82312 El Salvador Country Land Assessment World Bank El SALVADOR COUNTRY LAND ASSESSMENT 0 El Salvador Country Land Assessment World Bank El SALVADOR COUNTRY LAND ASSESSMENT Report Number: 82312-SV June 2012 (Revised) 1 El Salvador Country Land Assessment World Bank EXECUTIVE SUMMARY ......................................................................................................... 9 Focus of this Study ...................................................................................................................... 9 Methodology and Framework ..................................................................................................... 9 Agrarian Reform ....................................................................................................................... 10 Legal and Institutional Framework ........................................................................................... 10 Rural Land Tenure and Poverty Reduction ............................................................................... 11 Land Use and Agricultural Productivity ................................................................................... 12 Land Transactions in Cooperatives of the Agrarian Reform .................................................... 13 Urban Expansion –Extent, Tradeoffs and Disaster-Vulnerability ............................................ 13 Location of New Settlement and Public Policy Concerns ........................................................ 14 Development Control in AMSS ................................................................................................ 14 Responsiveness of Institutional Framework for Urban Land Development ............................. 14 Urban Expansion and Inclusion ................................................................................................ 15 Urban Land Prices and Affordability ........................................................................................ 15 Inclusion and the Regulatory Framework for Land Development ............................................ 16 Land Use and Deforestation ...................................................................................................... 16 Deforestation and Public Policies in the Agrarian Sector ......................................................... 17 Deforestation and Conservation Limits..................................................................................... 17 Land Market Trends .................................................................................................................. 18 Land Market Trends in the Cities of Santa Ana and Sonsonate and Contiguous Municipalities ................................................................................................................................................... 18 An Integrated Perspective of Land Use and Public Policy ....................................................... 18 Where are public policy objectives being achieved? ................................................................ 19 Where are public policy goals clashing? ................................................................................... 19 Institutions for mediating public policy conflicts and promoting land use efficiency .............. 19 Policy Options ........................................................................................................................... 20 ACKNOWLEDGEMENTS .......................................................................................................... 23 ABBREVIATIONS AND ACRONYMS ..................................................................................... 25 Chapter 1 Overview ...................................................................................................................... 26 1.1 The Importance of Land in the Development of El Salvador: Historical Context ......... 26 1.2 Objectives, Key Questions, Methodology, and Structure of this Report ....................... 27 2 El Salvador Country Land Assessment World Bank 1.3 Agrarian Reform ............................................................................................................ 30 1.4 Legal and Institutional Framework for Land Issues in El Salvador ............................... 32 1.5 Property Rights ............................................................................................................... 36 1.6 Land Tenure Informality Challenges ............................................................................. 37 1.7 Institutions and Outcomes .............................................................................................. 39 Chapter 2 Rural Land Tenure and Poverty Reduction .................................................................. 41 2.1 Overview of Chapter ...................................................................................................... 41 2.2. The Changing Structure of the Economy ....................................................................... 41 2.3 Equitable Distribution of Land Tenure and Access to Land .......................................... 42 2.4 Security of Land Tenure and Property Rights ................................................................ 46 2.5 General Discussion ......................................................................................................... 49 Chapter 3 Land and the Agrarian Reform Cooperatives .............................................................. 51 3.1 Overview of Chapter ...................................................................................................... 51 3.2 Context of the Agrarian Reform Cooperatives .............................................................. 51 3.3 Analysis of Land Use and Agricultural Productivity ..................................................... 52 3.4 Land Transactions in Cooperatives of the Agrarian Reform ......................................... 58 3.5 General Discussion ......................................................................................................... 61 Chapter 4 Urban Expansion – ....................................................................................................... 65 4.1 Overview of Chapter ..................................................................................................... 65 4.2 Urbanization trends ........................................................................................................ 65 4.3 The Extent of Urban Expansion 2001-2010 ................................................................... 66 4.4 New Settlement on Steep Slopes – Natural Disaster Vulnerability ............................... 69 4.5 New Settlement on viable agricultural land ................................................................... 71 4.6 Development Control in AMSS ..................................................................................... 73 4.7 General Discussion: Responsiveness of Institutional Framework for Urban Land Development ............................................................................................................................. 76 Chapter 5 Urban Expansion and Inclusion ................................................................................... 80 5.1 Overview of Chapter ..................................................................................................... 80 5.2 Urban Expansion and Broader Socio-Economic Vulnerability ..................................... 80 5.3 Urban Land Prices and Affordability ............................................................................. 83 5.4 General Discussion: Inclusion and the Regulatory Framework for Land Development 86 3 El Salvador Country Land Assessment World Bank Chapter 6 Land Use and Deforestation ......................................................................................... 91 6.1 Overview of Chapter ...................................................................................................... 91 6.2 Scale, Rate and Distribution of Deforestation ................................................................ 91 6.3 Deforestation and Public Policies in the Agrarian Sector .............................................. 96 6.4 General Discussion ......................................................................................................... 97 Chapter 7 Land Market Trends ................................................................................................... 100 7.1 Overview of Chapter .................................................................................................... 100 7.2 Land Market Trends in Ahuachapan, Santa Ana, and Sonsonate- 2001-2010 ............ 100 7.3 Land Market Trends in the Cities of Santa Ana and Sonsonate and Contiguous Municipalities .......................................................................................................................... 103 7.4 General Discussion ....................................................................................................... 104 Chapter 8 An Integrated Perspective of Land Use and Public Policy ........................................ 106 8.1 Overview of this Chapter ............................................................................................. 106 8.2 Where are public policy objectives being achieved? ................................................... 106 8.3 Where are public policy goals clashing? ...................................................................... 108 8.4 Institutions for mediating public policy conflicts and promoting land use efficiency . 