294,408 294,408 294,408 294,408 Dry Season Rice 135,722 135,722 135,722 135,722 135,722 135,722 135,722 135,722 135,722 135,722 135,722 135,722 135,722 135,722 135,722 135,722 135,722 135,722 Vegetables 18,646 18,849 19,055 19,263 19,474 19,686 19,901 20,119 20,338 20,560 20,785 21,012 21,241 21,473 21,708 21,945 22,184 22,427 464,861 465,433 466,020 466,620 467,237 467,868 468,517 469,182 469,865 470,566 471,286 472,025 472,785 473,566 474,369 475,195 476,044 476,918 - 0.1% 0.2% 0.4% 0.5% 0.6% 0.8% 0.9% 1.1% 1.2% 1.4% 1.5% 1.7% 1.9% 2.0% 2.2% 2.4% 2.6% Scenario L4 Cassava 4,578 4,578 4,578 4,578 4,578 4,578 4,578 4,578 4,578 4,578 4,578 4,578 4,578 4,578 4,578 4,578 4,578 4,578 Maize 11,507 11,507 11,507 11,507 11,507 11,507 11,507 11,507 11,507 11,507 11,507 11,507 11,507 11,507 11,507 11,507 11,507 11,507 Wet Season Rice 294,408 294,408 294,408 294,408 294,408 294,408 294,408 294,408 294,408 294,408 294,408 294,408 294,408 294,408 294,408 294,408 294,408 294,408 Dry Season Rice 135,722 135,722 135,722 135,722 135,722 135,722 135,722 135,722 135,722 135,722 135,722 135,722 135,722 135,722 135,722 135,722 135,722 135,722 Vegetables 18,646 18,646 18,646 18,646 18,646 18,646 18,646 18,646 18,646 18,646 18,646 18,646 18,646 18,646 18,646 18,646 18,646 18,646 464,861 464,861 464,861 464,861 464,861 464,861 464,861 464,861 464,861 464,861 464,861 464,861 464,861 464,861 464,861 464,861 464,861 464,861 - 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% Scenario L5 Cassava 4,578 4,469 4,361 4,257 4,154 4,055 3,957 3,862 3,770 3,679 3,591 3,505 3,421 3,339 3,258 3,180 3,104 3,029 Maize 11,507 11,231 10,961 10,698 10,441 10,191 9,946 9,708 9,475 9,247 9,025 8,809 8,597 8,391 8,189 7,993 7,801 7,614 Wet Season Rice 294,408 294,408 294,408 294,408 294,408 294,408 294,408 294,408 294,408 294,408 294,408 294,408 294,408 294,408 294,408 294,408 294,408 294,408 Dry Season Rice 135,722 135,722 135,722 135,722 135,722 135,722 135,722 135,722 135,722 135,722 135,722 135,722 135,722 135,722 135,722 135,722 135,722 135,722 Vegetables 18,646 18,646 18,646 18,646 18,646 18,646 18,646 18,646 18,646 18,646 18,646 18,646 18,646 18,646 18,646 18,646 18,646 18,646 464,861 464,475 464,098 463,731 463,372 463,022 462,680 462,346 462,020 461,702 461,392 461,089 460,794 460,505 460,224 459,949 459,681 459,419 - -0.1% -0.2% -0.2% -0.3% -0.4% -0.5% -0.5% -0.6% -0.7% -0.7% -0.8% -0.9% -0.9% -1.0% -1.1% -1.1% -1.2% Scenario L6 Cassava 4,578 4,469 4,361 4,257 4,154 4,055 3,957 3,862 3,770 3,679 3,591 3,505 3,421 3,339 3,258 3,180 3,104 3,029 Maize 11,507 11,231 10,961 10,698 10,441 10,191 9,946 9,708 9,475 9,247 9,025 8,809 8,597 8,391 8,189 7,993 7,801 7,614 Wet Season Rice 294,408 294,408 294,408 294,408 294,408 294,408 294,408 294,408 294,408 294,408 294,408 294,408 294,408 294,408 294,408 294,408 294,408 294,408 Dry Season Rice 135,722 135,722 135,722 135,722 135,722 135,722 135,722 135,722 135,722 135,722 135,722 135,722 135,722 135,722 135,722 135,722 135,722 135,722 Vegetables 18,646 18,646 18,646 18,646 18,646 18,646 18,646 18,646 18,646 18,646 18,646 18,646 18,646 18,646 18,646 18,646 18,646 18,646 464,861 464,475 464,098 463,731 463,372 463,022 462,680 462,346 462,020 461,702 461,392 461,089 460,794 460,505 460,224 459,949 459,681 459,419 - -0.1% -0.2% -0.2% -0.3% -0.4% -0.5% -0.5% -0.6% -0.7% -0.7% -0.8% -0.9% -0.9% -1.0% -1.1% -1.1% -1.2% 194 Table 103: Total demand and average return to labor ($/day) for scenarios with environmental sustainability DEMAND FOR LABOR (Million DAYS) 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 Baseline 166.51 166.51 166.51 166.51 166.51 166.51 166.51 166.51 166.51 166.51 166.51 166.51 166.51 166.51 166.51 166.51 166.51 166.51 Scenario L1 166.51 167.52 168.54 169.57 170.61 171.67 172.73 173.79 174.87 175.96 177.06 178.17 179.29 180.42 181.56 182.71 183.87 185.04 Scenario L2 166.51 167.60 168.71 169.84 171.00 172.17 173.37 174.60 175.85 177.12 178.42 179.75 181.11 182.49 183.91 185.35 186.83 188.34 Scenario L3 166.51 167.46 168.46 169.50 170.58 171.70 172.86 174.08 175.34 176.65 178.02 179.44 180.92 182.46 184.07 185.74 187.48 189.30 Scenario L4 166.51 166.51 166.51 166.51 166.51 166.51 166.51 166.51 166.51 166.51 166.51 166.51 166.51 166.51 166.51 166.51 166.51 166.51 Scenario L5 166.51 165.95 165.40 164.87 164.35 163.85 163.35 162.87 162.40 161.94 161.49 161.06 160.63 160.21 159.81 159.41 159.02 158.64 Scenario L6 166.51 165.95 165.40 164.87 164.35 163.85 163.35 162.87 162.40 161.94 161.49 161.06 160.63 160.21 159.81 159.41 159.02 158.64 RETURN TO LABOR ($/day) 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 Baseline 6.41 6.41 6.41 6.41 6.41 6.41 6.41 6.41 6.41 6.41 6.41 6.41 6.41 6.41 6.41 6.41 6.41 6.41 Scenario L1 6.41 6.41 6.41 6.41 6.41 6.41 6.41 6.41 6.41 6.41 6.41 6.41 6.41 6.41 6.42 6.42 6.42 6.42 Scenario L2 6.41 6.43 6.44 6.46 6.47 6.49 6.51 6.52 6.54 6.56 6.57 6.59 6.61 6.63 6.64 6.66 6.68 6.70 Scenario L3 6.41 6.43 6.46 6.48 6.50 6.52 6.55 6.57 6.60 6.62 6.65 6.68 6.71 6.73 6.76 6.79 6.83 6.86 Scenario L4 6.41 6.39 6.37 6.35 6.33 6.31 6.29 6.27 6.25 6.23 6.21 6.19 6.17 6.16 6.14 6.12 6.11 6.09 Scenario L5 6.41 6.40 6.39 6.38 6.36 6.35 6.34 6.33 6.32 6.31 6.30 6.29 6.28 6.27 6.26 6.25 6.24 6.23 Scenario L6 6.41 6.43 6.45 6.47 6.49 6.51 6.53 6.55 6.57 6.59 6.60 6.62 6.64 6.65 6.67 6.69 6.70 6.72 195 Table 104: Farm budget data comparing traditional and modern technology users (revenue and modern inputs) Revenue by technology use and total Revenue Yield by technology use (tons/ha) Output Prices by technology