THE ECONOMICS OF CLOVE FARMING IN INDONESIA Health, Population, and Nutrition Global Practice Cover photos (clockwise from left): Senior farmer standing in his farm. Photo by Jacob Ammentorp Lund / iStock. Beautiful Panoramic View of Gunung Batur Vlcano in Bali. Photo by AleksandarGeorgiev / iStock. Bag of Cloves. Photo by Bveth / iStock. Clove tree with spicy raw flowers and sticks. Photo by Bicho_raro / iStock. THE ECONOMICS OF CLOVE FARMING IN INDONESIA Health, Population, and Nutrition Global Practice SMOKING HAS BEEN A MAJOR CONTRIBUTOR O DISEASE BURDEN INDONESIA. SMO PREVALENCE AMONG WORKIN AGE INDIVIDUAL HAS EXCEEDED 3 PERCENT SINCE 2 2 // The Economics of Tobacco Taxation LIST OF ACRONYMS APCI Indonesia Clove Farmer Association BPJS National Health Insurance System FAO Food and Agriculture Organization’s HRW Human Rights Watch ILO International Labour Organization IAKMI Indonesian Public Health Association KIS Indonesia Health Card KKS Family Welfare Card KPS Social Security Card PPP Purchasing Power Parity USD United States (US) Dollars 3 EXECUTIVE SUMMARY Indonesia is the world’s largest clove producer, cultivating more than 70% of the world’s crop. The vast preponderance of Indonesia’s annual clove harvest, around 90%, is pur- chased each year by the Indonesian tobacco industry to produce kreteks. Many thousands of Indonesian households grow cloves, but it typically comprises only a small proportion of their total economic activity. Recent government data suggest that more than one million farmers grow cloves in Indonesia (IAKMI, p. 67). Given that clove production is a fraction of broader household production/resources of most clove-farming households and the labour involved in clove farming is intermittent and concentrated only at specific relatively brief times of the years (e.g., harvest), actual full-time equivalent employment is much smaller. This report on the economics of clove farming is motivated by the recent evolution of the taxation of kreteks and other tobacco products in Indonesia. Over more than a decade, the Indonesian government has slowly been raising excise taxes on tobacco products. More recently, the government has considered reforming the excise tax structure even further by collapsing the number of tiers of taxation rates while also generally raising these rates. The government has articulated concern that these tax reforms, while mitigating consumption of a harmful product, could also affect the economic lives of those in the kretek value chain. Clove farmers are part of this value chain. This report aims to inform the current debate over the taxation of kreteks in Indone- sia by examining clove-farming livelihoods. It presents results from a comprehensive, household-level, economically-focused survey of 600 clove farmers across the two largest clove-growing regions, Sulawesi and Central Java. The survey examines the role that clove farming plays in these households’ economic lives, among other related topics. This report is part of the Indonesia Tobacco Employment Studies implemented by the World Bank and the American Cancer Society. The findings from this report complement the findings from three closely-related reports focusing on specific segments of tobacco sector employment in Indonesia: tobacco farmers (World Bank 2017b), kretek workers or “hand-roll- ers” (World Bank 2017c) and tobacco manufacturing more broadly (World Bank, 2017d). The main findings of this report include: • Though hundreds of thousands of farmers grow cloves in Indonesia, tobacco tax reform is unlikely to cause a large economic disruption for most clove-growing households because more than half of households derived less than 20 percent of their income from cloves. Clove farming is not the primary source of economic livelihood for most households that grow cloves; 4 // Executive Summary • Clove farming is not highly lucrative for many farmers, particularly in certain regions such as Minahasa. After incorporating the costs of household labor, the aver- age clove-farming household was losing more than two dollars per kilogram; • Individuals in clove-farming households in the survey were on average poorer than the average Indonesian household—11.1 percent of Indonesians fall below the national poverty line compared to 37 percent of these clove-farming households; • A loss to a household from lower clove production should not be presumed because many clove-growing households with their varied economic activities are likely to be able to effectively reallocate factors of production (including capital, land and/or labour) to other productive activities such as alternative crops or a small business; {{ Land might be better allocated to more proven economically-lucrative pur- suits, particularly in places where there are clear viable options such as growing more of locally-grown cash crops (e.g., rice, corn, and green vegetables, among others). Recommendations In terms of considering what might happen if there were a change in tax policy that affected demand for Indonesian cloves, it is important to emphasize that clove farming is not particu- larly profitable for most clove-producing households in at least a couple of major clove-pro- ducing districts. The government needs to research which alternatives could be viable and tar- get the least-profitable areas for switching and help these farmers make successful transitions to growing other crops and/or economic activities. • In case of negative effects from higher tobacco taxes, the government could help clove farmers to identify other economically fruitful pursuits. {{ Research which alternative locally-grown crops are most viable on former clove-producing land; {{ Enhance supply and value chains for these alternative local crops; • Support research to identify the areas where clove production is least viable (e.g., Minahasa) and target those areas for alternatives first; • Improve agricultural extension services for non-clove crops; • Provide educational grants for improving clove farmers’ agricultural management skills; • Improve access to credit for farmers to grow other crops or pursue other small economic enterprises such as a small shop, fish farming, or a local transportation business. {{ Provide grants or low-interest loans to farmers willing to switch to alternative crops; and {{ Introduce financial programs or economic development programs as forms of transi- tion assistance away from clove cultivation. 5 ACKNOWLEDGMENTS The preparation of this report was carried out under the World Bank Global Tobacco Control Program coordinated by Patricio V. Marquez, with the support of the Bill & Melinda Gates Foundation and the Bloomberg Foundation. The report is part of the Indonesia Tobacco Employment Studies and was prepared by a team comprised of: Jeffrey Drope (Scientific Vice-President, American Cancer Society); Qing Li (Senior Data Analyst, American Cancer Society); Pandu Harimurti (Senior Health Specialist and Task-Team Leader, World Bank); Edson C. Araujo (Senior Economist and Task-Team Leader, World Bank); Gumilang Aryo Sahadewo (Research Faculty Universitas Gadjah Mada and Consultant World Bank); Nigar Nargis, PH.D. | Scientific Director, Economic and Health Policy Research Josefine Durazo (Survey Specialist, World Bank); Firman Witoelar (Director of Research, SurveyMeter); and Bondan Sikoki (Chairperson, SurveyMeter). The report greatly benefited from comments, inputs, and advice provided by Teh-wei Hu (Professor Emeritus, School of Public Health, U.C. Berkeley). 6 // Executive Summary TABLE OF CONTENTS List of Acronyms 3 Executive Summary 4 Acknowldgements 6 List of Tables 8 List of Figures 9 1. Introduction 11 2. Methods 15 3. Results 17 3.1 – Socio-Demographic Characteristics of Clove Farmers 17 3.2 – Household Assets 18 3.3 – Economic Livelihoods of Clove-Farming Households 22 3.4 – Costs of Clove Cultivation 27 3.5 – Clove Prices and Income 32 3.6 – Credit and Debt 34 3.7 – Household Economic Situation 36 3.8 – Alternative Livelihood Options 38 4. Household Well-being 41 4.1 – Access to Social Safety nets 41 4.2 – Food Security 44 5. Child Labour 49 6. Limitations 51 7. Conclusion 53 References 55 7 The Economics of Tobacco Taxation in Indonesia List of Tables Table 1: Socio-Demographic Characteristics of Clove Farmers 17 Table 2: Household Size by Age Group 18 Table 3: Household Assets 19 Table 4: Non-Livestock Agricultural Assets 20 Table 5: Main Source of Livelihood 23 Table 6: Frequencies of Households by Crop Cultivation by Region—All Seasons 26 Table 7: Mean Nonlabour Costs by Region in USD 27 Table 8: Per Kilogram Nonlabour by Region in USD 27 Table 9: Clove Farming Principal Inputs and Corresponding Costs (annual) 28 Table 10: Average Hours of Hired Labour per Household by Region 31 Table 11: Average Hours of Hired Labour by Gender 31 Table 12: Average Price per Kilogram by Grade and Type (USD) 32 Table 13: Average Clove