111 Chapter 9 Policy Options ............................................................................................................ 114 9.1 Key Findings ................................................................................................................ 114 9.2 Policy Options .............................................................................................................. 114 Bibliography ............................................................................................................................... 119 Annex 1: Discrepancies in Estimates of Land Under Agriculture .......................................... 120 Annex 2: Details of Urban Expansion Analyses ..................................................................... 124 Annex 3: Methodological Details for Deforestation Analysis ................................................ 132 Annex 4: Details of Eco-System Change detection in hectares for the periods 1998 – 2008 and 2008 – 2011 ............................................................................................................................. 136 4 El Salvador Country Land Assessment World Bank TABLES Table 1.1 Land Redistributed and Beneficiaries of the 1980s Agrarian Reform.......................... 31 Table 1.2: Summary of Legislation related to Land Issues .......................................................... 33 Table 1.3: Key Land Administration Entities ............................................................................... 35 Table 1.4: Main forms of tenure informality in rural areas .......................................................... 38 Table 1.5: Main Forms of Tenure Informality in Urban Areas .................................................... 39 Table 2.1: Comparison of landless households in rural sector, 1971 and 2008............................ 42 Table 2.2: Land Redistribution Programs by Department ............................................................ 43 Table 2.3: Land Redistributed by Program ................................................................................... 44 Table 2.4: Agrarian Structure, 1971 and 2008 .............................................................................. 45 Table 2.5: Landholdings pending individual titling in ISTA ........................................................ 48 Table 3.1: Registry of Agrarian Reform Cooperatives by Source ................................................ 53 Table 3.2: Distribution of Agrarian Reform Cooperatives and Comparison Group by Department ....................................................................................................................................................... 53 Table 3.3: Summary Statistics - Cultivated Land ......................................................................... 55 Table 3.4: Yields in Basic Grains and Sugarcane ......................................................................... 56 Table 3.5: Public Auctions in Cooperatives of the Agrarian Reform ........................................... 59 Table 4.1: Percentage of Urban Growth 2000-2010 by Elevation................................................ 70 Table 4.2: Percentage of Urban Development in SS 2010 by Land Use Zone ............................. 74 Table 4.3: Spatial Distribution of Land Use Change Development Permission Applications in AMSS 2008-2010 ......................................................................................................................... 75 Table 5.1: Proportion of new urban growth 2001 to 2010 in the AMSS that was in Precarious Settlements .................................................................................................................................... 81 Table 5.2: Proportion of new urban growth 2001 to 2010 in the municipalities of Colon and San Juan Opico that was in Precarious Settlements............................................................................. 82 Table 5.3: Proportion of new urban growth 2001 to 2010 in Santa Ana that was in Precarious Settlements .................................................................................................................................... 83 Table 5.4: Average Monthly Incomes and Percentage of Population Living in Poverty 2005 and 2009............................................................................................................................................... 85 Table 5.5: Land Affordability in 2009 .......................................................................................... 86 Table 5.6: Minimum Plot Sizes in San Salvador (AMSS) by Settlement Type ........................... 87 Table 6.1: Municipalities with Deforestation Rates Greater than 50% between 1998 and 2008 . 94 Table 6.2: Municipalities with Deforestation Rates Greater than of 25 and 50% between 1998 and 2008 ........................................................................................................................................ 95 Table 6.3: Summary Results of Correlation Analysis of Deforestation ....................................... 96 Table 7.1: Land Market Transactions in the Departments of Ahuachapan, Santa Ana, and Sonsonate – 2001, 2005 and 2010 .............................................................................................. 101 Table 7.2: Land market data for the municipalities of Santa Ana, San Sebastian, Salitrillo Sonsonate and Sonzacate ............................................................................................................ 104 Table 9.1: Matrix of Policy Options ........................................................................................... 116 5 El Salvador Country Land Assessment World Bank FIGURES Figure 1.1: Property Rights Ranking of Countries in Latin America and the Caribbean ............. 37 Figure 2.1: Agriculture Value Added (% of GDP) ....................................................................... 42 Figure 2.2: Poor and Non-Poor's Access to Agricultural Land .................................................... 44 Figure 2.3: Access to Land and Tenure Security .......................................................................... 46 Figure 2.4: Land Tenure Security by Region................................................................................ 47 Figure 3.1: Box Plot of Total Land ............................................................................................... 54 Figure 3.2: Average distribution of land by use ........................................................................... 54 Figure 3.3: Distribution of cultivated area by crop ....................................................................... 55 Figure 3.4: Yields in Sugarcane by Farm Size ............................................................................. 57 Figure 3.5: Yields in Maize by Farm ............................................................................................ 57 Figure 3.6: Machinery, Inputs, and Credit .................................................................................... 57 Figure 3.7: Reserved and Sale Prices by Number of Bidders ....................................................... 60 Figure 3.8: Land Use of Plot Prior to Public Auction................................................................... 61 Figure 3.9: Sale Price by Type of Prior Land Use ........................................................................ 61 Figure 4.1: Urbanization Trends for El Salvador and its Neighbors ............................................ 66 Figure 4.2: Annual Percentage Change in Built-up Area per Person in Select Latin American Cities during the 1990s ................................................................................................................. 66 Figure 4.3: Urban Growth in San Salvador Metropolitan Area (AMSS) 2001-2010 ................... 67 Figure 4.4: Location of Satellite Municipalities of Colon and San Juan Opico ........................... 68 Figure 4.5: Urban Expansion of Santa Ana between 2001 and 2010 ........................................... 68 Figure 4.6: Urban Expansion of San Miguel between 2001 and 2010 ......................................... 69 Figure 4.7: Urban Expansion 2000-2010 by Slope in San Salvador and Satellite Municipalities 70 Figure 4.8: Urban Expansion 2000-2010 by Slope in Santa Ana and San Miguel ....................... 70 Figure 4.9: Area by Soil Type that became urban in San Salvador between 2000 and2010 ........ 71 Figure 4.10: Area by Soil Type that became urban in the Satellite Municipalities of Colon and San Juan Opico - 2000-2010 ......................................................................................................... 71 Figure 4.11: Area by Soil Type that became urban in Santa Ana between 2001 and 2010 ......... 72 Figure 4.12: Area by Soil Type that became urban in San Miguel between 2000-2010 .............. 72 Figure 4.13: Overlay of Urban Expansion in AMSS 2000-2010 with Land Use Planning Zones 74 Figure 4.14: Regional Comparison (2006) of Typical Time it takes to obtain all permits to convert land from rural to urban use ............................................................................................. 77 Figure 5.1: Location of Precarious Settlements in AMSS in Relation to Slope ........................... 81 Figure 5.2: Location of all Precarious Settlements in AMSS in the municipalities of Colon and San Juan Opico ............................................................................................................................. 