use ($/ton) ($/ha) MODERN TRADITIONAL TOTAL MODERN TRADITIONAL TOTAL MODERN TRADITIONAL TOTAL Cassava 21.90 17.21 18.65 70.23 69.12 69.53 1,538.03 1,189.47 1,296.72 Maize 4.38 3.50 3.94 208.56 164.04 188.79 913.50 574.15 743.83 Wet Season Rice 3.32 2.70 2.89 278.76 252.46 261.43 925.47 681.65 755.53 Dry Season Rice 5.00 4.61 4.80 201.75 211.67 206.73 1,008.75 975.82 992.28 Vegetables 11.96 6.24 9.10 367.59 206.57 312.38 4,396.37 1,289.02 2,842.70 Seeds Seeds by farm size (unit/ha) Seed Prices by farm size ($/unit) Costs of seeds by farm size and total MODERN TRADITIONAL TOTAL MODERN TRADITIONAL TOTAL MODERN TRADITIONAL TOTAL Cassava 25,469.00 21,703.00 22,861.00 0.01 0.01 0.01 254.69 217.03 228.61 Maize 20.60 17.16 18.88 4.06 4.06 4.06 83.62 69.66 76.64 Wet Season Rice 64.35 27.73 38.82 0.66 0.66 0.66 42.47 18.30 25.62 Dry Season Rice 166.34 125.80 146.07 0.74 0.74 0.74 123.09 93.09 108.09 Vegetables 20.56 3.75 12.15 11.60 11.60 11.60 238.47 43.45 140.96 Fertilizer Fertilizer use by farm size (kg/ha) Fertilizer Prices by farm size ($/kg) Costs of fertilizers by farm size and total MODERN TRADITIONAL TOTAL TOTAL MODERN TRADITIONAL TOTAL Cassava 22.96 9.36 13.55 0.56 0.56 0.56 12.86 5.24 7.59 Maize 104.14 2.68 53.41 0.56 0.56 0.56 58.32 1.50 29.91 Wet Season Rice 176.36 93.32 118.48 0.56 0.56 0.56 98.76 52.26 66.35 Dry Season Rice 351.46 196.39 273.93 0.56 0.56 0.56 196.82 109.98 153.40 Vegetables 550.95 137.66 344.30 0.56 0.56 0.56 308.53 77.09 192.81 Other Inputs Other input Costs ($/ha) Cassava 65.16 24.48 37.00 Maize 43.61 16.03 29.82 Wet Season Rice 21.27 29.31 26.88 Dry Season Rice 30.35 25.12 27.73 Vegetables 424.30 69.36 246.85 196 Table 105: Farm budget data comparing traditional and modern technology users (labor, services, and gross margins) Labor Labor use by technology (days/ha) Wage rate by technology ($/day) Costs of labor by technology and total MODERN TRADITIONAL TOTAL MODERN TRADITIONAL TOTAL MODERN TRADITIONAL TOTAL Cassava 56.45 45.40 48.80 4.56 4.56 4.56 257.41 207.03 222.53 Maize 34.16 30.49 31.33 4.56 4.26 4.56 155.78 129.93 142.85 Wet Season Rice 28.69 58.81 48.29 4.56 4.41 4.56 130.84 259.06 220.21 Dry Season Rice 3.82 51.68 27.75 4.56 4.56 4.56 17.41 235.66 126.54 Vegetables 263.68 128.11 175.89 4.56 3.14 4.56 1,202.40 401.76 802.08 Services Service Costs ($/ha) MODERN TRADITIONAL TOTAL Cassava 253.12 314.08 295.32 Maize 175.80 145.55 160.68 Wet Season Rice 232.57 144.37 171.10 Dry Season Rice 245.02 167.55 206.29 Vegetables 68.10 64.79 66.45 Irrigation Irrigation Costs ($/ha) MODERN TRADITIONAL TOTAL Cassava Maize Wet Season Rice Dry Season Rice 141.04 8.33 74.69 Vegetables Gross Margins Gross Margins ($/ha) MODERN TRADITIONAL TOTAL Cassava 694.79 421.61 505.67 Maize 396.37 211.48 303.93 Wet Season Rice 399.56 178.35 245.37 Dry Season Rice 255.02 336.09 295.54 Vegetables 2,154.57 632.57 1,393.55 197 Table 106: Gross margins and NPV (million $) for TFP scenario (baseline) GROSS MARGINS (million $) ---> Baseline 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 NPV Gross Margins: MODERN 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 Cassava 82.15 82.15 82.15 82.15 82.15 82.15 82.15 82.15 82.15 82.15 82.15 82.15 82.15 82.15 82.15 82.15 82.15 82.15 $786 Maize 34.16 34.16 34.16 34.16 34.16 34.16 34.16 34.16 34.16 34.16 34.16 34.16 34.16 34.16 34.16 34.16 34.16 34.16 $327 Wet Season Rice 397.14 397.14 397.14 397.14 397.14 397.14 397.14 397.14 397.14 397.14 397.14 397.14 397.14 397.14 397.14 397.14 397.14 397.14 $3,801 Dry Season Rice 76.84 76.84 76.84 76.84 76.84 76.84 76.84 76.84 76.84 76.84 76.84 76.84 76.84 76.84 76.84 76.84 76.84 76.84 $735 Vegetables 46.67 46.67 46.67 46.67 46.67 46.67 46.67 46.67 46.67 46.67 46.67 46.67 46.67 46.67 46.67 46.67 46.67 46.67 $447 636.95 636.95 636.95 636.95 636.95 636.95 636.95 636.95 636.95 636.95 636.95 636.95 636.95 636.95 636.95 636.95 636.95 636.95 $6,096 Gross Margins: TRADITIONAL Cassava 92.57 92.57 92.57 92.57 92.57 92.57 92.57 92.57 92.57 92.57 92.57 92.57 92.57 92.57 92.57 92.57 92.57 92.57 $886 Maize 27.34 27.34 27.34 27.34 27.34 27.34 27.34 27.34 27.34 27.34 27.34 27.34 27.34 27.34 27.34 27.34 27.34 27.34 $262 Wet Season Rice 265.90 265.90 265.90 265.90 265.90 265.90 265.90 265.90 265.90 265.90 265.90 265.90 265.90 265.90 265.90 265.90 265.90 265.90 $2,545 Dry Season Rice 99.91 99.91 99.91 99.91 99.91 99.91 99.91 99.91 99.91 99.91 99.91 99.91 99.91 99.91 99.91 99.91 99.91 99.91 $956 Vegetables 20.55 20.55 20.55 20.55 20.55 20.55 20.55 20.55 20.55 20.55 20.55 20.55 20.55 20.55 20.55 20.55 20.55 20.55 $197 506.28 506.28 506.28 506.28 506.28 506.28 506.28 506.28 506.28 506.28 506.28 506.28 506.28 506.28 506.28 506.28 506.28 506.28 $4,845 Gross Margins: TOTAL Cassava 174.72 174.72 174.72 174.72 174.72 174.72 174.72 174.72 174.72 174.72 174.72 174.72 174.72 174.72 174.72 174.72 174.72 174.72 $1,672 Maize 61.49 61.49 61.49 61.49 61.49 61.49 61.49 61.49 61.49 61.49 61.49 61.49 61.49 61.49 61.49 61.49 61.49 61.49 $589 Wet Season Rice 663.04 663.04 663.04 663.04 663.04 663.04 663.04 663.04 663.04 663.04 663.04 663.04 663.04 663.04 663.04 663.04 663.04 663.04 $6,345 Dry Season Rice 176.76 176.76 176.76 176.76 176.76 176.76 176.76 176.76 176.76 176.76 176.76 176.76 176.76 176.76 176.76 176.76 176.76 176.76 $1,692 Vegetables 67.23 67.23 67.23 67.23 67.23 67.23 67.23 67.23 67.23 67.23 67.23 67.23 67.23 67.23 67.23 67.23 67.23 67.23 $643 1,143.23 1,143.23 1,143.23 1,143.23 1,143.23 1,143.23 1,143.23 1,143.23 1,143.23 1,143.23 1,143.23 1,143.23 1,143.23 1,143.23 1,143.23 1,143.23 1,143.23 