Production, Price and Income 33 Table 14: Logistic Regression of Willingness to Switch 39 Table 15: KPS and KKS Participation 42 Table 16: Participation in Indonesia Health Card (KIS) and National Health Insurance System (BPJS) 43 Table 17: Staple Food by Number of Months Produced 17 Table 18: Logistic Regression of Determinants of Food Security 46 Table 19: Child Labour 49 8 // Table of Contents List of Figures Figure 1: Indonesia’s Annual Smallholder Clove Production by Ton 11 Figure 2: Indonesian Clove Exports and Imports, 1990–2013 12 Figure 3: Livestock Assets 20 Figure 4: Total Land – Owned (hectares) & Under Cultivation 21 Figure 5: Land Ownership 22 Figure 6: Clove Farming Participation by Age and Gender 24 Figure 7: Proportion of Total Household Resources from Clove Farming Income 24 Figure 8: Land Size Dedicated to Clove Cultivation by Hectare 25 Figure 9: Distribution of Nonlabour Costs in USD 29 Figure 10: Clove-Farming Household Labour Hours by Gender and Age 30 Figure 11: Clove Farming Labour Cost by Gender and Age (in USD) 30 Figure 12: Percentage of Households Hiring Men and/or Women to Cultivate Cloves 31 Figure 13: Profits per Kilogram—Real and Perceived—by Region 33 Figure 14: Demand for Credit by Region 34 Figure 15: Why Farmers Borrow 35 Figure 16: Who Clove Farmers Owe 35 Figure 17: Per Capita Resources and Incomes of Clove-Farming Households 36 Figure 18: Poverty Status of Clove-Farming Households 37 Figure 19: Reasons for Engaging in Clove Farming 37 Figure 20: Percentage of Clove Farmers Considering Alternatives 38 Figure 21: Access to Sufficient Food 44 Figure 22: Total Household Income against Actual Rice Grown (bivariate probability density function, p.d.f. ) 47 9 SMOKING HAS The Economics of Tobacco Taxation in Indonesia BEEN A MAJOR CONTRIBUTOR O DISEASE BURDEN INDONESIA. SMO V LENCE PREVA AGE INDIVIDUAL HAS EXCEEDED 3 PERCENT SINCE 2 10 // Table of Contents 1 INTRODUCTION O Indonesia is the largest producer of cloves in the world. The Food and Agriculture Organization’s (FAO) most recent statistics on clove production (from 2013) report that Indonesia produced more than 70% of the world’s cloves, largely by smallholder farmers N (i.e., not plantations). In contrast, the next largest producer, Madagascar, produced ~16% of the world’s cloves. Important to the broader inquiry of this research, the vast prepon- derance of Indonesia’s annual clove harvest (usually ~100,000 metric tons)—typically around 90 percent—is purchased each year by the Indonesian tobacco industry to pro- duce kreteks. Figure 1 shows smallholder (less than four hectares) production from 1990 to 2015 by ton. O Figure 1: Indonesia’s Annual Smallholder Clove Production by Ton SMALLHOLDER PRODUCTION Ton) 120,000 100,000 Production by ton 80,000 60,000 40,000 20,000 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 Because of the large domestic demand for cloves, in most years, only a small pro- portion, ~5 percent, is exported. Occasionally, after a strong yield over one or more L years, Indonesian clove exports increase. For example, in 2016, the UN trade statistics database reports that approximately 40% of the value of total world clove exports were from Indonesia, which was atypical. Inversely, in the occasional low yield year, clove imports have increased to meet domestic demand. See Figure 2 below to track trends in clove imports and exports from 1990 to 2013. 3 2 11 The Economics of Clove Farming in Indonesia Figure 2: Indonesian Clove Exports and Imports, 1990–2013 25,000 20,000 Volume (Ton) 15,000 10,000 5,000 1990 1992 1994 1996 1998 2000 2002 2004 2006 2008 2010 2012 Import Export The Indonesian clove crop is typically grown on around 500,000 hectares of land, though the monetary value of the crop fluctuates with price. In early 2016, the Indo- nesia Clove Farmer Association (APCI) estimated that the average price was around $7.19/ kg. The same association suggested that this was a profit margin of more than 25% for farmers. The results from the survey reported below do not match in terms of price— farmers consistently reported average lower prices. Many thousands of Indonesian households grow cloves, but it typically comprises only a small proportion of their total economic activity. Though employment statis- tics are scarce, perhaps because growing cloves is only a small part of these households’ economic activities, according to a recent report from the Indonesian Public Health Asso- ciation (IAKMI, p. 67), the government reported that more than one million Indonesian individuals were contributing labour to growing cloves in 2012. It is critical to note—a dynamic supported strongly by the survey results presented in this report—that clove production is a fraction of broader household production/resources of most clove-farm- ing households. Furthermore, clove farming is by nature very rarely the source of full- time employment for most of these individuals, since the labour involved in clove farm- ing is intermittent and concentrated at specific times of the years, such as at harvest. This report aims to inform the current debate over the taxation of kreteks in Indonesia. As the Indonesian government considers reforming its excise tax structure and rates, there have been concerns that these tax reforms could also affect the eco- nomic lives of those in the kretek value chain. Clove farmers are part of this value chain. Accordingly, in this report, we examine empirically the economic livelihoods of clove farmers in Indonesia, largely by analyzing the results of a comprehensive, household-level economically-focused survey of 600 clove farmers across the two largest clove-growing 12 // Introduction regions, Sulawesi and Central Java. The survey examines the role that clove farming plays in these households’ economic lives, among other related topics. This report is part of the Indonesia Tobacco Employment Studies implemented by the World Bank and the Amer- ican Cancer Society. The findings from this report complements the findings from the other reports which focused on other specific segments of the tobacco sector employ- ment in Indonesia, more specifically tobacco farmers (World Bank 2017b), kretek workers (World Bank 2017c) and tobacco manufacturing more broadly (World Bank, 2017d). The report is structured as follows: the next section, section 2, presents a summary of survey methods and the analysis undertaken. Then the following section, section 3, presents the results of the analyses of the survey data. It begins with an examination of the socio-demographic characteristics of the households. Next, it tracks the economic situation of the households including their broad economic endeavors, costs of growing cloves, prices of and incomes from cloves, credit and debt to grow cloves, and alterna- tives to cloves. It also tracks household assets. The subsequent sections focus on specific issues, such as farmers’ wellbeing, child labour, health and food security. The final sec- tions, discusses limitations of the study and present conclusions, respectively. 13 SMOKING HAS BEEN A MAJOR CONTRIBUTOR O DISEASE BURDEN INDONESIA. SMO PREVALENCE AMONG WORKIN AGE INDIVIDUAL HAS EXCEEDED 3 PERCENT SINCE 2 14 // Introduction 2 METHODS The report utilizes a household-level survey of a nationally representative sample of clove-producing households. To examine the economic livelihoods of clove-farming households in Indonesia, we developed a survey instrument derived in large part from previous surveys used to evaluate the economic livelihoods of tobacco farmers in Africa and Southeast Asia, and the World Bank’s Living Standards Measurement Survey (see Chavez et al., 2016; Goma et al., 2015; Magati et al., 2016; Makoka et al., 2016; World Bank, 2017). The survey was particularly focused on collecting data on household economic activity, including all sources of clove-farming revenue and production. At the same time, it queried respondents comprehensively on the costs of running both their agricultural and non-agricultural businesses, including both hired and household labor, to provide a complete snapshot of each household’s economic life. To implement the survey, we purposively selected the two top clove-producing provinces, North Sulawesi and Central Java, which typically account for more than 70 percent of clove production. In each province, we selected the two highest-produc- ing kabupaten (districts). In each kabupaten, we randomly selected one top-producing kecamatan (subdistrict) and one second-tier producing kecamatan. Within each selected kecamatan, we chose the three highest-producing villages. In each selected village, we selected the sub-village with the largest number of farmers according to the village infor- mant. We went door to door to verify that the farmers listed were indeed current farmers. If we verified 50 farmers in the first sub-village, we stopped and used that list to draw our sample. If we could not find and verify 50 farmers, we went to the next sub-village to get the additional ones. From the sub-villages in which we could list 50 farmers, we randomly selected 25 to interview. A complete list of these districts and municipalities is available in the supplementary appendices available upon request. The report utilizes mainly thick, descriptive statistics of the survey results, but also utilizes causal analysis where appropri- ate, particularly for food security. 