82 Figure 5.3: Location of Precarious Settlements Santa Ana in Relation to Slope ......................... 82 Figure 5.4: Location of all Precarious Settlements San Miguel in Relation to Slope .................. 83 6 El Salvador Country Land Assessment World Bank Figure 5.5: Proportion of new urban growth 2001 to 2010 in San Miguel that was in Precarious Settlements .................................................................................................................................... 83 Figure 5.6: Spot Land Prices 2009/2010 ...................................................................................... 84 Figure 5.7: Urban land prices in San Salvador Metropolitan Area (US$ per vara cuadrada) ...... 85 Figur 5.8: Regional Comparison (2006) of Minimum Road Width (m) in a new legal subdivision on the urban fringe ........................................................................................................................ 88 Figure 6.1: Vegetation Cover for Natural Ecosystem for the Years 1998, 2008 and 2011 .......... 93 Figure 7.1: Box Plot of Land Sale Prices in the Departments of Ahuachapan, Santa Ana, and Sonsonate – 2001, 2005 and 2010 .............................................................................................. 102 Figure 7.2: Box plot display of average land sizes in sales in the Departments of Ahuachapan, Santa Ana, and Sonsonate – 2001, 2005 and 2010 ..................................................................... 103 7 El Salvador Country Land Assessment World Bank MAPS Map 6.1:Changes in the physiognomical structure of the ecosystems in El Salvador between 1998 and 2008 ............................................................................................................................... 91 Map 6.2: Changes in the physiognomical structure of the ecosystems in El Salvador between 2008 and 2011 ............................................................................................................................... 92 8 El Salvador Country Land Assessment World Bank EXECUTIVE SUMMARY Focus of this Study This Study assesses the alignment of land use, land tenure and land market outcomes in El Salvador with public policy aspirations in recent decades for efficient, inclusive and environmentally sustainable development in both urban and rural spaces. In doing so the Study indirectly gauges the effectiveness of land sector institutions in facilitating such developmental outcomes in agricultural production, urbanization, and forest management. Among the key questions with which the Government is grappling and which this Study attempts to answer are: To what extent have public policy interventions been effective in reducing rural inequality? What has happened to the land assets entrusted to Agrarian Reform Cooperatives under the last iterations of the Agrarian Reform? How rapidly are Salvadorian cities growing and how effectively has the supply of urban land kept pace with the growing demand associated with urbanization? To what extent have land use changes associated with urbanization and agricultural expansion been happening un-sustainably and in ways that are increasing natural disaster vulnerability? How extensive has been deforestation in the last decade? Methodology and Framework The Study uses empirical assessments of recent land use and market dynamism as a foundation for evaluating the effectiveness of land sector institutions and for generating policy options. Among other things, empirical assessments utilize a combination of temporal analysis of land use changes through change detection analysis of spatial images, inter-census comparisons of land tenure and production data, land market analyses, and cross-references with poverty data. Due to limitations in coverage of available data and resource constraints of this Study, the findings are neither comprehensive nor necessarily reflective of the totality of land use, land tenure and land market dynamism across the entire country. However, for the most part the Study maximizes the use of existing data and selectivity in the analysis is usually informed by strategic relevance. The way that land features in the development process provides a reference framework for this Study. Land as a productive resource ought to change uses efficiently in response to opportunities for greater productivity created by market forces and shifts in demographic and economic structures. Land also ought to be accessible to the vast majority of the population for a variety of purposes and its possession and use should be coupled with sufficient security of tenure to make productive and welfare-enhancing investments. Finally, as a finite resource, land uses also need to be sustainable. 9 El Salvador Country Land Assessment World Bank Agrarian Reform Chapter 1 briefly reviews some of the prominent struggles over land in El Salvador and outlines the salient features of today’s institutional framework for land governance. With the second highest population density in the western hemisphere, land is a scarce resource in El Salvador. Consequently, access to land and struggles over alternative uses of land have been at the center of the country’s social, economic, and political history. Until recently, the land tenure structure of El Salvador has been linked to the dynamics of export markets for agro-industrial commodities. After failed attempts to reduce land inequality through agrarian reform in the 1960s and 1970s, strong support for another reform materialized when a civil-military junta took power in 1979. But by this time, the civil conflict had started and the 1980 Agrarian Reform was launched as a strategy to neutralize the insurgency. The reforms of the 1980s occurred in three distinct phases with more than 85,000 families benefitting from land redistribution in excess of 300,000 ha. Land issues featured prominently in the 1992 Peace Accords that included a Land Transfer Program (PPT) through a willing- seller/willing-buyer model for the benefit of approximately 47,500 ex-combatants, renters, and tenants. In addition, organizations of landless peasants also benefited as part of a 1991 agreement. After seven years, the PTT distributed 3,305 properties to 36,100 beneficiaries covering an area of approximately 100,000 hectares. Overall, Agrarian Reform beneficiaries have had difficulty becoming successful farmers. The land plots transferred under PTT through a Land Bank were on average of lower agricultural value than those transferred during the 1980s programs, which undermined the agricultural potential of these new farms. Many beneficiaries had trouble repaying their loans with the debt eventually cancelled. The Land Bank was closed in 1998 and the backlog of properties was transferred to the Agrarian Transformation Institute of El Salvador (ISTA). Legal and Institutional Framework The legal framework related to land issues in El Salvador is largely contained in eleven key instruments. These are: (i) The Constitution, (ii) the Civil Code, (iii) the Urban and Construction Law, (iv) the Development and Territorial Planning Law for the Metropolitan Area of San Salvador, (v) the Environment Law, (vi) the National Protected Areas Law, (vii) Cultural Patrimony Law, (viii) the Restructuring of the Real Estate and Mortgages Registry Law, (ix) the Cadastre Law, (x) the Urban Titling Law, and (xi) the Agrarian Transformation Institute Law. In addition, there are a number of other institutional decrees and regulations which have a direct impact on land tenure rights and uses. A Territorial Planning and Development Law (decree 644 of 2011) was approved in March 2011, but this law will only become effective in August 2012. Since the early 1990s, El Salvador has made considerable progress in updating its legal and institutional framework for land administration. With the creation of the Centro Nacional de Registros (CNR) in 1994, El Salvador was a pioneer in Latin America in merging the legal 10 El Salvador Country Land Assessment World Bank registry with the geographical description of land parcels (urban and rural) under a single entity. This registration methodology, known as folio real, replaces the centuries-old custom of registering real property rights under a deed-based or individual-based system. Today, CNR is one of the most sophisticated registries in the region, is entirely self-financed, and has significant approval from the public for the efficiency and transparency of its services. More than two-thirds of the country’s land parcels have been systematically surveyed and today El Salvador enjoys a top-third ranking of 54 out of 183 nations in the 2012 ranking of the World Bank Group’s Doing Business Indicator for Registering Property. Despite the relative strength of some aspects of the institutional framework for land governance, significant challenges remain related to land tenure informality. As in most of Latin America, informality in access to land, use, and market transactions is quite common in El Salvador. Since 2000, ISTA has registered land invasions on 53 rural properties. There are several land regularization initiatives to address the issues of informality and land invasions in rural areas. Likewise, despite the long existence of the Urban Development and Construction Law, much informal urban development, subdivisions and sales have taken place giving rise to at least three regularization programs for urban lands. There has been discussion for some time on passing a special law to address the issues of approximately 300,000 plots in informal subdivisions mostly in urban areas, but little progress has been achieved. The most significant bottlenecks in regularizing urban parcels include mortgages which have not been cancelled in the Registry, unregistered subdivisions or amendments to existing parcels, and inaccurate liens on properties Rural Land Tenure and Poverty Reduction Chapter 2 asks the question ‘How effective have public policy interventions , including the Agrarian Reform, been in reducing rural inequality and tenure insecurity?’ El Salvador has largely succeeded in reversing the extreme inequality in access to land which has characterized most of its history. Land distribution has become more equitable in the last four decades with the Land Gini falling from 0.81 to 0.75 and the percentage of landless rural households from 41% to 18%. The departments of La Union, San Vicente, and Ahuachapan had the largest improvements. Land redistribution programs under the Agrarian Reform and the Land Transfer Program (PTT) redistributed 482,882 hectares of land or about 31% of the estimated agricultural land area in the country. The departments of Usulutan, La Libertad, and Sonsonate account for 41% of redistributed land. Between 2005 and 2009, the average farm size for the non-poor decreased by almost half while the average size for the poor though smaller in absolute terms, was maintained. In fact, at present, the main challenges have to do with land fragmentation and alternative uses of land. The current fragmentation of agricultural land indicates that rural households are depending on land in different ways than they did in the past. Rural households long term economic prospects are tied to income earning opportunities related to off-farm supplementary 11 El Salvador Country Land Assessment World Bank employment, urbanization and remittance transfers1. In this context, government policies should not discourage any of these activities which may be seen as rational choices of rural households. Overall, perceived land tenure security is high for the population as a whole although the formalization of property rights of the land redistributed under the Agrarian Reform is still a pending issue. ISTA is yet to legalize individual titles to over 170,000 landholdings. There were many anomalies in the ISTA titles. The cooperatives’ titles were held until the 30 - year period of co-management was over or debt was repaid. The titles were listed in the ISTA President’s special volume, instead of the notaries’ protocol as private conveyances must be. Once the titles were granted, beneficiaries often did not know that they had to register them in the Registry of Real Property and Mortgages or they did not receive support to complete the process. The distribution of land to groups of individual beneficiaries under the modality of joint titles (proindiviso) also made legalization more difficult. A similar situation occurs with members of the cooperatives that want to obtain title to their individual plots. The lack of reliable and updated property records is a main constraint to ongoing regularization efforts. Updating the inventory of properties and the cadastral information in ISTA’s land records is an important investment to improve land governance. The systematization of existing land records in a digital inventory would provide ISTA with a tool to better track land transactions. A digital inventory would also improve transparency and reduce transaction costs by making the information more easily available to other institutions and the general public. Furthermore, ISTA could invest in a more systematic scheme to update the cadastral information of records that do not comply with the current technical specifications instead of trying to solve each case individually.. These activities require not only substantial investments in technology and personnel, but also an appropriate institutional coordination with the CNR. In particular, the CNR can only register documents that meet the legal requirements to maintain legal securtiy. Land Use and Agricultural Productivity Chapter 3 explores what has happened to the lands transferred to Agrarian Reform cooperatives under the last iterations of the Agrarian Reform . These cooperatives have more land available and are more likely to be landowners than “Other Commercial Enterprises” (as defined by the agricultural census) although they cultivate a lower proportion of that land. Based on available data, the average land size of a holding for “Agrarian Reform Cooperatives” is 328 hectares compared to 73 hectares for “Other Commercial Enterprises.” Agrarian Reform Cooperatives also produce almost exclusively on land owned by them (98%). Agrarian Reform Cooperatives on average cultivate 10% less of their holdings than the comparison group (controlling for land size). 1 This has some similarities with the situation in China today whereby average farm size is 0.6 ha although there rental of farm machinery is helping to raise productivity and government is encouraging consolidation of farms - the latter being a less feasible proposition in El Salvador. 12 El Salvador Country Land Assessment World Bank Agrarian Reform Cooperatives have higher productivity in the production of maize and sugarcane than Other Commercial Enterprises. These differences, however, become insignificant when land size is considered. Sugarcane and coffee are the two most important crops for both Agrarian Reform Cooperatives and Other Commercial Enterprises, but coffee is cultivated by a larger number of producers. Agrarian Reform Cooperatives and Other Commercial Enterprises have similar access to machinery, inputs, and credit once the size of their land holdings is taken into account. In particular, the difference in access to tractors, harvesting equipment, fertilizer use, and use of credit is statistically significant. These differences, however, become insignificant when land size is included, which suggests that the scale of operations may be the factor correlated with higher use. Also, when land size is taken into account, the results show that Agrarian Reform Cooperatives use 11.9% less spraying equipment than the comparison group. Land Transactions in Cooperatives of the Agrarian Reform Stringent restrictions were placed on land owned by Agrarian Reform Cooperatives to avoid fragmentation and maintain economies of scale. When Phase I started, cooperatives were not allowed to sell, rent or subdivide their lands. These restrictions were progressively lifted during the 1990s, when the country adopted a policy aimed at liberalizing and reactivating the reformed sector (Programa Promoción de la Reactivación Económica y Social). The selling of cooperative land to non-members requires a non-judicial public auction of which there were 78 between 1997 and 2006 but despite the legal reforms of the 1990s, public auctions are not an effective mechanism for cooperatives to participate in land market transactions. The average size of plots successfully auctioned through the formal process was 68 hectares and the average sale price per hectare was US$8,459 (in 2005 dollars). Cooperatives have not reported using this mechanism since 2006. Also, most of the auctions involved a single bidder. In these cases, the cooperatives obtained uncompetitive prices. In 12 of the 20 auctions where municipal land price data was matched, the sale price in the auction was significantly lower than the average sale price for a rural property in the same municipality. Meanwhile, cooperatives continue to use informal mechanisms to sell land to external agents Urban Expansion –Extent, Tradeoffs and Disaster-Vulnerability Chapter 4 asks the question ‘how has land governance in El Salvador responded to the challenges of urban land supply in the last decade?’ Urban expansion occurred more rapidly outside of the capital in the first decade of this century. The urban footprint of the Metropolitan Area of San Salvador (AMSS) grew by approximately 30% from about 115 square km to over 148 square km. The rate of expansion was comparable to global rates in the 1990s, however, very significantly, two satellite municipalities of the AMSS, Colon and San Juan Opico, both in the Department of La Libertad, went from being completely undeveloped to 26 square kilometers of urban development in this same period. Growth in Santa Ana and San 13 El Salvador Country Land Assessment World Bank Miguel, the second and third most important cities in the country respectively, was more rapid. Santa Ana grew very substantially by 120% from approximately 17 square km to over 37 square km. San Miguel grew by approximately 60% from approximately 24 square km to over 38 square km. Location of New Settlement and Public Policy Concerns Despite the relative elasticity in the supply of urban land, a number of challenging policy issues have emerged over the past two decades on account of the location of the new growth. In all three cities there is a significant amount of growth that is occurring on steep slopes. This is especially so in San Salvador, and Santa Ana with San Miguel being in much more low-lying terrain. Given its comparatively high ratio of population to land area, the opportunity cost of agricultural production is expected to be quite high in El Salvador. Market pressures to convert agricultural land to more productive uses are expected to be strong in the future. A significant part of the urban expansion of the last decade has occurred on land with good agricultural potential (Classes I or II soils). This proportion was greatest for Santa Ana (almost one half), then San Miguel (one quarter) and least for San Salvador (about one fifth). Also some of the urban expansion in San