1,143.23 $10,941 93.4% 1,067.90 1,067.90 1,067.90 1,067.90 1,067.90 1,067.90 1,067.90 1,067.90 1,067.90 1,067.90 1,067.90 1,067.90 1,067.90 1,067.90 1,067.90 1,067.90 1,067.90 1,067.90 $10,220 Share of Modern 0.56 0.56 0.56 0.56 0.56 0.56 0.56 0.56 0.56 0.56 0.56 0.56 0.56 0.56 0.56 0.56 0.56 0.56 198 Table 107: Gross margins and NPV (million $) for TFP scenario (scenario P1) GROSS MARGINS (million $) ---> Scenario T1 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 NPV MODERN 0 1 2 3 4 5 6 7 8 9 Cassava 82.15 83.09 84.04 85.01 85.98 86.97 87.97 88.98 90.00 91.03 92.08 93.13 94.20 95.29 96.38 97.49 98.60 99.74 $849 Maize 34.16 34.53 34.90 35.28 35.66 36.05 36.44 36.83 37.23 37.63 38.04 38.45 38.87 39.29 39.72 40.15 40.58 41.02 $351 Wet Season Rice 397.14 401.16 405.22 409.32 413.46 417.65 421.87 426.14 430.46 434.82 439.22 443.66 448.15 452.69 457.27 461.90 466.58 471.30 $4,065 Dry Season Rice 76.84 77.18 77.52 77.87 78.21 78.56 78.91 79.26 79.61 79.96 80.31 80.67 81.03 81.39 81.75 82.11 82.47 82.84 $757 Vegetables 46.67 47.76 48.87 50.01 51.17 52.36 53.58 54.82 56.10 57.40 58.74 60.10 61.50 62.93 64.40 65.89 67.43 68.99 $523 636.95 643.71 650.55 657.48 664.49 671.58 678.76 686.03 693.39 700.84 708.39 716.02 723.76 731.58 739.51 747.53 755.66 763.89 $6,545 - 1.1% 2.1% 3.2% 4.3% 5.4% 6.6% 7.7% 8.9% 10.0% 11.2% 12.4% 13.6% 14.9% 16.1% 17.4% 18.6% 19.9% TRADITIONAL Cassava 92.57 92.57 92.57 92.57 92.57 92.57 92.57 92.57 92.57 92.57 92.57 92.57 92.57 92.57 92.57 92.57 92.57 92.57 $886 Maize 27.34 27.34 27.34 27.34 27.34 27.34 27.34 27.34 27.34 27.34 27.34 27.34 27.34 27.34 27.34 27.34 27.34 27.34 $262 Wet Season Rice 265.90 265.90 265.90 265.90 265.90 265.90 265.90 265.90 265.90 265.90 265.90 265.90 265.90 265.90 265.90 265.90 265.90 265.90 $2,545 Dry Season Rice 99.91 99.91 99.91 99.91 99.91 99.91 99.91 99.91 99.91 99.91 99.91 99.91 99.91 99.91 99.91 99.91 99.91 99.91 $956 Vegetables 20.55 20.55 20.55 20.55 20.55 20.55 20.55 20.55 20.55 20.55 20.55 20.55 20.55 20.55 20.55 20.55 20.55 20.55 $197 506.28 506.28 506.28 506.28 506.28 506.28 506.28 506.28 506.28 506.28 506.28 506.28 506.28 506.28 506.28 506.28 506.28 506.28 $4,845 - 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% TOTAL Cassava 174.72 175.66 176.61 177.58 178.56 179.54 180.54 181.55 182.57 183.61 184.65 185.71 186.78 187.86 188.95 190.06 191.18 192.31 $1,735 Maize 61.49 61.86 62.24 62.62 63.00 63.38 63.77 64.17 64.57 64.97 65.38 65.79 66.21 66.63 67.05 67.48 67.92 68.36 $613 Wet Season Rice 663.04 667.06 671.12 675.22 679.36 683.55 687.78 692.05 696.36 700.72 705.12 709.56 714.06 718.59 723.17 727.80 732.48 737.20 $6,610 Dry Season Rice 176.76 177.10 177.44 177.78 178.13 178.47 178.82 179.17 179.52 179.87 180.23 180.58 180.94 181.30 181.66 182.02 182.38 182.75 $1,713 Vegetables 67.23 68.31 69.42 70.56 71.72 72.91 74.13 75.38 76.65 77.96 79.29 80.66 82.06 83.49 84.95 86.45 87.98 89.55 $720 1,143.23 1,149.99 1,156.83 1,163.76 1,170.76 1,177.86 1,185.04 1,192.31 1,199.67 1,207.12 1,214.67 1,222.30 1,230.03 1,237.86 1,245.79 1,253.81 1,261.94 1,270.17 $11,391 - 0.6% 1.2% 1.8% 2.4% 3.0% 3.7% 4.3% 4.9% 5.6% 6.2% 6.9% 7.6% 8.3% 9.0% 9.7% 10.4% 11.1% 93.4% 1,067.90 1,074.21 1,080.60 1,087.07 1,093.62 1,100.25 1,106.95 1,113.75 1,120.62 1,127.58 1,134.63 1,141.76 1,148.98 1,156.29 1,163.70 1,171.19 1,178.78 1,186.47 $10,640 Share of Modern 0.56 0.56 0.56 0.56 0.57 0.57 0.57 0.58 0.58 0.58 0.58 0.59 0.59 0.59 0.59 0.60 0.60 0.60 199 Table 108: Gross margins and NPV (million $) for TFP scenario (scenario P2) GROSS MARGINS (million $) ---> Scenario T2 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 NPV MODERN Cassava 82.15 85.43 88.85 92.40 96.10 99.94 103.94 108.10 112.42 116.92 121.59 126.46 131.52 136.78 142.25 147.94 153.86 160.01 $1,037 Maize 34.16 35.52 36.95 38.42 39.96 41.56 43.22 44.95 46.75 48.62 50.56 52.58 54.69 56.88 59.15 61.52 63.98 66.54 $431 Wet Season Rice 397.14 413.02 429.54 446.72 464.59 483.18 502.50 522.60 543.51 565.25 587.86 611.37 635.83 661.26 687.71 715.22 743.83 773.58 $5,013 Dry Season Rice 76.84 79.92 83.11 86.44 89.90 93.49 97.23 101.12 105.16 109.37 113.75 118.30 123.03 127.95 133.07 138.39 143.92 149.68 $970 Vegetables 46.67 48.54 50.48 52.50 54.60 56.78 59.06 61.42 63.87 66.43 69.09 71.85 74.72 77.71 80.82 84.05 87.42 90.91 $589 636.95 662.43 688.93 716.49 745.15 774.95 805.95 838.19 871.72 906.58 942.85 980.56 1,019.78 1,060.57 1,103.00 1,147.12 1,193.00 1,240.72 $8,040 - 4.0% 8.2% 12.5% 17.0% 21.7% 26.5% 31.6% 36.9% 42.3% 48.0% 53.9% 60.1% 66.5% 73.2% 80.1% 87.3% 94.8% TRADITIONAL Cassava 92.57 90.58 88.51 86.35 84.11 81.77 79.35 76.82 74.20 71.47 68.63 65.68 62.61 59.42 56.10 52.65 49.06 45.32 $734 Maize 27.34 26.61 25.85 25.06 24.24 23.39 22.50 21.58 20.62 19.62 18.58 17.51 16.38 15.22 14.00 12.74 11.43 10.06 $206 Wet Season Rice 265.90 258.81 251.44 243.77 235.79 227.50 218.87 209.90 200.57 190.86 180.77 170.27 159.36 148.01 136.20 123.92 111.15 97.87 $2,004 Dry Season Rice 99.91 97.25 94.48 91.60 88.60 85.48 82.24 78.87 75.36 71.72 67.92 63.98 59.88 55.61 51.18 46.56 41.76 36.77 $753 Vegetables 20.55 20.01 19.44 18.84 18.23 17.59 16.92 16.22 15.50 14.75 13.97 13.16 12.32 11.44 10.53 9.58 8.59 7.57 $155 506.28 493.25 479.71 465.62 450.96 435.73 419.88 403.40 386.25 368.43 349.89 330.60 310.55 289.70 268.01 245.45 221.99 197.59 $3,851 - -2.6% -5.2% -8.0% -10.9% -13.9% -17.1% -20.3% -23.7% -27.2% -30.9% -34.7% -38.7% -42.8% -47.1% -51.5% -56.2% -61.0% TOTAL Cassava 174.72 176.01 177.35 178.75 180.20 181.72 183.29 184.92 186.62 188.39 190.23 192.14 