15 SMOKING HAS BEEN A MAJOR CONTRIBUTOR O DISEASE BURDEN INDONESIA. SMO V LENCE PREVA AGE INDIVIDUAL HAS EXCEEDED 3 PERCENT SINCE 2 16 // Results 3 RESULTS O 3.1 - Socio-Demographic Characteristics of Clove Farmers Clover farmers’ households are typically headed by married males, usually older N than average and with an elementary school education or less. In Table 1, we pres- ent the basic socio-demographic characteristics of the heads of the clove-farming house- holds—household heads were overwhelmingly male (95.5%) and married (88.2%). They tended to be older than the average population with 69% between 36 and 60 years of age (the average age in 2016 in Indonesia was 27), and another ~25% who were older O than 60 years. The majority (~60%) had an elementary school education or less. More than two-thirds (69%) reported agriculture as their principal activity. Table 1: Socio-Demographic Characteristics of Clove Farmers N = 600 PERCENT GENDER Male 573 95.5 Female 27 4.5 AGE (YEARS) 20-35 33 5.54 36-60 412 69.13 >60 151 25.34 MARITAL STATUS Never married 12 2.00 L Married 529 88.17 Divorced/separated 11 1.83 Widowed 47 7.83 Living with partner 1 0.17 3 HOUSEHOLD SIZE BY AGE <21 418 1.71 21–35 294 1.44 36–60 527 1.74 >60 233 1.30 2 17 The Economics of Clove Farming in Indonesia Table 1: Socio-Demographic Characteristics of Clove Farmers (CONT.) N = 600 PERCENT EDUCATION Not yet/no schooling 29 4.83 Elementary school 332 55.33 Public junior high 107 17.83 Senior high school 78 13.00 Vocational school 22 3.67 College or above 32 5.33 MAIN ACTIVITY Agricultural work 414 69.00 Non-agricultural work 160 26.67 Home duties 4 0.67 Retired/aged 8 1.33 No work 14 2.33 Clove-farming households average four people. In Table 2, we present the size of clove-farming households. The median household size was four, typically with two peo- ple older than 36 years, one between 21 and 35, and then one who was younger than 21. Households were as few as one and as large as 10. Table 2: Household Size by Age Group TOTAL GENDER AGE RANGE F M <21 21–35 36–60 >60 Median 4 2 2 1 1 2 0 Minimum 1 0 0 0 0 0 0 Maximum 10 7 6 5 4 4 3 3.2 - Household Assets Most clove-farming households demonstrated some asset accumulation. For exam- ple, as presented in Table 3, approximately 90% of households reported having a TV and/ or a mobile (cell) phone. More than three quarters of households owned at least one motorcycle. Nearly half of households had a VCR/DVD player or similar device. 18 // Results Table 3: Household Assets ITEM OWNERSHIP TV 91.50% DVD/VCD player/home theater/radio tape 44.17% Satellite disk 31.67% Electric oven/microwave 2.67% Refrigerator 46.00% Liquefied petroleum gas tube (≥3 kg) 86.67% Washing machine 25.50% AC (air conditioner) 0.67% Telephone (landline) 0.33% Mobile (cellular) phone 88.67% Computer/laptop 16.83% Tablet 8.17% Video camera (handycam)/camera 3.83% Water heater 11.33% Electric pump/jet pump 11.17% Generator 16.50% Vehicle 14.00% Boat/motorboat 3.67% Motorcycle 76.83% Many clove farmers reported other agricultural assets, particularly poultry and/or small livestock. In Figure 3, we report on households’ livestock assets. Slightly more than half of the clove-farming households reported owning chickens while nearly 40% indi- cated owning small livestock such as goats and sheep. Only approximately 20% reported owning large livestock such as buffalo, cows or horses, which is lower than for the general agricultural population which is appoximately 40 percent according to the most recent Indonesia Family Life Survey (IFLS, 2016). 19 The Economics of Clove Farming in Indonesia Figure 3: Livestock Assets 60% 56.50% 38.17% 40% 19.83% 20% 0% Large livestock Small livestock Poultry (cow, bu alo, horse, etc.) (goat, sheep, pigs, etc.) (chicken,ducks, geese, quail, etc.) Most clove farmers reported a variety of non-livestock agricultural assets. In Table 4, we report clove farmers’ non-livestock agricultural assets. More than half (57%) of farmers reported owning a cangkul (a “hoe” in English) used to loosen earth, among other activities. More than one quarter of households reported owning a chopping machine that can chop raw agricultural products. Roughly one quarter of households reported owning a sabit (a “sickle” in English), which is a hand-held agricultural chopping imple- ment. Very few households reported owning more expensive agricultural assets such as a pump, a plough or a tractor. Table 4: Non-Livestock Agricultural Assets ITEM OWNERSHIP Item Ownership Wagon 6.33% Plough 0.50% Tractor 1.83% Water pump for farming 1.33% Chopper machine 28.17% Cangkul 56.67% Sabit 25.33% 20 // Results The total size of land of clove-farming households owned and cultivated for cloves varied within and across regions. As Figure 4 reports, the median size of the land plots owned and cultivated was generally much greater in Sulawesi. In Bolaang Mongondo and Minahasa, the median household owned 2 hectares. In Minahasa, they cultivated most of it (on average about 1.75 ha). In Bolaang Mongondow, farmers on average cul- tivated about three-quarters of their land. In Java, the median plot sizes were smaller, 1 hectare in Wonogiri and just over half of a hectare in Pemalang. In Pemalang, perhaps due to the small plot size, all farmers reported cultivating all of their cultivable land. Figure 4: Total Land – Owned (hectares) & Under Cultivation 2.50 2.00 2.00 2.00 1.75 1.50 1.50 1.00 1.00 0.75 0.54 0.54 0.50 0.00 Bolaang Mongondow Minahasa Pemalang Wonogiri Cultivated Owned Most clove-farming households owned the land on which they are growing cloves. As Figure 5 illustrates, households owned the vast majority, ~92%, of land parcels that they used to cultivate their cloves. Only 1% of the parcels were rented to grow cloves, while approximately 5% of the parcels were someone else’s land but the farmers reported that they did not pay rent to grow cloves on the land. 21 The Economics of Clove Farming in Indonesia Figure 5: Land Ownership 1.06% 4.86% 1.00% 0.93% 92.15% Granted by local leaders Owned Rented from others Tenant (no rent) Other 3.3 - Economic Livelihoods of Clove-Farming Households Clove-farming households demonstrated a wide variety of principal economic activity, but rarely actual clove farming. In terms of the reported main livelihoods of household members (all, head and otherwise), there was considerable heterogeneity, with no one category capturing more than 15% of the survey respondents. As Table 5 illustrates, agriculture was the largest represented sector. Clove farming comprised 15 percent of those respondents who answered the question, as did non-clove farming. Clove farming is not the primary source of economic livelihood for most house- holds that grow cloves. In the column on the farthest right, the table reports the average daily income of individuals from the survey working in that sector and the aver- age number of months worked in that sector in the previous 12 months. Individuals reporting clove farming as the main source of livelihood averaged $7.93 per day in daily income, which was slightly above the average of all survey respondents, $7.32. However, on average, respondents worked fewer than four months at clove farming. Respondents reporting non-clove agriculture as a main livelihood generated an average of $4.77 per day from this activity. Importantly, many of these activities were not full time (perhaps especially some agricultural activities), which must be considered, particularly when eval- uating/comparing these values across sectors (for a version of Table 5 for all household members, please see the online appendix). 