Salvador and Santa Ana has displaced forests. This trend raises questions about the effectiveness of local land use planning mechanisms in all cities but in secondary cities in particular. Development Control in AMSS The vast majority of urban expansion in the AMSS between 2000 and 2010 occurred in accordance with the Master Plan although urban expansion did infringe upon risk-prone areas that were zoned for maximum protection especially in the south and occurred upon land zoned for agricultural development in the north east. Almost 14% of the city’s built footprint by 2010 had occurred in areas not zoned for urban use. Analysis of the land use change/intensification applications in recent years suggests that a significant part of the urban expansion may have occurred without apriori formal development permission. Responsiveness of Institutional Framework for Urban Land Development The initial findings suggest that rather than adopt policies that were either resistant or indifferent to urbanization, in the past two or three decades the Government of El Salvador adopted a more proactive approach particularly through the creation of the Planning Office of the San Salvador Metropolitan Area (OPAMSS) to act as an umbrella agency. Feedback from private sector developers and other real estate professionals is that this coordinating mechanism in San Salvador has provided a predictable and fairly efficient framework for land development. Moreover, a regional comparison of the typical time it takes to obtain all building permits to convert land from rural to urban use saw San Salvador rating very favorably at 2.5 months on average in 2006. Long times for planning permission usually mean lesser elasticity in land supply leading to less affordable prices and more informal development. 14 El Salvador Country Land Assessment World Bank However, there is the risk that shorter times may imply less than adequate scrutiny of the sustainability of current trends in land use conversion. On the other hand, urban planning and development control outside of San Salvador is much weaker and construction permitting processes can reportedly take up to two years. Without the required permits, subdivided plots cannot be registered at CNR or used as collateral for loans in commercial banks. There are few direct links or coordination between the Vice- Ministry of Housing and Urban Development and the municipalities of San Miguel, Santa Ana, and those in the San Andres valley. Urban Expansion and Inclusion In Chapter 5 the extent to which urban spatial expansion in El Salvador has been occurring in an inclusive way is explored. In the main cities, more than a third and sometimes close to a half of all urban growth in the last decade occurred in precarious settlements, with a majority being in settlements of extreme or high vulnerability in Santa Ana and San Miguel. Almost half (45%) of AMSS’s and Santa Ana’s new urban land growth in the first decade of this century occurred in precarious settlements which typically lack one or more basic urban utility service as well as well as having houses constructed from unstable materials. In AMSS about half of this precarious growth (44%) was in settlements of extreme or high vulnerability whereas in Santa Ana the equivalent proportion was two thirds (66%). For San Miguel more than a third of its new urban land growth occurred in precarious settlements with almost three quarters (72%) occurring in areas of high or extreme vulnerability. In the two main satellite municipalities of the AMSS, Colon and San Juan Opico, more than a third (38%) of the urban growth occurred in precarious settlements with more than a third of this (37%) in settlements of extreme or high vulnerability. Urban Land Prices and Affordability Based on available land market and income distribution data, a typical small plot at the low end of the market is relatively affordable in the AMSS, moderately affordable in Santa Ana and largely unaffordable in San Miguel. These results are based on a working definition of affordability being that the cost of the land does not exceed three times annual household income. In San Salvador (AMSS), one quarter of households in the corresponding Department would find land prices at the lower end of the formal market unaffordable. In Santa Ana this proportion grows to 40% and in San Miguel it is much higher at 80%2. These trends may partially reflect “deflation’ in the 2000s which followed the boom of the 1990s. 2 Notably the San Miguel analysis was made on the basis of very few available sub-division prices. However local stakeholders perceived the reported prices to be realistic estimates for that sub-market. 15 El Salvador Country Land Assessment World Bank Inclusion and the Regulatory Framework for Land Development These trends are a cause for public policy concern as they suggest that a significant part of El Salvador’s urban expansion is occurring with considerable socio-economic vulnerability. Moreover, an apparent correlation between the location of precarious settlements and steep slopes in some parts of the AMSS and Santa Ana was also noted, exposing a double-edged vulnerability for some. Still, by at least one measure, three quarters of the population of San Salvador Department can afford a typical small plot in the tail end of the formal market in the metropolitan area which suggests that some aspects of the regulatory framework are not over- prescribed. For example, a key regulation, the minimum plot size, starts as low as 75 square meters in AMSS and can be less than 100 square meters in Santa Ana and San Miguel. A regional comparison of another key regulatory provision, the minimum road width in urban subdivisions also shows San Salvador in a favorable light. Normally high public space requirements including road width requirements translate into less saleable land in the subdivision and thereby materialize as higher per unit costs of land for consumers. Informality offers entry into a system at different price points and the mixed storyline of this Chapter is consistent with the prevalence of informal residential subdivisions that reportedly number over 300 nationally. Given the affordability indicators observed in this Chapter, it is likely that significant segments of El Salvador’s urban population are gaining access to urban spaces by entering at price points below those on offer in the formal market i.e. by purchasing plots in the informal sector, most commonly in informal sub-divisions. Given that several key regulatory measures for land development in El Salvador are realistically prescribed, low incomes may continue to be the binding constraint sustaining informal markets for some time. Applications for development permission after the fact of land conversion are reportedly common. The efficiency of the regularization process3 that follows, if it follows, will be a key determinant of how long these vulnerable segments of the population remain in situations of compromised welfare. Land Use and Deforestation Chapter 6 presents the findings of the original analyses of land use in relation to deforestation. During the period 1998 - 2008 a total of about 44.000 thousand hectares of natural forests were lost at an average annual deforestation rate of 1.5%. By 2008, the rate dropped off significantly with only 1334 ha of natural forest lost during the period 2008 - 2011 at 0.17% per year. While the rate in this second period is significantly lower, the fact that there were only three points of measurement in 13 years does not preclude the 3 Among the most significant bottlenecks in regularizing urban parcels are included mortgages which have not been cancelled in the Registry, unregistered subdivisions or amendments to existing parcels, and inaccurate liens on properties. It is worth noting that the Salvadoran Registry System is declarative, so there is no obligation to submit documents for registration to the Registry. 16 El Salvador Country Land Assessment World Bank possibility that the trend may have started earlier than 2008. A comparison with FAO figures suggests that the sharp decline in annual deforestation rates likely started after 2005 but no later than 2008. It should also be noted that the degree of resolution of the spatial analysis did not necessarily allow for detection of some types of deforestation such as the loss of shade coffee. The conversion of forests was mostly in lowland tropical deciduous broad-leaved forest, on well drained soils and most of the lost forests were converted to annual crops and pastures for what appears to be a subsistence farming landscape. Only field verification during the growing season can make this assessment with confidence including the identification of the specific crops. However, it is likely that in relatively steep hilly terrain, the deforestation has been carried out to create grazing land, whereas in more level terrain, the forest may have been cleared for the preparation of agricultural fields. During both periods there was some clear-cutting of mangroves for the creation of shrimp farms, but the sizes of the losses were not all that large with 91ha and 32 ha respectively for each period. Finally, during the periods of the evaluations, there was no change towards recovery from productive land use to natural ecosystems. All changes were conversions form natural ecosystems to crops, grazing and other productive land use. Deforestation was greatest in the north-west, central and south-east regions of the country. It was almost non-existent in the South-West and only minor in the North-East. There were eight municipalities with deforestation rates greater than 50% and twelve between 25% and 50%.4 The degree of thinning of the forests through selective logging of mature trees and the collection of firewood, the main cooking fuel for rural El Salvador, is not known. Deforestation and Public Policies in the Agrarian Sector The causes of deforestation and of the decrease in average annual deforestation rates between the periods 1998 – 2008 and 2008 – 2011 cannot be defined without further knowledge of the socio-economical processes that took place in rural El Salvador. The potential role of land redistribution programs, subsidized distribution of seed and fertilizers and agrarian structure were explored by performing correlation analyses but none of the associated variables was found to have high value or statistically significant correlation. Deforestation and Conservation Limits It is noteworthy that a significant part of the deforestation in recent years occurred within formally demarcated Conservation Areas that were developed and delineated in 2004 as part of the “Plan Nacional de Ordenamiento y Desarrollo Territorial (PNODT)” - 4 Municipalities with baseline forests in 1998 of less than 1000ha were excluded from the analyses. 