194.13 196.20 198.35 200.59 202.91 205.33 $1,771 Maize 61.49 62.13 62.80 63.48 64.20 64.95 65.72 66.53 67.37 68.24 69.15 70.09 71.07 72.09 73.15 74.26 75.40 76.60 $637 Wet Season Rice 663.04 671.83 680.98 690.49 700.38 710.67 721.37 732.50 744.07 756.11 768.63 781.65 795.19 809.27 823.91 839.14 854.98 871.45 $7,016 Dry Season Rice 176.76 177.16 177.59 178.03 178.49 178.97 179.47 179.99 180.53 181.09 181.67 182.28 182.91 183.56 184.24 184.95 185.69 186.46 $1,723 Vegetables 67.23 68.55 69.92 71.34 72.83 74.37 75.97 77.64 79.38 81.18 83.06 85.01 87.04 89.15 91.35 93.63 96.01 98.48 $744 1,143.23 1,155.68 1,168.63 1,182.10 1,196.11 1,210.68 1,225.83 1,241.58 1,257.97 1,275.01 1,292.73 1,311.17 1,330.34 1,350.27 1,371.01 1,392.57 1,414.99 1,438.32 $11,891 - 1.1% 2.2% 3.4% 4.6% 5.9% 7.2% 8.6% 10.0% 11.5% 13.1% 14.7% 16.4% 18.1% 19.9% 21.8% 23.8% 25.8% 93.4% 1,067.90 1,079.53 1,091.63 1,104.21 1,117.29 1,130.90 1,145.05 1,159.77 1,175.08 1,190.99 1,207.55 1,224.77 1,242.67 1,261.30 1,280.66 1,300.81 1,321.75 1,343.54 $11,107 Share of Modern 0.56 0.57 0.59 0.61 0.62 0.64 0.66 0.68 0.69 0.71 0.73 0.75 0.77 0.79 0.80 0.82 0.84 0.86 200 Table 109: Gross margins and NPV (million $) for TFP scenario (scenario P3) GROSS MARGINS (million $) ---> Scenario T3 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 NPV MODERN Cassava 82.15 86.41 90.90 95.62 100.59 105.81 111.31 117.09 123.17 129.57 136.30 143.38 150.82 158.66 166.90 175.57 184.68 194.28 $1,132 Maize 34.16 35.91 37.75 39.68 41.72 43.86 46.11 48.47 50.95 53.57 56.31 59.20 62.23 65.42 68.78 72.30 76.01 79.90 $468 Wet Season Rice 397.14 417.20 438.28 460.43 483.69 508.13 533.81 560.78 589.11 618.88 650.15 683.00 717.51 753.76 791.85 831.86 873.89 918.04 $5,417 Dry Season Rice 76.84 80.27 83.85 87.59 91.50 95.58 99.84 104.30 108.95 113.81 118.88 124.19 129.73 135.51 141.56 147.87 154.47 161.36 $1,003 Vegetables 46.67 49.67 52.86 56.25 59.86 63.70 67.79 72.14 76.77 81.70 86.95 92.53 98.47 104.79 111.51 118.67 126.29 134.39 $706 636.95 669.46 703.64 739.57 777.36 817.08 858.85 902.77 948.96 997.52 1,048.59 1,102.29 1,158.76 1,218.14 1,280.59 1,346.27 1,415.34 1,487.98 $8,727 - 5.1% 10.5% 16.1% 22.0% 28.3% 34.8% 41.7% 49.0% 56.6% 64.6% 73.1% 81.9% 91.2% 101.0% 111.4% 122.2% 133.6% TRADITIONAL Cassava 92.57 90.58 88.51 86.35 84.11 81.77 79.35 76.82 74.20 71.47 68.63 65.68 62.61 59.42 56.10 52.65 49.06 45.32 $734 Maize 27.34 26.61 25.85 25.06 24.24 23.39 22.50 21.58 20.62 19.62 18.58 17.51 16.38 15.22 14.00 12.74 11.43 10.06 $206 Wet Season Rice 265.90 258.81 251.44 243.77 235.79 227.50 218.87 209.90 200.57 190.86 180.77 170.27 159.36 148.01 136.20 123.92 111.15 97.87 $2,004 Dry Season Rice 99.91 97.25 94.48 91.60 88.60 85.48 82.24 78.87 75.36 71.72 67.92 63.98 59.88 55.61 51.18 46.56 41.76 36.77 $753 Vegetables 20.55 20.01 19.44 18.84 18.23 17.59 16.92 16.22 15.50 14.75 13.97 13.16 12.32 11.44 10.53 9.58 8.59 7.57 $155 506.28 493.25 479.71 465.62 450.96 435.73 419.88 403.40 386.25 368.43 349.89 330.60 310.55 289.70 268.01 245.45 221.99 197.59 $3,851 - -2.6% -5.2% -8.0% -10.9% -13.9% -17.1% -20.3% -23.7% -27.2% -30.9% -34.7% -38.7% -42.8% -47.1% -51.5% -56.2% -61.0% TOTAL Cassava 174.72 176.99 179.41 181.97 184.69 187.59 190.66 193.91 197.37 201.04 204.93 209.06 213.44 218.08 223.00 228.21 233.74 239.60 $1,866 Maize 61.49 62.52 63.60 64.75 65.96 67.25 68.61 70.05 71.57 73.19 74.90 76.70 78.62 80.64 82.78 85.04 87.44 89.97 $674 Wet Season Rice 663.04 676.01 689.72 704.19 719.48 735.63 752.67 770.67 789.68 809.74 830.92 853.27 876.87 901.77 928.05 955.78 985.04 1,015.91 $7,420 Dry Season Rice 176.76 177.52 178.33 179.19 180.10 181.06 182.08 183.16 184.31 185.52 186.81 188.17 189.60 191.13 192.73 194.43 196.23 198.13 $1,756 Vegetables 67.23 69.67 72.29 75.09 78.09 81.29 84.71 88.37 92.28 96.45 100.92 105.69 110.78 116.23 122.04 128.25 134.88 141.96 $861 1,143.23 1,162.71 1,183.34 1,205.19 1,228.32 1,252.81 1,278.73 1,306.17 1,335.21 1,365.95 1,398.47 1,432.89 1,469.31 1,507.84 1,548.60 1,591.72 1,637.33 1,685.57 $12,578 1.7% 3.5% 5.4% 7.4% 9.6% 11.9% 14.3% 16.8% 19.5% 22.3% 25.3% 28.5% 31.9% 35.5% 39.2% 43.2% 47.4% 93.4% 1,067.90 1,086.10 1,105.37 1,125.77 1,147.38 1,170.25 1,194.47 1,220.10 1,247.23 1,275.94 1,306.32 1,338.47 1,372.49 1,408.48 1,446.56 1,486.83 1,529.44 1,574.50 $11,749 Share of Modern 0.56 0.58 0.59 0.61 0.63 0.65 0.67 0.69 0.71 0.73 0.75 0.77 0.79 0.81 0.83 0.85 0.86 0.88 201 Table 110: Summary of gross margins for TFP scenarios Gross Margins in Gross Margins in 2030 ($ NPV at CPI ($ 2013 ($ million) Million) Million) % Change NPV Baseline 1,067.90 1,067.90 10,220 No Move + Efficiency 1,067.90 1,186.47 10,640 4.1% Move from Traditional to Modern 1,067.90 1,343.54 11,107 8.7% Move + Efficiency 1,067.90 1,574.50 11,749 15.0% Table 111: Examples of costs of irrigation investment (per unit and total investment) Examples of irrigation costs Cost in Newly Irrigated Areas (ha) Total Investment (Million $) with Scenario T4 $/ha South Asia 1,400 454,543 636.36 Average 2,500 454,543 1,136.36 East Asia 4,000 454,543 1,818.17 202 Table 112: Demand for labor and return to labor for TFP scenarios 203 Table 113: Demand for fertilizers for TFP scenarios 204 Table 114: Cultivated land and gross margins for scenario P4 (TFP through irrigation) GROSS MARGINS (million $) Baseline 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 Cultivated Areas 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 Cassava 337,800 337,800 337,800 337,800 337,800 337,800 337,800 337,800 337,800 337,800 337,800 337,800 337,800 