22 // Results Table 5: Main Source of Livelihood N PERCENT AVERAGE DAILY AVERAGE MONTHS WAGE (IN USD) WORKED FOR LAST 12 MONTHS Agriculture: clove 95 14.68 7.93 3.79 Agriculture: non-clove 97 14.99 4.77 4.00 Forestry; animal husband- 70 10.82 5.09 5.44 ry; fishing; Mining and quarry 27 4.17 12.13 6.11 Electricity; gas; water 6 0.93 8.03 7.50 (utilities) Construction/building 87 13.45 9.52 4.36 Transport; storage; com- 24 3.71 7.15 8.00 munications Financial services; real 5 0.77 8.49 9.00 estate; lease Restaurant; food sale 16 2.47 4.67 5.38 Industry: food processing/ 4 0.62 3.49 6.25 production Industry: clothing 6 0.93 5.10 7.50 Industry: other 9 1.39 13.49 5.89 Sales: nonfood 11 1.70 6.44 6.27 Service: government 60 9.27 7.05 10.30 Service: teacher 50 7.73 5.61 11.22 Service: professional 17 2.63 7.29 7.12 Service: transportation 5 0.77 7.27 6.80 Service: other 48 7.42 5.81 6.49 Other 10 1.55 3.64 7.10 Total 647 100 7.32 6.11 23 The Economics of Clove Farming in Indonesia Figure 6: Clove Farming Participation by Age and Gender 94.89% 94.96% 99.29% 95.59% 95.24% 100% 91.18% 88.00% 83.16% 80% 60% 40% 20% 0% <21 21–35 36–60 >60 Male Female Most members of clove-farming households participated in clove farming, though it was usually a small proportion of their total economic activity. In Figure 6, we report the percentage of all adult household members who indicated participating at some point in the previous year in clove production. Across all age categories, participa- tion rates in clove farming were very high. Thus, though only a small proportion reported clove farming as a main activity, it was an activity for most individuals living in the sur- veyed households. Figure 7: Proportion of Total Household Resources from Clove Farming Income 40 34.56 30 Percent 20 17.7 9.85 10 8.347 6.01 5.175 5.008 4.174 4.841 4.341 0 0 .2 .4 .6 .8 1 Proportion of Clove Income in Total Household Resources 24 // Results Most clove-farming households are cultivating cloves as a secondary economic pursuit. In Figure 7, we examine the proportion of total household resources generated by farming cloves. The results demonstrate that more than half of these clove-farming households generated less than 20% of their total household resources from clove farm- ing.  At the same time, less than a quarter of the clove-farming households in the survey generated more than half of their overall resources from clove farming. The nature of cloves as a relatively low-maintenance perennial plant reasonably lends itself to this dynamic, though there are bursts of more labor-intensive activity during the year, particularly at harvest and postharvest times. Figure 8: Land Size Dedicated to Clove Cultivation by Hectare 64.01 60 40 Percent 23.79 20 7.098 1.37 1.059 .4981 .8095 .1868 .3736 .8095 0 0 5 10 15 20 Hectares of Land Cultivated Most households—nearly two-thirds—reported cultivating less than one quarter of a hectare of land to grow cloves. In Figure 8, we present the number of hectares that the surveyed households dedicated to clove farming. There are two caveats to this finding in that we observed that many farmers had difficulty assigning the size of the land for clove trees because they were above the ground. Also, some farmers reported using some of the land under the canopy of the clove trees for other productive agricultural purposes (i.e., growing other crops). Clove farmers reported growing a wide variety of crops, which varied by region. The survey also asked about all the crops the households were cultivating across all grow- ing seasons. In many districts, respondents reported three growing seasons. As reported in Table 6, there was significant variation among regions. For example, in Pemalang and Wonogiri, cassava was relatively widespread (46.7% and 44% respectively). In Pemalang, 56.7 percent of farmers reported growing bananas, and another 44.7 percent reported cultivating bamboo or wood. In Wonogiri, 71.3% reported growing paddy (rice). In gen- eral, clove-farming households in Bolaang Mongondow Timur and Minahasa reported growing a smaller variety of crops beyond cloves. 25 The Economics of Clove Farming in Indonesia Table 6: Frequencies of Households by Crop Cultivation by Region—All Seasons BOLAANG MINAHASA PEMALANG WONOGIRI TOTAL MONGONDOW Cassava 3 12 70 66 151 Sweet potato 0 7 22 9 38 Ground nut 1 1 1 3 6 Cashew/other 1 5 0 6 nut Corn 7 25 37 23 92 Chili 26 19 37 1 83 Shallot 1 0 2 0 3 Coconut 30 36 15 6 87 Banana 31 33 85 13 162 Spices 11 15 36 91 153 Green vegetables 6 6 20 0 32 Other vegetables 1 3 13 5 22 Clove 150 150 150 150 600 Fruits 5 23 22 1 51 Paddy 1 7 55 107 170 Wood/bamboo 20 40 67 79 206 Tea/coffee/ 51 1 47 6 105 chocolate Other 0 1 9 16 26 26 // Results 3.4 – Costs of Clove Cultivation Mean and per kilogram nonlabour costs varied by region. In Tables 7 and 8, we exam- ine mean and per kilogram nonlabour costs. The average costs varied among regions and major categories of expenses (depreciation of equipment, inputs, interest, levies and land rental). On average, households spent $231.76 on clove farming. Physical inputs (e.g., fertilizer, pesticides, equipment, etc.) were typically the largest cost both per kg, but also varied, from 0.35/kg in Wongiri to $1.69/kg in Bolaang Mongondow. Rent was only a large cost in Bolaang Mongondow ($3.37/kg). The other major cost in at least two regions was interest on loans—$0.66/kg in Minahasa and $0.50 in Pemalang. We also calculated costs per hectare but we found the calculations fundamentally problematic because farmers have difficulty assigning a value to the size of the land dedicated to love farming. Table 7: Mean Nonlabour Costs by Region in USD DISTRICT DEPRECIATION INPUT INTEREST LEVIES RENT TOTAL COST COST Bolaang 18.27 95.08 118.89 4.15 337.41 237.37 Mongondow Minahasa 10.28 198.11 220.40 6.46 4.50 428.49 Pemalang 6.76 55.17 73.62 0.54 269.92 131.78 Wonogiri 11.67 46.18 72.67 0.36 749.79 130.03 Total 11.74 99.28 121.39 2.88 340.41 231.76 Table 8: Per Kilogram Nonlabour by Region in USD DISTRICT DEPRECIATION INPUT COST INTEREST LEVIES RENT TOTAL COST Bolaang 0.57 1.69 0.38 0.45 3.37 6.04 Mongondow Minahasa 0.30 1.43 0.66 0.02 . 39.82 Pemalang 0.10 0.55 0.50 0.00 . 1.09 Wonogiri 0.25 0.35 0.27 0.00 . 1.35 Total 0.32 1.05 0.44 0.13 3.37 12.78 27 The Economics of Clove Farming in Indonesia Fertilizer, fuel and/or transportation were the most common costs associated with clove farming. Table 9 reports in further detail on the physical (i.e., nonlabour) inputs that clove farmers reported using. In total, the average per household dollar amount spent on clove inputs was $99.21. Almost all (93%) farmers reported buying a sickle and nearly two-thirds indicated purchasing bamboo sticks. Nearly 60% reported paying in some way to transport their cloves to market. Most clove farmers bought fertilizers, both nonorganic (56%) and organic (50%). Half of farmers purchased gasoline for vehicles and equipment related to clove farming activities. The highest average cost of a widely-used input was nonorganic fertilizer at $58.79. The average cost of equipment rental was higher ($85.13), but only 12 percent of farmers reported such a cost. Table 9: Clove Farming Principal Inputs and Corresponding Costs (annual) INPUT PROPORTION OF FARMERS AVERAGE COST (USD) WHO USED INPUT Fertilizer nonorganic 56.17% 58.79 Fertilizer 49.83% 21.45 Pesticides (chemicals) 35.50% 13.31 Gasoline for clove farming 49.50% 47.48 equipment Oil 25.00% 5.31 Firewood/fuel wood 4.17% . Bamboo, bamboo sticks, rice 63.67% . hay, descuke-ride Knapsack sprayer 52.67% . Drums 14.00% . Sprinkler 18.83% 46.64 Rental of equipment/live- 12.00% 85.13 stock Transportation (to market) 59.17% . Water pump 2.50% . Mattock, sickle 93.00% . Others 47.00% 24.30 28 // Results Most clove farmers reported total nonlabour costs of less than $200. Figure 9 is a graphical representation of the distribution of average costs related to clove farming by region. Across all four regions, we can observe that most households report clove-related, nonlabour input costs below $200 per year, which was comparable to the mean costs of the tobacco farmers from the companion survey. Figure 9: Distribution of Nonlabour Costs in USD 3,000 2,000 Non-labour Input 1,000 0 BMT MINAHASA PEMALANG WONOGIRI Older household members were likelier to dedicate the most hours to clove farming. The survey queried respondents about the number of hours of labour the individuals in the survey dedicated to cultivating their cloves. On average, households dedicated an average of 479 hours per year to clove cultivation. Figure 10 demonstrates that male household members spend about 281 hours on average, with particularly older men expending more hours, on clove farming. The less physically-intensive nature of this work—mainly pruning, weeding, and watering compared to ploughing and other labo- rious field preparation—may help to explain this finding. For example, men older than 60 years spent on average 332 hours, while those between 21 and 35 spent 195 hours. Women spent fewer hours than men, on average about 170 hours, and with the same pattern that older women dedicated more hours than younger women. The total value of labour varied by age and sex with the labour of men older than 36 years demonstrating the highest value. In Figure 11, we report the average labour costs of household members. We calculated the value of the labour by first establishing the district-level agricultural minimum wage (Government of Indonesia, 2016), then con- verted the number of total hours into the appropriate units (e.g., day, week, month, etc.) and multiplied by the wage rate. The value of labour for men older than 60 was the highest at approximately $260 for the year. 29 The Economics of Clove Farming in Indonesia Figure 10: Clove-Farming Household Labour Hours by Gender and Age 400 350 336.30 Hours Worked For Clove Farming 331.36 300 250 215.02 200 194.81 192.53 150 133.92 100.99 100 50 46.13 0 Age Group < 21 21–35 36 – 60 > 60 Male Female Figure 11: Clove Farming Labour Cost by Gender and Age (in USD) 300 259.12 261.51 250 221.36 200 178.43 157.58 143.06 150 130.81 109.78 100 87.00 50 38.06 0 < 21 21–35 36–60 > 60 Total Male Female Most clove farmers hired some external seasonal workers, particularly for harvest. The results presented in Figure 12 demonstrate that many clove-farming households hired laborers to help with clove cultivation. It was particularly common for households to hire help during the harvest season. Nearly 90 percent of households hired men to work in this season, while nearly 19 percent hired women. 