17 El Salvador Country Land Assessment World Bank the “National Plan for territorial zoning and development”, but for which no further regulation has been defined legally. Also in practice, no specific land use practices or conservation oriented management has been promoted by the Government outside the legally acquired protected areas. Land Market Trends Chapter 7 analyzes available land market data in three Departments, Ahuachapan, Santa Ana, and Sonsonate, to identify trends and land use dynamism in the first decade of this century. Prices for land sales in all three departments remained relatively stable between 2001 and 2010. The average size of properties being sold has rebounded in urban land markets and decreased in rural land markets between 2001 and 2010. This trend in rural land markets confirms the fragmentation of landholdings that was observed in the comparison of the 1971 and 2008 agricultural census data. Overall, land markets in the departments of Ahuachapan, Santa Ana, and Sonsonate experienced more dynamism in the rural than the urban sector. The number of registered transactions increased the most in rural areas compared to urban ones. Prices of land sales in rural areas also show an upward trend, with notable outliers in places contiguous to large urban areas. Land Market Trends in the Cities of Santa Ana and Sonsonate and Contiguous Municipalities Land prices in rural areas contiguous to the cities of Santa Ana and Sonsonate behaved differently from prices in other municipalities. The municipality of Santa Ana had an average price in urban areas of US$55 per square meter in 2001. The price in urban areas grew to US$63 and $US81 per square meter in 2005 and 2010 respectively. The mean price in rural areas of this municipality fell in 2005 and 2010 compared to their base in 2001. In contrast, the nearby municipality of San Sebastian Salitrillo experienced a significant increase in land sales of rural properties from 18 in 2001 to 773 in 2005. Average prices in rural areas also increased substantially from US$1.53 per square meters in 2001 to US$7.5 and US$13.4 per square meter in 2005 and 2010 respectively. The dynamism in the municipality of San Sebastian Salitrillo confirms the pattern of spatial expansion of the city of Santa Ana which physically expanded into this municipality during the period in which this land market dynamism was observed In the department of Sonsonate, the municipality of Sonzacate, which is contiguous to the city of Sonsonate, had the highest recorded value for rural properties in 2001 and 2005, and the second highest in 2010. These trends suggest that there is high dynamism in rural land markets nearby large cities, and that the level of development that is occurring on these lands is perceived to be of high value. Normally markets capitalize information about public policy including projected infrastructure expansion as well as planned private sector investments. An Integrated Perspective of Land Use and Public Policy The penultimate Chapter, Chapter 8, looks at the Study’s empirical findings from a more integrated, cross-sectoral perspective so that their implications for public policy are better 18 El Salvador Country Land Assessment World Bank understood. The Chapter is structured in three parts. The first section identifies and discusses areas in which there appears to be congruence between public policy and outcomes on the ground. Next the focus is placed on areas where this congruence is lacking. Finally, a brief assessment is made of the existing institutions that are meant to mediate conflicts in land use in the national interest. Where are public policy objectives being achieved? The first section identifies and discusses areas in which there appears to be congruence between public policy and outcomes on the ground. These include: (i) the elasticity of supply in conversion of rural and peri-urban land to urban use as evidenced by the dominant pattern of infrastructure preceding settlement as well as by moderate price trends particularly in San Salvador and Santa Ana; (ii) the effectiveness of institutional mechanisms for mediating competing demands for peri-urban land in contiguous AMSS with the notable exceptions of less regulated growth in the satellite municipalities of Colon and San Juan Opico and the fact that zoning did not adequately exclude steep slopes around AMSS from urbanization; (iii) Reduction in rural land-based inequality due to the combination of active policies and market forces; and (iv) Stemming of deforestation rates in recent years whereby the rate fell off drastically at least since 2008 although the link to public policy is not clear. Where are public policy goals clashing? The second section places focus on areas where public policy congruence is lacking. They include: (i) the limitations of land use zoning and development control mechanisms in directing urban settlement away from areas vulnerable to or exacerbating, natural disaster damage, especially steep slopes; (ii) the shortcomings of the same mechanisms and participatory processes to actively deliberate on the potential loss of prime agricultural lands to urban development, especially in Santa Ana and San Miguel; (iii) the spatial concentration of the urban poor whereby precarious settlements many of high or extreme vulnerability, form a large core of the new urban settlement, in a context of rapidly growing urban crime and violence; (iv) the inability of the more egalitarian land tenure structure created by the Agrarian Reform to enhance agricultural productivity despite the fact that Agrarian Reform Cooperatives have more land available and enjoy similar access to machinery, inputs, and credit as other commercial enterprises; (v) the inconsistency of the high degree of fragmentation of rural plots that exists today with public policy aspirations to eliminate rural poverty predominantly through increasing agricultural income; (vi) the lack of effective land governance provisions to protect forests and the environmental services that they provide at least between 1998 and 2005 with the consequent increase in natural disaster risk exposure; and (vii) the tension between expanding land access and formal security of tenure whereby the perceived high transaction costs of formal land conversion and subdivision create a barrier to efficient, formal market-based responses to the changing nature of demand for land especially during the civil war leading to an estimated 300,000 plots in informal subdivisions and hindrances to the efficient transfer of agrarian reform lands. Institutions for mediating public policy conflicts and promoting land use efficiency Finally, the Study presents a brief assessment of the existing institutions that mediate conflicts in land use in the national interest. These include: (i) Land Use Zoning and Development Control instruments with the potential for reconciling the land use imperatives 19 El Salvador Country Land Assessment World Bank associated with different development and sectoral agendas but which are insufficiently deployed in El Salvador, especially outside of the capital where urban expansion is most rapid; (ii) Land Tenure Regularization Initiatives (which are an ex-post mechanism for reconciling land use conflicts where informal occupation or use is later deemed suitable to continue) but progress in El Salvador has been slow prolonging compromised welfare for vulnerable communities; (iii) Land Information Systems and Governance, an arena where the reliability, accessibility and comprehensiveness of land information has seen significant improvement in the last two decades although the systematic use of that information in an integrated platform to support decision making and public inquiry remains limited; (iv) Property Taxation which though capable of incentivizing more efficient use of land and boosting revenues for municipal services, remains non-existent in El Salvador; (v) The new Law for Territorial Planning and Development which provides a common framework for land use across the country’s territory but whose implementation will be challenging not only from the perspective of needed local resources but also because it may induce resistance from some municipalities on the grounds that it counters the broader constitutional and public policy provisions for decentralization and subsidiarity. Many of the adverse developmental outcomes observed in this