337,800 337,800 337,800 337,800 337,800 Maize 215,442 215,442 215,442 215,442 215,442 215,442 215,442 215,442 215,442 215,442 215,442 215,442 215,442 215,442 215,442 215,442 215,442 215,442 Wet Season Rice 2,484,832 2,484,832 2,484,832 2,484,832 2,484,832 2,484,832 2,484,832 2,484,832 2,484,832 2,484,832 2,484,832 2,484,832 2,484,832 2,484,832 2,484,832 2,484,832 2,484,832 2,484,832 Dry Season Rice 495,465 495,465 495,465 495,465 495,465 495,465 495,465 495,465 495,465 495,465 495,465 495,465 495,465 495,465 495,465 495,465 495,465 495,465 Vegetables 54,155 54,155 54,155 54,155 54,155 54,155 54,155 54,155 54,155 54,155 54,155 54,155 54,155 54,155 54,155 54,155 54,155 54,155 3,587,694 3,587,694 3,587,694 3,587,694 3,587,694 3,587,694 3,587,694 3,587,694 3,587,694 3,587,694 3,587,694 3,587,694 3,587,694 3,587,694 3,587,694 3,587,694 3,587,694 3,587,694 Cultivated Land - 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% Scenario T4 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 Cassava 337,800 337,800 337,800 337,800 337,800 337,800 337,800 337,800 337,800 337,800 337,800 337,800 337,800 337,800 337,800 337,800 337,800 337,800 Maize 215,442 215,442 215,442 215,442 215,442 215,442 215,442 215,442 215,442 215,442 215,442 215,442 215,442 215,442 215,442 215,442 215,442 215,442 Wet Season Rice 2,484,832 2,435,135 2,386,433 2,338,704 2,291,930 2,246,091 2,201,169 2,157,146 2,114,003 2,071,723 2,030,289 1,989,683 1,949,889 1,910,891 1,872,674 1,835,220 1,798,516 1,762,545 Dry Season Rice 495,465 594,858 692,264 787,721 881,269 972,946 1,062,790 1,150,837 1,237,123 1,321,683 1,404,552 1,485,763 1,565,351 1,643,346 1,719,782 1,794,689 1,868,098 1,940,038 Vegetables 54,155 54,155 54,155 54,155 54,155 54,155 54,155 54,155 54,155 54,155 54,155 54,155 54,155 54,155 54,155 54,155 54,155 54,155 3,587,694 3,637,391 3,686,093 3,733,822 3,780,596 3,826,435 3,871,357 3,915,380 3,958,523 4,000,803 4,042,237 4,082,843 4,122,637 4,161,635 4,199,852 4,237,306 4,274,010 4,309,981 Gross Margins 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 Scenario T4 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 Cassava 170.82 170.82 170.82 170.82 170.82 170.82 170.82 170.82 170.82 170.82 170.82 170.82 170.82 170.82 170.82 170.82 170.82 170.82 Maize 65.48 65.48 65.48 65.48 65.48 65.48 65.48 65.48 65.48 65.48 65.48 65.48 65.48 65.48 65.48 65.48 65.48 65.48 Wet Season Rice 609.70 597.51 585.56 573.85 562.37 551.12 540.10 529.30 518.71 508.34 498.17 488.21 478.44 468.88 459.50 450.31 441.30 432.48 Dry Season Rice 183.44 220.23 256.30 291.64 326.27 360.21 393.48 426.07 458.02 489.33 520.01 550.07 579.54 608.42 636.71 664.45 691.63 718.26 Vegetables 75.47 75.47 75.47 75.47 75.47 75.47 75.47 75.47 75.47 75.47 75.47 75.47 75.47 75.47 75.47 75.47 75.47 75.47 1,104.90 1,129.51 1,153.62 1,177.25 1,200.41 1,223.10 1,245.34 1,267.14 1,288.50 1,309.43 1,329.94 1,350.04 1,369.75 1,389.05 1,407.98 1,426.52 1,444.69 1,462.50 0.0% 2.2% 4.4% 6.5% 8.6% 10.7% 12.7% 14.7% 16.6% 18.5% 20.4% 22.2% 24.0% 25.7% 27.4% 29.1% 30.8% 32.4% 205 Table 115: Gross margins on competitiveness (scenarios C1 through C6) 206 Continued… Table 116: Summary of gross margins and NPV for Competitiveness scenarios (C1 through C6) Gross Margins 2019 Gross Margins 2030 NPV at CPI (million $) (million $) (million $) NPV Changes Baseline 1,068 1,068 10,220 C1: Equal Increase 1,147 1,244 10,986 7.5% C2:↑ Output price half of ↑ inputs 629 (1,577) 3,761 -63.2% C3: Wage Rate ↑ 50% 569 (1,348) 3,655 -64.2% C4: Cost of Services ↑ 25% 983 799 9,290 -9.1% C5: Output price ↓ 15% 794 358 7,449 -27.1% C6: Output price ↑ 15% 1,650 3,020 16,739 63.8% 207 Table 117: Return to labor for Competitiveness scenarios (C1 through C6) 208 Continued… 209 Table 118: Demand for labor and demand for fertilizers with Competitiveness scenarios Assumptions: Price change will not affect demand for labor (Inelastic demand) - Changes are due to change on cultivated areas DEMAND FOR LABOR (Million days) All Scenarii 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 DEMAND FOR LABOR (Million days) 0 1 2 3 4 5 6 7 8 9 9 9 9 9 9 9 9 9 Cassava 16.48 16.48 16.48 16.48 16.48 16.48 16.48 16.48 16.48 16.48 16.48 16.48 16.48 16.48 16.48 16.48 16.48 16.48 Maize 6.75 6.75 6.75 6.75 6.75 6.75 6.75 6.75 6.75 6.75 6.75 6.75 6.75 6.75 6.75 6.75 6.75 6.75 Wet Season Rice 120.00 120.00 120.00 120.00 120.00 120.00 120.00 120.00 120.00 120.00 120.00 120.00 120.00 120.00 120.00 120.00 120.00 120.00 Dry Season Rice 13.75 13.75 13.75 13.75 13.75 13.75 13.75 13.75 13.75 13.75 13.75 13.75 13.75 13.75 13.75 13.75 13.75 13.75 Vegetables 9.53 9.53 9.53 9.53 9.53 9.53 9.53 9.53 9.53 9.53 9.53 9.53 9.53 9.53 9.53 9.53 9.53 9.53 166.51 166.51 166.51 166.51 166.51 166.51 166.51 166.51 166.51 166.51 166.51 166.51 166.51 166.51 166.51 166.51 166.51 166.51 - 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% Assumptions: Price change will not affect demand for fertilizers (Inelastic demand) - Changes are due to change on cultivated areas DEMAND FOR FERTILIZERS (tons) All Scenarii 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 DEMAND FOR FERTILIZERS (tons) 0 1 2 3 4 5 6 7 8 9 9 9 9 9 9 9 9 9 Cassava 4,578 4,578 4,578 4,578 4,578 4,578 4,578 4,578 4,578 4,578 4,578 4,578 4,578 4,578 4,578 4,578 4,578 4,578 Maize 11,507 11,507 11,507 11,507 11,507 11,507 11,507 11,507 11,507 11,507 11,507 11,507 11,507 11,507 11,507 11,507 11,507 11,507 Wet Season Rice 294,408 294,408 294,408 294,408 294,408 294,408 294,408 294,408 294,408 294,408 294,408 294,408 294,408 294,408 