30 // Results Figure 12: Percentage of Households Hiring Men and/or Women to Cultivate Cloves 18.60% 100% 80% Proportion of farmers 60% 4.58% 3.50% 0.27% 89.49% 40% 3.77% 1.62% 51.48% 20% 48.25% 12.40% 16.17% 0.54% 31.00% 0% 2.70% Nursery Trans-planting Field tending Harvest Postharvest Selling/marketing Adult male Adult female Child The amount of hired labour varies widely across regions. Table 10 furnishes more information about hired labour. The total number of hours varies by region with farmers in Minahasa hiring for the most hours at 354. As illustrated in Table 11, farmers were more likely to hire men. On average, a hired male laborer will work 135 hours at harvest time, significantly more than the average hired woman (57 hours). Field tending is the next most labor-intensive activity with hired men laborers averaging nearly 67 hours and hired women laborers 98 hours. The postharvest hours for child labourers is a potential issue and is explored in a later section of the report. Table 10: Average Hours of Hired Labour per Household by Region DISTRICT TOTAL LABOUR HOURS Bolaang Mongondow 139.99 Minahasa 354.28 Pemalang 143.00 Wonogiri 69.07 Total 203.67 Table 11: Average Hours of Hired Labour by Gender NURSERY TRANS- FIELD HARVEST POST- SELLING/ PLANTING TENDING HARVEST MARKETING Adult male 13.81 25.45 66.61 135.73 46.78 4.30 Adult 13.14 22.62 98.29 57.12 44.07 2.00 female Child 0.00 . . 44.25 80.83 . Total 13.63 25.26 69.20 121.43 47.49 3.92 31 The Economics of Clove Farming in Indonesia 3.5 - Clove Prices and Income There is significant variation in the price of cloves depending on the variety and grade. For example, Table 12 reports that white cloves earned the lowest prices with Grade A White cloves earning on average $3.27/kg. In contrast, Sikotok and Zazibar Grade A cloves earned nearly $6/kg. Notably, Sikotok Grades B and C cloves earn about the same, while lower quality Zanzibar cloves earn significantly less. Note that none of these prices matches the $7.19/kg estimate of the Indonesia Clove Farmers Association. Table 12: Average Price per Kilogram by Grade and Type (USD) TYPE Grade White Sikotok Ambonese Zazibar A 3.27 5.89 4.11 5.92 B 2.81 5.82 4.65 4.62 C 2.04 6.37 2.89 D 1.97 3.05 W 2.42 2.10 2.80 2.36 Total 2.91 5.43 3.87 5.30 There was wide variation in clove prices and income reported, both within and across regions. The survey asked each household to report on average clove production, price and income. As Table 13 suggests, average clove production varied considerably, from 203.61 kg in Minahasa to 320.72 kg in Pemalang. The average price received by the farmers depended on the type of clove and varied among regions. The cloves grown on Sulawesi (Bolanng Mongondow and Minahasa) are typically of the Zanzibar variety and fetched higher average prices—between $6.50/kg and $7/kg USD—than in Central Java where farmers in Pemalang received an average price of $3.14/kg and those in Wongiri, $1.96/kg. In Central Java, farmers grew both white clove and Zanzibar clove (and at least a couple of other types, too). Partly because of price (and total cultivation per house- hold), average income from cloves ranged widely among the districts from $439.13 in Wonogiri to $2,575.72 in Minahasa. Another variable that affected price was whether the farmer sold the clove bud dry or wet. Dry bud—more common in Sulawesi—typically fetched a higher price. 32 // Results Table 13: Average Clove Production, Price and Income DISTRICT QUANTITY OF AVERAGE PRICE REPORTED CLOVE (KG) (USD) ANNUAL CLOVE INCOME (USD) Bolaang Mongondow 249.56 6.52 1,665.54 Minahasa 203.61 6.73 2,575.72 Pemalang 320.72 3.14 778.44 Wonogiri 254.10 1.96 439.13 All 251.02 4.69 1,364.71 Profits from clove cultivation varied widely by district. In Figure 13, we examine the profits per kilogram by region. The red bars indicate profits in terms of “perceived” profits or what most agricultural economists would term “gross margins,” which is the gross revenues from sell- ing cloves less the direct costs incurred to grow them including physical inputs (e.g., fertilizer and pesticide), hired labour, marketing expenses and transportation. The blue bars represent what we term “realized profits” because they assign a value to the household labor dedicated to clove cultivation. We determine this value using the regional agricultural minimum wage. Overall, we observe that the average profit per kg using the gross margin method was $4.84. This was driven in large part by the high prices paid to clove farmers in Bolaang Mongondow where the gross margin was $12.34/kg. The lowest gross margin was in Minahasa where it was $0.26/kg. In the blue columns, we incorporate the value of labour in the cost calculations. Overall, we find that average realized profits from cultivating cloves were negative for the sur- veyed households at –2.24/kg. We find that Pemalang is the most profitable at $1.45/kg, while Bolaang Mongondow is the least profitable at –$2.72. Considering this district’s higher average prices, labour costs for clove cultivation are clearly very high in this area. Figure 13: Profits per Kilogram—Real and Perceived—by Region PER KILOGRAM PROFIT 15 12.34 10 4.69 4.84 5 0.26 1.45 1.67 US $ 0.27 0 -2.24 -5 -2.72 -7.11 -10 Bolaang Minahasa Pemalang Wonogiri Total Mongondow Real Profit Perceived Profit 33 The Economics of Clove Farming in Indonesia 3.6 - Credit and Debt Less than one-third of clove farmers sought credit to cultivate cloves. In Figure 14, we report the number of survey respondents who were seeking credit—ostensibly—to grow cloves. We find that the proportion never exceeds one-third and drops as low as 22.7% in Wonogiri. It appears that clove farming is not an economic activity that farmers commonly seek credit to pursue. Figure 14: Demand for Credit by Region 40% 32.00% 30.00% 30% 28.67% 22.67% 20% 10% 0% Bolaang Mongondow Minahasa Pemalang Wonogiri Clove farmers typically reported seeking credit for broader business and house- hold reasons than to support clove production directly. Figure 15 illustrates why some of the clove farmers are borrowing money and suggests that many of the reasons that they borrow do not pertain to actual clove farming. The most common explanation for borrowing money was a rather general explanation of “investing” in their clove-farming business (21.3%). But, the two next most common explanations were not farming-re- lated: to pay for school fees (11.3%) or to purchase a vehicle (11.3%). These findings are similar to patterns with tobacco farmers in the Philippine (Briones, 2013; Chavez et al., 2016) and Sub-Saharan Africa (Goma et al., 2015; Magati et al., 2016; Makoka et al., 2016). Notably, only five percent mentioned that they wanted to borrow to pay for the inputs to cultivate cloves. 