Study can be attributed in part to inadequate reconciliation and analysis of existing data sources and weak inter- Agency coordination within the Government. For example, this Study found significant discrepancies in estimates of the current stock of agrarian reform lands. Additionally, as noted above, weaknesses in developmental control mechanisms have contributed to some glaring deficiencies in promoting sustainable land use conversion. Policy Options The final Chapter of the Study presents policy options for consideration by the Government of El Salvador and the country’s civil society organizations in order to address the key challenges related to land tenure, land use, and territorial planning. The matrix of policy options that follows encompasses several policy instruments, including national legal/regulatory changes, taxation, municipal level planning measures, and increased civil society and private sector engagement. Public Policy Policy Option Tradeoffs Possible Challenges Concern to Implementation Backlog of untitled Adopt temporary measures for May encourage Requires careful analysis agrarian reform lands systematic regularization of emergence of old of legal and regulatory lands by area, rather than case disputes over rights options by case Limitations on Streamline the procedures for Some rural lands may Elimination of restrictions transferability of agrarian regularizing land rights exit agriculture may require legislative reform lands and those of altogether action residents of informal urban Reduce obstacles to voluntary subdivisions direct transfers from Land may again cooperatives to individuals become concentrated holders, and from absentee under a few producers owners to current holders Provision of technical assistance to agrarian reform beneficiaries based on new 20 El Salvador Country Land Assessment World Bank technologies and market opportunities to facilitate better informed decision making about selling or retention of agrarian reform lands Encroachment of urban Direct future urban expansion Possibly higher Costs of enforcement expansion onto good away from areas with high infrastructure costs quality agricultural land potential soils Some high quality areas And slower are already partially commercial developed development Increased commuter times Insufficient land use Create a shared pool of Loss of some Inter-municipal planning and development qualified technical personnel municipal autonomy competition for resources control capacity in and land information analysis and talent secondary cities and capability for contiguous satellite municipalities municipalities Not clear to which existing tier of government institutions such a shared resource can be attached Most contiguous spatial development is not inter- institutional Strengthen land information If undue emphasis is Capacity to use modern systems for decision making placed on highest systems will require including reliance on land accuracy data, it may strengthening market and spatial trends as delay crucial well as disaster risk mapping investments in to guide urban planning and infrastructure and prioritize infrastructure other public goods investments Decentralization of Introduce a property taxation Reduced financial Land taxes are unpopular responsibilities for land system which returns all or control from the development / service most collections to central government Land values are not expansion to weaker municipalities systematically recorded municipalities Administrative costs can be unjustifiably high Rapid horizontal Raise the permitted ratio of Increased urban Costs of enhancing expansion of cities often total built space to land area congestion, traffic, infrastructure and services onto steep slopes and unless adequately capacity to support higher agriculturally viable land addressed density population Cultural aversion to high rise living including residual fears from past earthquakes Increase enforcement in Restricts the High costs of enforcement, maximum protection zones development space, particularly for the poor, possibly driving land if affordable alternatives prices upward are not available 21 El Salvador Country Land Assessment World Bank Expedite regularization of land Discourages vertical Ownership status of lands titles to vacant land within city growth may be in conflict perimeters Introduce a vacant land tax in May encourage Administrative cost may urban areas cosmetic development be high to avoid tax Spatial concentration of Introduce incentives to May discourage some Requires broad and low-income housing promote mixed-income sub- commercial genuine political divisions such as tax breaks; development commitment for effective market guarantees; implementation contributions of public land Encourage greater participation May distract attention Requires sustained by civil associations to lobby away from increased encouragement from for improved services integration of central and municipal neighborhoods governments Multiple land use Use “Guided Land May increase costs of Limited capacities for conflicts associated with Development”5 to provide a development and consultation, unguided spatial expansion framework for more systematic infrastructure implementation and of secondary cities and sustainable expansion enforcement Use “Land Readjustment”6 to Land prices within Legislative framework for provide a framework for more land readjustment this tool may require systematic and sustainable zones may rise strengthening expansion although they may decrease in the broader area due to more elastic supply 5 Guided Land Development is a policy that involves the identification and setting aside of rights of way around the periphery of a city to create a basis for more systematic urban expansion. More Details can be found in Memo to the Mayor, World Bank 2011. 6 Land Readjustment or Land Pooling is a policy tool whereby land owners voluntarily contribute part of their land holdings to accommodate the rights of way for servicing their land while often also creating surplus vacant plots. The landowners receive a smaller but more valuable, better serviced plot and government’s land acquisition costs are significantly reduced. More Details can be found in Memo to the Mayor, World Bank 2011. 22 El Salvador Country Land Assessment World Bank ACKNOWLEDGEMENTS This Study was prepared under the overall guidance of Ethel Sennhauser, the World Bank’s Sector Manager for Agriculture and Rural Development at the Latin America and Caribbean Region. Robin Rajack was the Task Team Leader for the Study. The team also comprised Fernando Galeana, Jorge Muñoz, Katie McWilliams, Marcial Morataya and Daan Vreugdenhil. Anna Corsi, Sarah Antos and Monica Salazaar also contributed. The peer reviewers of the study were Edward Cook, Klaus Deininger, Edgardo Mosqueira, and Alexandra Ortiz. The team was assisted by Rosario Arevalo, Claudia Emilia Echeverria, Jania Ibarra, and Ketty Morales. The Study Team thanks Felipe Jaramillo, Country Director for Central America, Ede Jorge Ijjasz-Vasquez, Director for Sustainable Development in Latin America and the Caribbean and Fabrizio Zarcone, Representative for El Salvador, for their support. Many thanks are also due to Gregor Wolf, Sector Leader for Sustainable Development in Central America. The Team is also grateful for support from the Food and Agricultural Organization of the United Nation’s Cooperation Program (FAO/CP) and particularly Fabrice Edouard. The FAO/CP support enabled the Study Team to document and analyze legal and institutional aspects of land governance in El Salvador. The Study Team gives special thanks to Alexander Segovia, Technical Secretariat of the Presidency; main counterpart of the study. The Study Team also gives thanks to Pablo Alcides Ochoa, President of the Salvadorian Institute for Agrarian Transformation (ISTA); José Enrique Argumedo, Executive Director of the National Center for Registries (CNR); José Roberto Góchez, Vice-minister of Housing and Urban Development; Guillermo López Suárez, Minister of Agriculture and Livestock (MAG); Margarita Minero Executive Director of the Organization of Planning in the San Salvador Metropolitan Area (OPAMSS), and Herman Rosa Chávez, Minister of Environment and Natural Resources (MARN). These agencies contributed with information to the study: MAG provided data on the 2008 Agricultural Census, Agrarian Reform cooperatives and land auctions; CNR provided data on land market transactions and Agrarian Reform properties; MARN provided land use cover maps; ISTA data on Agrarian Reform’s land transfers; OPAMSS provided data on land permits and land use; and the General Directorate of Statistics and Census provided the 2005 and 2009 Annual Household Surveys. The team benefited greatly from the workshops and working meetings held in San Salvador in June 2011, January 2012 and June 2012 attended by officials from the institutions mentioned above as well as from the Vice-Secretariat for Territorial Planning and Decentralization (SSDT), the Corporation of Municipalities of the Republic of El Salvador (COMURES), the Social Investment Fund for Local Development (FISDL), the Liberty and Progress Institute (ILP) and the Office of Planning of the San Andrés Valley. 23 El Salvador Country Land Assessment World Bank The Study Team also thanks the National Real Estate Association of El Salvador for providing data on land prices and the United Nations Development Program for providing the geo- referencing of precarious settlements. The findings, interpretations, and conclusions expressed in this volume do not necessarily reflect the views of the Executive Directors of the World Bank or the governments they represent. 