294,408 294,408 294,408 294,408 Dry Season Rice 135,722 135,722 135,722 135,722 135,722 135,722 135,722 135,722 135,722 135,722 135,722 135,722 135,722 135,722 135,722 135,722 135,722 135,722 Vegetables 18,646 18,646 18,646 18,646 18,646 18,646 18,646 18,646 18,646 18,646 18,646 18,646 18,646 18,646 18,646 18,646 18,646 18,646 464,861 464,861 464,861 464,861 464,861 464,861 464,861 464,861 464,861 464,861 464,861 464,861 464,861 464,861 464,861 464,861 464,861 464,861 - 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 210 Table 119: Characteristics of rice production: fragrant vs. IRRI Revenue Yield by technology use (tons/ha) Output Prices by technology use ($/ton) Revenue by technology use and total 150% FRAGRANT OTHER TOTAL FRAGRANT OTHER TOTAL FRAGRANT OTHER TOTAL FRAGRANT Wet Season Rice 2.50 2.91 2.89 444.52 254.01 261.43 1,111.29 739.30 757.90 1,587.56 Dry Season Rice 3.50 4.87 4.80 352.14 201.22 206.73 1,232.47 979.62 992.27 1,760.68 Seeds Seeds by farm size (unit/ha) Seed Prices by farm size ($/unit) Costs of seeds by farm size and total FRAGRANT OTHER TOTAL FRAGRANT OTHER TOTAL FRAGRANT OTHER TOTAL FRAGRANT Wet Season Rice 38.82 38.82 38.82 1.32 0.63 0.66 51.24 24.26 25.61 51.24 Dry Season Rice 146.07 146.07 146.07 1.48 0.70 0.74 216.18 102.40 108.08 216.18 Fertilizer Fertilizer use by farm size (kg/ha) Fertilizer Prices by farm size ($/kg) Costs of fertilizers by farm size and total FRAGRANT IRRI TOTAL FRAGRANT IRRI TOTAL FRAGRANT IRRI TOTAL FRAGRANT Wet Season Rice 118.48 118.48 118.48 0.56 0.56 0.56 66.35 66.35 66.35 66.35 Dry Season Rice 273.93 273.93 273.93 0.56 0.56 0.56 153.40 153.40 153.40 153.40 Other Inputs Other input Costs ($/ha) FRAGRANT IRRI TOTAL FRAGRANT Wet Season Rice 26.88 26.88 26.88 26.88 Dry Season Rice 27.73 27.73 27.73 27.73 Labor Labor use by technology (days/ha) Wage rate by technology ($/day) Costs of labor by technology and total FRAGRANT IRRI TOTAL FRAGRANT IRRI TOTAL FRAGRANT IRRI TOTAL FRAGRANT Wet Season Rice 48.29 48.29 48.29 4.56 4.56 4.56 220.21 220.21 220.21 220.21 Dry Season Rice 27.75 27.75 27.75 4.56 4.56 4.56 126.54 126.54 126.54 126.54 Services Service Costs ($/ha) FRAGRANT IRRI TOTAL FRAGRANT Wet Season Rice 171.10 171.10 171.10 171.10 Dry Season Rice 206.29 206.29 206.29 206.29 Irrigation Irrigation Costs ($/ha) FRAGRANT IRRI TOTAL FRAGRANT Wet Season Rice - Dry Season Rice 74.69 74.69 74.69 74.69 Gross Margins Gross Margins ($/ha) FRAGRANT IRRI TOTAL FRAGRANT Wet Season Rice 575.51 230.50 247.75 1,051.78 Dry Season Rice 427.64 288.58 295.53 955.84 211 Table 120: Cultivated land areas for fragrant rice Income Growth scenarios 212 Table 121: Gross margins for fragrant rice Income Growth scenarios 213 Table 122: Summary of gross margins for fragrant rice Income Growth scenarios Gross Margins 2014 Gross Margins 2030 NPV at CPI (million $) (million $) (million $) % Change NPV Baseline 762.04 762.04 7,293 Fragrant 10%, Price 100% 763.99 808.94 7,443 2.1% Fragrant 25%, price 100% 766.63 946.21 7,774 6.6% Table 123: Demand for labor, return to labor and demand for fertilizers for fragrant rice Income Growth scenarios DEMAND FOR LABOR (Million days) Baseline 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 Scenario R1 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 Wet Season Rice 120.30 120.30 120.30 120.30 120.30 120.30 120.30 120.30 120.30 120.30 120.30 120.30 120.30 120.30 120.30 120.30 120.30 120.30 Dry Season Rice 13.78 13.78 13.78 13.78 13.78 13.78 13.78 13.78 13.78 13.78 13.78 13.78 13.78 13.78 13.78 13.78 13.78 13.78 134.08 134.08 134.08 134.08 134.08 134.08 134.08 134.08 134.08 134.08 134.08 134.08 134.08 134.08 134.08 134.08 134.08 134.08 AVERAGE RETURN TO LABOR ($/DAY) 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 Baseline 5.68 5.68 5.68 5.68 5.68 5.68 5.68 5.68 5.68 5.68 5.68 5.68 5.68 5.68 5.68 5.68 5.68 5.68 Fragrant 10%, Price 100% 5.68 5.70 5.71 5.73 5.75 5.76 5.78 5.80 5.82 5.84 5.86 5.88 5.90 5.93 5.95 5.98 6.01 6.03 Fragrant 25%, price 100% 5.68 5.72 5.76 5.80 5.84 5.89 5.95 6.01 6.07 6.15 6.23 6.31 6.41 6.52 6.63 6.76 6.90 7.06 DEMAND FOR FERTILIZERS (Tons) Baseline 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 Wet Season Rice 294,408 294,408 294,408 294,408 294,408 294,408 294,408 294,408 294,408 294,408 294,408 294,408 294,408 294,408 294,408 294,408 294,408 294,408 Dry Season Rice 135,722 135,722 135,722 135,722 135,722 135,722 135,722 135,722 135,722 135,722 135,722 135,722 135,722 135,722 135,722 135,722 135,722 135,722 430,130 430,130 430,130 430,130 430,130 430,130 430,130 430,130 430,130 430,130 430,130 430,130 430,130 430,130 430,130 430,130 430,130 430,130 Land under Cultivation (Ha) 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 Fragrant (10%) 149,015 149,015 155,273 161,795 168,590 175,671 183,049 190,737 198,748 207,096 215,794 224,857 234,301 244,142 254,396 265,080 276,214 287,815 Other (90%) 2,831,282 2,831,282 2,825,024 2,818,502 2,811,707 2,804,626 2,797,248 2,789,560 2,781,549 2,773,201 2,764,503 2,755,440 2,745,996 2,736,155 2,725,901 2,715,217 2,704,083 2,692,482 Fragrant (25%) 149,015 163,767 179,980 197,798 217,380 238,901 262,552 288,545 317,111 348,505 383,007 420,924 462,596 508,393 558,724 614,038 674,827 741,635 Other (75%) 2,831,282 2,816,530 2,800,317 2,782,499 2,762,917 2,741,396 2,717,745 2,691,752 2,663,186 2,631,792 2,597,290 2,559,373 2,517,701 2,471,904 2,421,573 2,366,259 2,305,470 2,238,662 214 Table 124: Cultivated land areas for cassava processing Income Growth scenarios 215 Table 125: Gross margins for