34 // Results Figure 15: Why Farmers Borrow PRIMARY USE OF LOAN 5.17% 5.47% 3.95% 29.18% 0.91% Inputs for Clove Farming Land for Farming Other Crop 11.25% Purchase vehicle Other; specify 3.04% Land for Clove Farming Schooling 8.51% Investing in business 11.25% Inputs for Farming Other Crop Purchasing house 21.28% Special Occasions (wedding,etc.) Clove farmers most typically borrowed from commercial banks or microfinance institutions. In Figure 16, we examine the sources of clove farmers’ loans. The most 4.86% banks (28.5%). The next most common source common source of credit was commercial of credit was microfinance institutions (21%). The third most common source of credit was from relatives (16.7%). In a related dynamic, we found that 33 households were under contract to sell their cloves to a buyer. In seven of those cases, the buyer had pro- vided them with inputs such as fertilizer, which was then deducted from the sales to that buyer; notably, none of these farmers reported this as a loan even though it effectively was because, on average, they paid above market prices for these inputs. Figure 16: Who Clove Farmers Owe 10.64% Relatives 16.72% 2.43% Neighbors (not a farmer) Fellow farmer 4.26% Farmers organization 1.52% BMT 1.82% 2.43% Local money lender 28.57% Microfinance institution 7.29% Rural bank Commercial bank Government 3.34% 20.97% Other 4.86% 35 The Economics of Clove Farming in Indonesia 3.7 – Household Economic Situation We generate and compare per household median and mean resources and incomes of clove-farming households. As presented in Figure 17, we calculated each household’s total resources, which includes all revenues and other production (e.g., food cultivated and consumed), and generated a per household measure. Similarly, we calcu- lated household income, which was resources less all expenses incurred by the house- hold to produce the goods/services (either sold or consumed). PER CAPITA HOUSEHOLD RESOURCE/INCOME Figure 17: Per Capita Resources and Incomes of Clove-Farming Households Annual median per capita income 432.54 Annual mean per capita income 1288.68 Annual median per capita resource 711.45 Annual mean per capita resource 1802.24 0 500 1,000 1,500 2,000 2016 US$ Poverty among clove-farming households is widespread. In Indonesia, generally, 8.3% of households fall below the World Bank poverty rate of $1.90/day (PPP), while 11.1% fall under the Indonesian government’s poverty line of ~$22.60 USD per month. In Figure 18, we illustrate the proportion of clove-farming households that fall below these pov- erty lines. Using the measure of household total resources, which does not incorporate any agricultural or other business costs, the poverty rate for clove-farming households is 16.5% at the World Bank threshold and 21.83% at the Indonesian government threshold. When we use the more realistic measure of household total income, which subtracts all of the expenses incurred to farm or to run other household businesses from total resources, the poverty rate of clove-farming households increases to 34.17% at the World Bank line and 37% at the official government line. 36 // Results Figure 18: Poverty Status of Clove-Farming Households POVERTY STATUS 38.05% Poverty at National Poverty Line 330,776 Rupiah a month per person 23.91% 35.29% Poverty at $1.90 a day per person, PPP 2011 18.77% 0% 10% 20% 30% 40% Headcount ratio measured by per capita income Headcount ratio measured by per capita resources Farmers provide a variety of explanations for choosing clove farming. The chal- lenging economic situations described above reasonably beg the question: why do clove farmers choose to cultivate this crop? As Figure 19 shows, the most common reason— given by just over a quarter of survey respondents—is that clove farming is lucrative. The results above from Bolaang Mongondow support this perspective. About 20% of respon- dents indicated that it was the only viable cash crop while another 20% reported that it had more to do with the availability of land to cultivate cloves. In order to determine how accurate their perception was, we cross-referenced perception and reality and those who identified clove farming as lucrative (either as a primary or “ever mentioned” reason) were not any more profitable (either per kilogram or per hectare) on average than those who did not identify it as lucrative. One weakness is individuals may define lucrative differently. Figure 19: Reasons for Engaging in Clove Farming 4.71% 12.08% Existence of ready market It was the only viable cash crop 25.73% Inherited it from parents 19.59% Availability of land 0.20% Influence by other clove producers It was a highly lucrative 17.82% enterprise 19.86% 37 The Economics of Clove Farming in Indonesia 3.8 - Alternative Livelihood Options Most clove farmers were not considering alternatives to clove farming. As Figure 20 illustrates, nearly 70 percent of clove farmers were not considering switching. Less than 20 percent of farmers indicated being serious or very serious about switching to an alter- native livelihood. Figure 20: Percentage of Clove Farmers Considering Alternatives No, not at all 69.67 Not so serious - thought about it 11.67 Serious - researched other options 16.17 Very Serious - Already planning to switch 2.5 0 10 20 30 40 50 60 70 80 We also used logistic regression to examine which farmers were willing to try alter- natives to clove farming. The survey inquired about the possibility of farmers switching from clove cultivation to another crop. Some farmers indicated a willingness to switch, a process that has been happening involuntarily in some parts of the country due to chal- lenges with clove disease and insect infestations (Rachman, 2016). In Table 14 we present the results from a logistic regression. The dependent variable is dichotomous with a “1” value representing those who are actively switching their clove crops to some extent already or have thought recently about switching. We used the previous literature on switching from tobacco growing as a starting point (e.g., Chavez et al., 2016; Goma et al., 2015; Magati et al., 2016; Makoka et al., 2016), and then used machine-learning methods and stepwise regression to further identify relevant independent variables from the survey results. Farmers who farm fewer cloves or who have larger agricultural enterprises were the most willing to switch away from cloves. We report only the statistically signifi- cant coefficients in the table, though all the foundational analyses are available in our online appendices. We found that clove farmers who cultivated smaller amounts of land for cloves are more likely to be willing to switch from cloves to another crop. Farmers who cultivate more land in general are also more likely to be willing to switch – perhaps because they have flexibility to hedge each season and grow different crops and/or they 38 // Results Table 14: Logistic Regression of Willingness to Switch VARIABLES COMPLETE CASE IMPUTED MISSING –2.520** Hectares of land for clove farming (1.029) 2.655*** Hectares of land cultivated (0.627) 0.0774* Nonlabour income ($100) (0.0431) 0.218* 0.0832*** Non-clove crop sales ($100) –0.114 (0.0247) –0.0739*** Clove sales ($100) (0.0229) 0.424** Depreciation ($100) (0.171) 0.0922*** Hired labour cost for clove ($100) (0.0262) 0.725*** 0.282*** Input cost for clove ($100) (0.255) (0.109) –3.610*** –0.954*** Cost of interest ($100) (1.021) (0.230) Hired labour cost for non-clove –0.239*** crops ($100) (0.0752) Nonagricultural household enterprises 0.00585* 0.00516*** income ($100) (0.00301) (0.00133) nonagricultural household enterprises –0.101** –0.0850*** cost ($100) (0.0488) (0.0260) Observations 254 600 Standard errors in parentheses *** p<0.01, ** p<0.05, * p<0.1 have more experience cultivating other crops. Similarly, farmers who already make more money selling crops other than cloves are more likely to be willing to switch. Farmers who had successful non-agricultural enterprises (and low-cost ones) were also more willing to quit. If the government wishes to move farmers away from clove production to decrease supply, this analysis gives several reasonable starting points in terms of whom to target. 39 SMOKING HAS BEEN A MAJOR CONTRIBUTOR O DISEASE BURDEN INDONESIA. SMO PREVALENCE AMONG WORKIN AGE INDIVIDUAL HAS EXCEEDED 3 PERCENT SINCE 2 40 // Introduction 4 HOUSEHOLD WELL-BEING 4.1 — Asset to Social Safety Nets Nearly half of clove farming households in the survey received a social assistance benefit. In Table 15, we examine the proportion of households receiving social protec- tion. We observe first that the total percentage of the surveyed households that received benefits either under a Social Security Card (KPS) or a Family Welfare Card (KKS) was just under 15 percent. In terms of a cash assistance/transfer, 3.7 percent of the total surveyed households received this benefit and had a KPS/KKS card. Another 1 percent of the surveyed households received this benefit but did not have the card. The largest benefit received by these farmers was through the rice for the poor program: 11.7 percent of surveyed households were receiving the rice benefit and had a KPS/KKS, while another 31.7 percent of households were receiving the benefit and did not have a card. Rates of receiving the other two major benefits—health and education assistance—were fairly similar to the cash assistance rates. These rates were below national averages—nearly a quarter of Indonesians are eligible for KKS and nearly 70 percent of households receive “rice for the poor” benefits. Enrollments in government health programs were higher among clove farmers than in the general population. As shown in Table 16, approximately 16 percent of individuals in clove-farming households were beneficiaries of KIS while approximately 13 percent benefits from BPJS. Enrollment in each program nationwide is about 10 percent. 41 The Economics of Clove Farming in Indonesia Table 15: KPS and KKS Participation INDICATORS STATUS OF CLOVE FARMERS’ HOUSEHOLDS Percentage of households who have KPS/KKS 14.67% BENEFIT 1: PERCENTAGE OF HOUSEHOLDS WHO RECEIVED CASH ASSISTANCE/ TRANSFER IN LAST YEAR With KPS/KKS 3.67% Not with KPS/KKS 1.00% BENEFIT 2: PERCENTAGE OF HOUSEHOLDS WHO RECEIVED RICE FOR THE POOR (RASKIN) IN LAST YEAR With KPS/KKS 11.67% Not with KPS/KKS 31.67% BENEFIT 3: PERCENTAGE OF HOUSEHOLDS WHO RECEIVED ASSISTANCE FOR HEALTH PAYMENT IN LAST YEAR With KPS/KKS 2.50% Not with KPS/KKS 3.00% BENEFIT 4: PERCENTAGE OF HOUSEHOLDS WHO RECEIVED ASSISTANCE FOR POOR STUDENT IN LAST YEAR With KPS/KKS 3.33% Not with KPS/KKS 2.83% OVERALL PERCENTAGE OF HOUSEHOLDS WHO RECEIVED ASSISTANCE FROM KPS/ KKS OR OTHER SOURCES IN LAST YEAR One benefit 36.67% Two benefits 8.50% Three benefits 2.00% Total 47.17% 42 // Household Well-Being Table 16: Participation in Indonesia Health Card (KIS) and National Health Insurance System (BPJS) INDICATORS STATUS OF CLOVE FARMERS’ HOUSEHOLDS PERCENTAGE OF HOUSEHOLDS WHO HAVE KIS/BPJS-PBI KIS 16.17% BPJS-PBI 13.17% Both KIS and BPJS-KIS 0.67% None 71.33% Percentage of households who have KIS and used it 0.53% in last year PERCENTAGE OF HOUSEHOLDS WITH VOLUNTARY BPJS (BPJS SUKARELA/JKN) Plus KIS 0.67% Plus BPJS-PBI 1.33% Plus both KIS and BPJS-PBI 0.00% Only voluntary BPJS 16.83% None 81.17% PERCENTAGE OF HOUSEHOLDS WHO HAVE VOLUNTARY BPJS AND USED IT IN LAST 16.83% YEAR AVERAGE NUMBER OF HOUSEHOLD MEMBERS WHO OWN KIS 2.78 BPJS-PBI 2.75 Voluntary BPJS 2.77 43 The Economics of Clove Farming in Indonesia Figure 21: Access to Sufficient Food Always has su cient food 34.5 Usually has su cient food 53.5 Usually lacks su cient food 10.33 Always lacks su cient food 1.67 0 10 20 30 40 50 60 Percentage 4.2 – Food Security Most clove farmers report “usually” or always having access to sufficient food. Food security, as presented in Figure 21, is an important well-being–related issue among some farmers in Indonesia (e.g., Hanie and Tsegai, 2016) and among tobacco farmers in many parts of the world (Eriksen et al., 2015). Approximately 12 percent of individuals from clove-farming households reported serious food insecurity, which was comparable to what we found in the companion survey in several tobacco-growing regions though considerably less than in Lombok where food security was a serious challenge for many tobacco farmers. Less than one-third of clove-farming households grew their own food. Among the survey respondents, 32 percent reported producing their own food. The survey asked house- holds for how many months the cultivated food fed the household (see Table 17). Thirty percent reported that they grew enough for the entire year. Importantly, this does not necessarily equate to a lack of food security. In many cases in the FGDs, farmers reported growing other cash crops and using the monetary proceeds of the cultivation to buy food in the marketplace. This dynamic was similar to what others found in recent surveys of tobacco farmers (e.g., Chavez et al., 2016; Goma et al., 2015; Magati et al., 2016; Makoka et al., 2016). 44 // Household Well-Being Table 17: Staple Food by Number of Months Produced OBSERVATION MEAN MINIMUM LONGETIVITY OF FOOD SUPPLY 190 6.64 0.33 Month Freq. Percent <1 months 8 1.58 2 Months 3 17.37 3 Months 33 21.05 4 Months 40 9.47 5 Months 18 10.53 Months that staple food 6 Months 20 1.05 production lasts for household 7 Months 2 2.11 8 Months 4 0.53 9 Months 1 1.58 10 Months 3 0.53 11 Months 1 30 12 Months 57 29.69 Total 190 100 Clove farmers who cultivated more cloves were more likely to be food secure. We also examined food security through multivariate analysis. As a dependent variable we used a dichotomous measure of whether a household always had enough to eat (a “1” value). We used the previous literature on food security (Eriksen et al., 2015; Khisa, 2011), machine-learning methods and stepwise regression to identify the relevant independent variables from the survey results. In Table 18, we report only the statisti- cally significant coefficients, though the foundational analyses are available in our online appendices. We find that clove farmers who use more land to farm cloves are more likely to be food secure, though we find that the coefficient for the overall hectarage cultivated (for all crops) is negative and statistically significant. Household size is positively and significantly related to food security suggesting that larger households are more likely to be secure. 45 The Economics of Clove Farming in Indonesia Table 18: Logistic Regression of Determinants of Food Security VARIABLES COMPLETE CASE IMPUTED MISSING Hectares of land for clove farming 10.76*** 2.703 (4.061) (1.942) –3.300*** 0.381 Hectares of land cultivated (1.084) (0.823) –0.000947** –0.000236 Income from other crops (0.000439) (0.000215) 0.000386 0.000327** Total income ($100) (0.000304) (0.000141) Growing Sikotok cloves (vs. white –1.755 –3.097*** cloves) (1.515) (0.773) –1.941*** –1.154*** Unclear about input costs (0.605) (0.376) 1.705*** 0.537** Household size (0.450) (0.213) Observations 250 600 Standard errors in parentheses *** p < 0.01, ** p < 0.05, * p < 0.1 There was limited variation in the amount of staple rice crops grown by clove farmers. We examined the decision to grow the staple crop, typically rice. First, only 170 households grew rice out of the 597 households that reported rice as their staple food, suggesting that the majority of households buy their rice and grow other crops on their land. In Figure 22, we observe that most households regardless of their overall resources grow about the same amount of rice, and typically not a sufficient amount to feed the household. Recent reports suggest that the average individual in Indonesia eats between 114 kg (Wall Street Journal, 2015) and 163 kg (OECD-FAO 2015) of rice. We also examined broader decisions to grow all food crops. Accordingly, in Figure 22, we plot the number of households by their total resources and all of the food crops that they grow. We find a slightly different finding here wherein lower resource households are much likelier to grow other food crops to feed their families. In brief, the results suggest that the relation- ships between cropping decisions and food security are complex. 46 // Household Well-Being Figure 22: Total Household Income against Actual Rice Grown (bivariate probability density function, p.d.f. ) 47 SMOKING HAS BEEN A MAJOR CONTRIBUTOR O DISEASE BURDEN INDONESIA. SMO V LENCE PREVA AGE INDIVIDUAL HAS EXCEEDED 3 PERCENT SINCE 2 5 CHILD LABOUR O Child labour in clove farming occurred in only a small proportion of surveyed households. The International Labour Organization estimates that up to 1.5 million Indonesian children, aged 10-17, work on farms (ILO, 2015). A recent Human Rights N Watch report examined this dynamic on tobacco farms specifically (see HRW, 2016). Accordingly, we asked respondents about each household member’s contribution to farming cloves, including children. Recognizing that child labour is prone to underreport- ing because of possible social stigma and children contribute to the work on farms to a certain extent in most countries in the world, we particularly sought to determine if chil- dren were working consistently during school hours. In Table 18, in the columns under O “# of Case Worked,” we report any time a respondent indicated that a child (both under 18, and under 15, which is the more common age of adulthood in Indonesia) had worked at the specific agricultural activity pertinent to clove farming. In the second set of columns, we report the number of times that a child was reported working more than nine times in a month during official school hours at each activity. We were able to cross-reference working and school attendance because the survey asked whether the child went to school in the morning or afternoon, and then asked when each individual was working at clove farming at specific times. We found that a child working occurred in less than 10 percent of cases for harvest and postharvest activities. For children working consistently during school hours, the proportion dropped to 0.2% for harvest and postharvest activi- ties. It is important to reinforce that this was not a focus of the survey, and child labour is a very difficult phenomenon to measure in a survey (for example, the HRW (2016) report was observational of a random selection of tobacco farms). Table 19: Child Labour L # OF CASES WORKED # WORKED DURING SCHOOL HOURS <18 years old ≤15 years old <18 years old ≤15 years old Nursery 19 9 5 5 Land preparation & transplant 18 7 4 4 3 Field tending 26 10 5 5 Harvest 64 34 14 14 Postharvest (incl. curing) 70 34 11 11 Selling & marketing 15 7 1 1 Clove farming 98 56 20 20 2 49 50 // Introduction 6 LIMITATIONS The findings from this report provides a comprehensive picture of the farmers’ economic livelihoods in Indonesia, when considering these findings a few con- siderations are necessary. First, the survey coverage included Central Java and the two largest-producing areas of Sulawesi, which is about 70% of national production. Produc- tion is widely dispersed otherwise and it was too costly to include other areas. However, we do not have any a priori reasons to believe that clove farmers in other parts of the country are different than those in this survey. Many clove farmers were older than the average Indonesian so calculation of their replacement wage may be affected. The oldest household members—especially those over 60 years—would be most likely to replace clove farming with either other agri- cultural endeavors or household-based microenterprises and not wage-based employ- ment. More research into the age dynamic of clove farming is necessary. Similarly, the survey generally did not deeply examine the specific alternatives that clove farmers might undertake if they stopped clove farming. It would be useful in future research to explore what options farmers perceive and place them into other research or economic opportunities by region. 