cassava processing Income Growth scenarios (million $) 216 Table 126: Comparison between fresh cassava and dry chips production 2013 Level Cultivated land areas (Ha) % Processed Cassava 337,800 5% Revenue Yield by technology use (tons/ha) Output Prices by technology use ($/ton) Revenue by technology use and total PROCESSED OTHER PROCESSED OTHER TOTAL PROCESSED OTHER TOTAL Cassava 9.15 16.90 158.60 72.32 1,451.19 1,222.21 Costs w/o labor PROCESSED OTHER TOTAL PROCESSED OTHER TOTAL PROCESSED OTHER TOTAL Cassava 581.00 575.00 1.00 1.00 581.00 575.00 Labor Costs PROCESSED OTHER TOTAL PROCESSED OTHER TOTAL PROCESSED OTHER TOTAL Cassava 220.00 196.78 1.00 1.00 220.00 196.78 Gross Margins Gross Margins ($/ha) PROCESSED OTHER TOTAL Cassava 650.19 450.43 217 Table 127: Demand for labor and return to labor for cassava processing Income Growth scenarios DEMAND FOR LABOR (Million days) Baseline 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 Baseline 14.66 14.66 14.66 14.66 14.66 14.66 14.66 14.66 14.66 14.66 14.66 14.66 14.66 14.66 14.66 14.66 14.66 14.66 Processed 5%, Price 50% higher 14.66 14.67 14.67 14.67 14.68 14.68 14.69 14.69 14.70 14.70 14.71 14.71 14.72 14.72 14.73 14.74 14.74 14.75 Processed 10%, Price 50% higher 14.66 14.67 14.68 14.69 14.70 14.72 14.73 14.74 14.76 14.78 14.80 14.82 14.84 14.87 14.90 14.93 14.97 15.01 AVERAGE RETURN TO LABOR ($/DAY) 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 Baseline 10.61 10.60 10.60 10.60 10.60 10.59 10.59 10.59 10.58 10.58 10.58 10.57 10.57 10.56 10.56 10.55 10.55 10.54 Processed 5%, Price 50% higher 10.61 10.61 10.62 10.63 10.64 10.64 10.65 10.66 10.67 10.68 10.69 10.70 10.71 10.72 10.74 10.75 10.76 10.78 Processed 10%, Price 50% higher 10.61 10.63 10.65 10.67 10.70 10.73 10.76 10.80 10.84 10.89 10.94 10.99 11.05 11.12 11.19 11.27 11.36 11.45 Land under Cultivation (Ha) 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 Cultivated land: PROCESSED 10% 16,890 17,599 18,339 19,109 19,911 20,748 21,619 22,527 23,473 24,459 25,486 26,557 27,672 28,834 30,045 31,307 32,622 33,992 Cultivated land: OTHER 90% 320,910 320,201 319,461 318,691 317,889 317,052 316,181 315,273 314,327 313,341 312,314 311,243 310,128 308,966 307,755 306,493 305,178 303,808 5.3% 5.5% 5.7% 6.0% 6.3% 6.5% 6.8% 7.1% 7.5% 7.8% 8.2% 8.5% 8.9% 9.3% 9.8% 10.2% 10.7% 11.2% Cultivated land: PROCESSED 25% 16,890 18,562 20,400 22,419 24,639 27,078 29,759 32,705 35,943 39,501 43,412 47,709 52,433 57,624 63,328 69,598 76,488 84,060 Cultivated land: OTHER 75% 320,910 319,238 317,400 315,381 313,161 310,722 308,041 305,095 301,857 298,299 294,388 290,091 285,367 280,176 274,472 268,202 261,312 253,740 5.3% 5.8% 6.4% 7.1% 7.9% 8.7% 9.7% 10.7% 11.9% 13.2% 14.7% 16.4% 18.4% 20.6% 23.1% 25.9% 29.3% 33.1% 218 Table 128: Distribution and farm budget by farm size 2012 Cultivated Areas % of land cultivated by farm size Total cultivated land by farm size (ha) Cultivated Past growth Rate SMALL MEDIUM LARGE TOTAL SMALL MEDIUM LARGE land (ha) Cassava 57% 36% 7% 100% 192,546 121,608 23,646 337,800 33.1% 2.64 6.57 Maize 47% 41% 12% 100% 101,258 88,331 25,853 215,442 9.7% 0.75 2.34 Wet Season Rice 60% 31% 9% 100% 1,490,899 770,298 223,635 2,484,832 2.6% 0.56 1.28 Dry Season Rice 60% 31% 9% 100% 297,279 153,594 44,592 495,465 6.8% 0.71 2.28 Vegetables 75% 21% 4% 100% 40,616 11,373 2,166 54,155 7.8% 0.19 0.84 2,122,598 1,145,204 319,892 3,587,694 1.0 5 Revenue Yield by farm size (tons/ha) Output Prices by farm size ($/ton) Revenue by farm size and total ($/ha) SMALL MEDIUM LARGE AVERAGE SMALL MEDIUM LARGE AVERAGE SMALL MEDIUM LARGE TOTAL Cassava 19.65 18.07 17.79 18.95 69.58 64.40 73.99 68.02 1,367.17 1,163.69 1,316.24 1,290.35 Maize 3.73 3.73 4.32 3.80 210.43 202.28 161.06 201.16 784.91 754.49 695.77 761.74 Wet Season Rice 2.68 2.89 3.12 2.78 285.43 266.47 229.43 274.51 764.94 770.10 715.82 762.12 Dry Season Rice 4.70 4.48 5.18 4.68 211.60 210.16 199.94 210.10 994.52 941.51 1,035.68 981.79 Vegetables 9.74 6.00 6.67 8.83 331.28 193.75 160.94 295.58 3,226.66 1,162.48 1,073.45 2,707.06 Seeds Seeds by farm size (unit/ha) Seed Prices by farm size ($/unit) Costs of seeds by farm size and total ($/ha) SMALL MEDIUM LARGE AVERAGE SMALL MEDIUM LARGE AVERAGE SMALL MEDIUM LARGE TOTAL Cassava 2,614.55 1,832.86 2,506.25 2,325.56 0.09 0.11 0.10 0.10 236.53 200.12 242.66 223.85 Maize 23.82 23.52 20.57 23.31 3.05 3.29 3.87 3.25 72.57 77.47 79.65 75.43 Wet Season Rice 53.37 40.46 106.83 54.18 0.47 0.48 0.36 0.46 24.85 19.62 38.50 24.46 Dry Season Rice 261.29 233.13 226.19 249.40 0.41 0.44 0.50 0.43 108.31 101.88 113.44 106.78 Vegetables 12.53 1.78 2.51 9.87 12.43 15.80 62.48 15.14 155.80 28.07 156.73 129.01 Fertilizer Fertilizer use by farm size (kg/ha) Fertilizer Prices by farm size ($/kg) Costs of fertilizers by farm size and total ($/ha) SMALL MEDIUM LARGE AVERAGE SMALL MEDIUM LARGE AVERAGE SMALL MEDIUM LARGE TOTAL Cassava 15.91 - 11.93 9.90 0.75 0.68 0.48 11.99 - 8.17 7.41 Maize 107.91 12.97 14.29 57.75 0.69 0.60 0.66 0.65 74.16 7.74 9.41 39.15 Wet Season Rice 126.75 106.42 89.42 117.08 0.57 0.61 0.70 0.59 72.17 64.77 62.98 69.04 Dry Season Rice 235.92 212.29 295.29 233.94 0.56 0.60 0.65 0.58 133.29 127.40 192.16 136.76 Vegetables 357.87 88.61 112.50 291.51 0.63 0.62 0.70 0.63 224.39 54.52 78.75 182.89 Other Inputs Other input Costs ($/ha) SMALL MEDIUM LARGE AVERAGE Cassava 35.14 32.50 43.49 34.77 Maize 38.77 26.15 24.92 31.93 Wet Season Rice 38.83 24.15 18.88 32.48 Dry Season Rice 31.98 19.87 31.41 28.17 Vegetables 292.85 31.79 59.77 228.70 219 Continued… Labor Labor use by farm size (days/ha) Wage rate by farm size ($/day) Costs of labor by farm size and total ($/ha) SMALL MEDIUM LARGE AVERAGE SMALL MEDIUM LARGE AVERAGE SMALL MEDIUM LARGE TOTAL Cassava 46.23 58.44 43.90 50.46 4.56 4.56 4.56 4.56 210.82 266.48 200.18 230.1128 Maize 37.71 32.15 24.63 33.86 4.56 4.56 4.56 4.56 171.98 146.61 112.33 154.4203 Wet Season Rice 63.05 50.82 26.63 55.98 4.56 4.56 4.56 4.56 287.51 231.75 121.41 255.2754 Dry Season Rice 44.87 30.81 11.72 37.53 4.56 4.56 4.56 4.56 204.60 140.48 53.44 171.1184 Vegetables 201.85 52.83 73.33 165.41 4.56 4.56 4.56 4.56 920.43 240.89 334.38 754.2846 Services Service Costs ($/ha) SMALL MEDIUM LARGE AVERAGE Cassava 328.56 226.88 309.52 295.32 Maize 183.54 160.60 139.51 160.68 Wet Season Rice 134.73 160.03 234.16 171.10 Dry Season Rice 173.81 212.15 226.33 206.29 Vegetables 61.24 62.79 136.67 66.45 Gross Margins Gross Margins ($/ha) SMALL MEDIUM LARGE AVERAGE Cassava 544.13 437.71 512.23 498.89 Maize 243.90 335.92 329.95 300.13 Wet Season Rice 206.85 269.79 239.89 209.76 Dry Season Rice 306.82 283.48 297.51 283.57 Vegetables 1,571.96 744.42 307.16 1,345.72 220 Table 129: Gross margins with Farm Size Distribution scenario (baseline) GROSS MARGINS (million $) Baseline 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 SMALL 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 Cassava 104.77 104.77 104.77 104.77 104.77 104.77 104.77 104.77 104.77 104.77 104.77 104.77 104.77 104.77 104.77 104.77 104.77 104.77 Maize 24.70 24.70 24.70 24.70 24.70 24.70 24.70 24.70 24.70 24.70 24.70 24.70 24.70 24.70 24.70 24.70 24.70 24.70 Wet Season Rice 308.40 308.40 308.40 308.40 308.40 308.40 308.40 308.40 308.40 308.40 308.40 308.40 308.40 308.40 308.40 308.40 308.40 308.40 Dry Season Rice 91.21 91.21 91.21 91.21 91.21 91.21 91.21 91.21 91.21 91.21 91.21 91.21 91.21 91.21 91.21 91.21 91.21 91.21 Vegetables 63.85 63.85 63.85 63.85 63.85 63.85 63.85 63.85 63.85 63.85 63.85 63.85 63.85 63.85 63.85 63.85 63.85 63.85 592.92 592.92 592.92 592.92 592.92 592.92 592.92 592.92 592.92 592.92 592.92 592.92 592.92 592.92 592.92 592.92 592.92 592.92 - 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% MEDIUM Cassava 53.23 53.23 53.23 53.23 53.23 53.23 53.23 53.23 53.23 53.23 53.23 53.23 53.23 53.23 53.23 53.23 53.23 53.23 Maize 29.67 29.67 29.67 29.67 29.67 29.67 29.67 29.67 29.67 29.67 29.67 29.67 29.67 29.67 29.67 29.67 29.67 29.67 Wet Season Rice 207.82 207.82 207.82 207.82 207.82 207.82 207.82 207.82 207.82 207.82 207.82 207.82 207.82 207.82 207.82 207.82 207.82 207.82 Dry Season Rice 43.54 43.54 43.54 43.54 43.54 43.54 43.54 43.54 43.54 43.54 43.54 43.54 43.54 43.54 43.54 43.54 43.54 43.54 Vegetables 8.47 8.47 8.47 8.47 8.47 8.47 8.47 8.47 8.47 8.47 8.47 8.47 8.47 8.47 8.47 8.47 8.47 8.47 342.73 342.73 342.73 342.73 342.73 342.73 342.73 342.73 342.73 342.73 342.73 342.73 342.73 342.73 342.73 342.73 342.73 342.73 - 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% LARGE Cassava 12.11 12.11 12.11 12.11 12.11 12.11 12.11 12.11 12.11 12.11 12.11 12.11 12.11 12.11 12.11 12.11 12.11 12.11 Maize 8.53 8.53 8.53 8.53 8.53 8.53 8.53 8.53 8.53 8.53 8.53 8.53 8.53 8.53 8.53 8.53 8.53 8.53 Wet Season Rice 53.65 53.65 53.65 53.65 53.65 53.65 53.65 53.65 53.65 53.65 53.65 53.65 53.65 53.65 53.65 53.65 53.65 53.65 Dry Season Rice 13.27 13.27 13.27 13.27 13.27 13.27 13.27 13.27 13.27 13.27 13.27 13.27 13.27 13.27 13.27 13.27 13.27 13.27 Vegetables 0.67 0.67 0.67 0.67 0.67 0.67 0.67 0.67 0.67 0.67 0.67 0.67 0.67 0.67 0.67 0.67 0.67 0.67 88.22 88.22 88.22 88.22 88.22 88.22 88.22 88.22 88.22 88.22 88.22 88.22 88.22 88.22 88.22 88.22 88.22 88.22 - 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% Gross Margins: OVERALL Cassava 170.11 170.11 170.11 170.11 170.11 170.11 170.11 170.11 170.11 170.11 170.11 170.11 170.11 170.11 170.11 170.11 170.11 170.11 Maize 62.90 62.90 62.90 62.90 62.90 62.90 62.90 62.90 62.90 62.90 62.90 62.90 62.90 62.90 62.90 62.90 62.90 62.90 Wet Season Rice 569.86 569.86 569.86 569.86 569.86 569.86 569.86 569.86 569.86 569.86 569.86 569.86 569.86 569.86 569.86 569.86 569.86 569.86 Dry Season Rice 148.02 148.02 148.02 148.02 148.02 148.02 148.02 148.02 148.02 148.02 148.02 148.02 148.02 148.02 148.02 148.02 148.02 148.02 Vegetables 72.98 72.98 72.98 72.98 72.98 72.98 72.98 72.98 72.98 72.98 72.98 72.98 72.98 72.98 72.98 72.98 72.98 72.98 1,023.87 1,023.87 1,023.87 1,023.87 1,023.87 1,023.87 1,023.87 1,023.87 1,023.87 1,023.87 1,023.87 1,023.87 1,023.87 1,023.87 1,023.87 1,023.87 1,023.87 1,023.87 - 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 221 Table 130: Gross margins with Farm Size Distribution scenario (I5) 222 Table 131: Gross margins with Farm Size Distribution scenario (I6) 223 Table 132: Summary of gross margins from changes on Farm Size Distribution ($ million) Gross Margins in Gross Margins in Farm Size 2013 (Million $) 2030 (Million $) % Change NPV Change NPV Baseline Small 592.92 592.92 5,674 Scenario I5 Small 592.92 499.80 -15.7% 5,319 -6.3% Scenario I6 Small 592.92 499.80 -15.7% 5,319 -6.3% Baseline Medium 342.73 342.73 3,280 Scenario I5 Medium 342.73 392.60 14.6% 3,470 5.8% Scenario I6 Medium 342.73 899.84 162.6% 4,977 51.7% Baseline Large 88.22 88.22 844 Scenario I5 Large 88.22 134.60 52.6% 1,021 21.0% Scenario I6 Large 88.22 308.51 249.7% 1,497 77.3% Baseline Total 1,023.87 1,023.87 9,799 Scenario I5 Total 1,023.87 1,027.00 0.3% 9,811 0.1% Scenario I6 Total 1,023.87 1,708.15 66.8% 11,793 20.4% 224 Table 133: Demand for labor and return to labor for Farm Size Distribution scenarios 225