51 SMOKING HAS BEEN A MAJOR CONTRIBUTOR O DISEASE BURDEN INDONESIA. SMO PREVALENCE AMONG WORKIN AGE INDIVIDUAL HAS EXCEEDED 3 PERCENT SINCE 2 52 // Introduction 7 CONCLUSION Clove production is a widespread agricultural activity in Indonesia, particularly in Central Java and Sulawesi. Accordingly, policy makers are concerned that changes in the tobacco product tax structure, including the minimization of tiers and the increase of excise tax rates on kreteks, will negatively affect consumption and therefore the live- lihoods of clove farmers. The research undergirding this report finds several crucial fea- tures of clove farming that should help policy makers to better evaluate any potential disruptions to the clove supply and value chains that might affect these farmers. First, for most clove-farming households, clove farming is a small fraction of the household’s broader economic activity. Thus, a decrease in demand for kreteks that might lead to a decrease in demand for cloves is only going to affect most clove-farming households to a small extent economically. This effect could also be mitigated or elimi- nated if farmers re-allocated their resources to other economically fruitful pursuits and/ or if the government helped these farmers to shift to other activities. A “loss” should not be presumed, particularly considering that the evidence here suggests that most of these households have varied economic activities to which they could reallocate capital, land and/or labour. Second, clove farming does not appear to be highly lucrative for many of these farmers, particularly in certain regions (e.g., Minahasa). Even the gross margins reported by most households suggest that clove farming is, at best, a small side business for the vast preponderance of these households. Even in the unlikely scenario that farm- ers were not able to grow something else on their land and/or use the time previously allocated to clove farming for new economically productive activities, the loss to the vast majority of these households would be small because the contribution in real terms was typically small. In fact, it is quite possible that clove farmers could easily use their land just as or more productively growing other crops, and/or use their time more productively either economically or for other enriching purposes such as education, family, or recreation. Third, individuals in clove farming households, on average, are poorer than the average Indonesian household, which reinforces the idea that land and/or time might be better allocated to more economically lucrative pursuits, particularly in places where there are more viable options. 53 The Economics of Clove Farming in Indonesia RECOMMENDATIONS In terms of considering what might happen if there were a change in tax policy that af- fected demand for Indonesian cloves, it is important to reiterate that clove farming is not particularly profitable for most clove-producing households in at least a couple of major clove-producing districts. The government needs to research which alternatives could be viable and target these least-profitable areas for switching and help these farmers make successful transitions to growing other crops and/or economic activities. • In case of negative effects from higher tobacco taxes, the government could help clove farmers to identify other economically fruitful pursuits. {{ Research which alternative locally-grown crops are most viable on former clove-pro- ducing land; {{ Enhance supply and value chains for these alternative local crops; • Support research to identify the areas where clove production is least viable (e.g., Minahasa) and target those areas for alternatives first; • Improve agricultural extension services for non-clove crops; • Provide educational grants for improving clove farmers’ agricultural management skills; • Improve access to credit for farmers to grow other crops or pursue other small economic enterprises such as a small shop, fish farming, or a local transportation business. {{ Provide grants or low-interest loans to farmers willing to switch to alternative crops; and {{ Introduce financial programs or economic development programs as forms of transition assistance away from clove cultivation. 54 // Conclusion REFERENCES Briones, R. 2015. Small Farmers in High-Value Chains: Binding or Relaxing Constraints to Inclusive Growth? World Development 72: 43–52. Chavez, J. J., J. Drope, Q. Li, M. J. Aloria. 2016. The Economics of Tobacco Farming in the Philippines. Quezon City: Action for Economic Reforms and Atlanta: American Cancer Society. Eriksen, M., J. Mackay, N. Schluger, F. Islami, J. Drope. 2015. The Tobacco Atlas. Atlanta: American Cancer Society and New York: World Lung Foundation. Goma, F., J. Drope, R. Zulu, Q. Li, J. Banda. The Economics of Tobacco Farming in Zambia. Lusaka: University of Zambia School of Medicine and Atlanta: American Cancer Society. December, 2015. Hanie, E., and D. Tsegai. 2016 “Food Demand Analysis of Indonesian Households with Particular Attention to the Poorest.” ZEF—Discussion Papers on Development Policy No. 151. Bonn: Center for Development Research. Available at: https://papers.ssrn.com/sol3/papers. cfm?abstract_id=1945226 (Last accessed May 1, 2017). Human Rights Watch (HRW). 2016. “The Harvest Is in My Blood” Hazardous Child Labor in Tobacco Farming in Indonesia. New York: Human Rights Watch. Ikatan Ahli Kesehatan Masyarakat Indonesia (IAKMI). 2014. Tobacco Facts and Challenges. Jakarta: IAKMI. Indonesia Clove Farmer Association. Indonesia’s Clove Production in 2016 Stagnant. See http://www.indonesia-investments.com/tag/item31?tag=6378. (Last accessed May 3, 2017) Indonesian Family Life Survey. 2016. Santa Monica: Rand Corporation. See: https://www.rand.org/labor/FLS/IFLS/download.html Government of Indonesia. Minimum Wages by Province. Available at http://www.cascadeasia.com/2015/12/indonesias-2016-minimum-wag- es-by-province/ (last accessed May 2, 2017). International Labour Organization (ILO) Country Office for Indonesia and Timor- Leste, “Child Labour in Plantation,” undated, http://www.ilo.org/jakarta/areasofwork/ WCMS_126206/lang--en/index.htm (accessed May 1, 2017). 55 The Economics of Clove Farming in Indonesia Khisa, G. 2011. Tobacco production and food security in Bungoma District, Kenya: Effect of tobacco production on household food security in Malakisi Division Bungoma District, Kenya. Saarbrücken, Germany: Lambert Publishing. Magati, P., Q. Li, J. Drope, R. Lencucha, R. Labonté. 2016. The Economics of Tobacco Farming in Kenya. Nairobi: International Institute for Legislative Affairs and Atlanta: American Cancer Society. Makoka, D., A. Appau, R. Lencucha, J. Drope. 2016. Farm-Level Economics of Tobacco Production in Malawi. Lilongwe: Centre for Agricultural Research and Development and Atlanta: American Cancer Society. Organisation for Economic Co-operation and Development (OECD) and Food and Agriculture Organization (FAO). 2015. Agricultural Outlook 2015—Indonesia. See: http://www.oecd-ilibrary.org/agriculture-and-food/oecd-fao-agricultural-outlook-2015/ rice-projections-consumption-per-capita_agr_outlook-2015-table125-en (Last accessed, June 2017). Rachman, A. 2016. Indonesia Rushes to Save National Treasure. Wall Street Journal. June 28. Wall Street Journal. 2015. Indonesia’s Rice Consumption. Wall Street Journal blogs. See: https://blogs.wsj.com/briefly/2015/05/27/indonesias-rice-consumption-the-numbers/ (Last accessed, June 2017). World Bank. 2017a. Living Standards Measurement Survey. Washington DC: World Bank. See http://go.worldbank.org/UJ3E1I2850 (Last accessed on May 3, 2017). World Bank (2017b). The Economics of Tobacco Farming in Indonesia. Indonesia Tobacco Employment Studies. World Bank, Washington, DC. World Bank (2017c). The Economics of Kretek Rolling in Indonesia. Indonesia Tobacco Employment Studies. World Bank, Washington, DC. World Bank (2017d). The Economics of Tobacco Taxation and Employment in Indonesia. Indonesia Tobacco Employment Studies. World Bank, Washington, DC. 56 // Conclusion 57