24 El Salvador Country Land Assessment World Bank ABBREVIATIONS AND ACRONYMS AMSS (Área Metropolitana de San Salvador) Metropolitan Area of San Salvador AOIs Areas of Interest AUP (Asentamientos Urbanos Precarios) Precarious Urban Settlements BCIE (Banco Centroamericano de Integración Económica) Bank for Central American Integration CAPRA Central American Probabilistic Risk Assessment CNR (Centro Nacional de Registros) National Registry Center COMURES Corporation of Municipalities of the Republic of El Salvador DEM Digital Elevation Model EHPM (Encuesta de Hogares de Propósitos Múltiples) Household Survey for Multiple Objectives FINATA (Financiera Nacional de Tierras Agrícolas) National Financier of Agricultural Lands FISDL Social Investment Fund for Local Development FLACSO (Facultad Latinoamericana de Ciencias Sociales) Latin American Faculty of Social Sciences FMLN Frente Farabundo Martí para la Liberación Nacional FSV (Fondo Social para la Vivienda) Social Housing Fund FUNDASAL (Fundacion Salvadorena de Desarrollo y Vivienda Minima) Foundation for Salvadorean Development and Affordable Housing GFDDR Global Facility for Disaster Reduction and Recovery ILP (Instituto Libertad y Progreso) Liberty and Progress Institute ISTA (Instituto Salvadoreño de Transformación Agraria) Agrarian Transformation Institute of El Salvador LIDAR Light Detection And Ranging MAG Ministry of Agriculture MARN Ministry of Environment and Natural Resources OECD Organization for Economic Co-operation and Development OPAMSS Planning Office of the AMSS PDT (Planes de Desarrollo Territorial) Territorial Development Plans PLAMADUR (Plan Maestro de Desarrollo Urbano del Área Metropolitana de San Salvador) Master Urban Development Plan PNODT (Plan Nacional de Ordenamiento y Desarrollo Territorial) National Plan for Territorial Planning and Development PTT (Programa de Transferencia de Tierras) Land Transfer Program REDD+ Reducing Emissions from Deforestation and Forest Degradation SSDT Vice-Secretariat for Territorial Planning and Decentralization UNDP United Nations Development Program USAID United States Agency for International Development 25 El Salvador Country Land Assessment World Bank Chapter 1 Overview 1.1 The Importance of Land in the Development of El Salvador: Historical Context Land is a scarce resource in El Salvador. With a population of 5.7 million (2007) and a land area of just 8,124 square miles (331 inhabitants per square km), El Salvador is the second most densely populated country in the western hemisphere after Haiti. A combination of a stagnant agricultural sector, large inflows of remittances from abroad, and rapid urban expansion has resulted in considerable pressure for land use conversion and a high opportunity cost for agricultural land. At present, about 61% of the population lives in urban areas and it is growing at an annual rate of 1%. Access to land and struggles over alternative uses of land have been at the center of the country’s social, economic, and political history. Inequitable land distribution has characterized El Salvador’s economy since the 1800s, and has historically been at the root of conflict and civil war in the country. Modest attempts at land reform were carried out in the 1930s, 1960s, and 1970s. However, it was not until the 1980s that the Government launched a comprehensive land reform effort that was continued throughout the 1990s as part of the Peace Accords. This process of agrarian reform was one of the most extensive in Latin America. Between 1981 and 1999, more than 400,000 hectares of land were distributed, equivalent to about one-third of the agricultural land in the country. Until recently, the land tenure structure of El Salvador has been linked to the dynamics of export markets for agro-industrial commodities. The rise of international prices for coffee was a crucial factor in the abolishment of communal lands and the concentration of land in large estates in the late 19th and early 20th century. The world depression of the 1930s hit peasant families hard in El Salvador as wages were reduced and employment levels were cut back. Poverty levels increased significantly during this period and popular unrest mounted. In subsequent decades, land concentration occurred in the coastal lowlands with the rise of prices in sugarcane and cotton in the 1940s and 1950s and cattle grazing in the 1960s and 1970s. As these large estates developed, less land became available for small farmers. Peasant families diversified their livelihood strategies by accessing land through rental and sharecropping arrangements and working as day laborers in the large estates. Other families migrated to frontier regions of the country looking to clear new land for cultivation, without holding any legal right to it, or moved to urban areas. 26 El Salvador Country Land Assessment World Bank 1.2 Objectives, Key Questions, Methodology, and Structure of this Report Objective The main objective of this Study is to assess the alignment of land use, land tenure and land market outcomes in El Salvador with public policy aspirations for efficient, inclusive and environmentally sustainable development in both urban and rural spaces. In doing so the Study indirectly gauges the effectiveness of land sector institutions in facilitating such developmental outcomes in agricultural production, urbanization, and forest management. Key Questions Among the key questions with which the Government is grappling and which this Study attempts to answer are: To what extent have public policy interventions including the land tenure changes introduced through the last iterations of the Agrarian Reform been effective in reducing rural inequality? What has happened to the land assets entrusted to Agrarian Reform Cooperatives under the last iterations of the Agrarian Reform? How rapidly are Salvadorian cities growing and how effectively has the supply of urban land kept pace with the growing demand associated with urbanization? To what extent have land use changes associated with urbanization and agricultural expansion been happening un-sustainably and in ways that are increasing natural disaster vulnerability? How extensive has been deforestation in the last decade? Given the competition for use of land, the Study inevitably identifies some areas in which public policy goals are clashing or are being ineffective in addressing developmental challenges. The scope of this particular Study does not address several other related policy questions which due to their importance and complexity require devoted studies of their own. For example this Study although interested in land in the rural sector is not an Agricultural Sector Assessment diagnosing the causes of agricultural sector decline and proposing strategies to reinvigorate the sector. Such an undertaking would require consideration of many factors beyond land including terms of trade, subsidy structures, labor market dynamics, technology and innovation and an assessment of extension services among other things. Neither is this Study an analysis of urban land supply for competitiveness which would require consideration of firm behavior and constraints as well as a variety of regulations that affect the efficiency and viability of doing business in a specific location. Finally, it is important to note that while this Study does 27 El Salvador Country Land Assessment World Bank assess deforestation rates it does not specifically investigate the causes of deforestation and the potential incentives that might remedy it. Methodology and Data Sources The Study uses empirical assessments of recent land use and market dynamism as a foundation for evaluating the effectiveness of land sector institutions and for generating policy options. Empirical assessments utilize a combination of temporal analysis of land use changes through change detection analysis of spatial images, inter-census comparisons of land tenure and production data, land market analyses, and cross-references with poverty data. Implicit in the analysis of land use, tenure and land market outcomes in relation to public institutions are conceptual facets of how land should ideally feature in the development agenda. These include: (i) a land tenure structure that promotes efficiency and equity in land use; (ii) a property rights regime that stimulates investments, reduces conflicts, and minimizes transaction costs; and (iii) an institutional framework (including legal provisions) that allows for the transfer of land to its most productive use; (iv) a regulatory framework that keeps barriers to access low while effectively managing tradeoffs Data Sources The following primary datasets were created and used in the Study:  A database of all urban and rural registered real property transactions (sales, mortgages, bequests etc.) in three of the 14 Departments of San Salvador for 2000, 2005 and 2010  A database of all Applications for Development Permission for Land Use Change in the metropolitan area of San Salvador 2008-1010  A database of all formal auctions of agrarian reform cooperative lands between 1997 and 2006  A baseline ecosystem classification map for all of El Salvador for the year 1998. Many secondary datasets were also utilized sometimes as attribute layers for the spatial analyses reported on the Study. These include:  A database of precarious settlements in El Salvador produced by the Government of El Salvador in collaboration with the United Nations Development Program (UNDP) and the Latin American Faculty of Social Sciences (FLACSO).  A Digital Elevation Model (DEM) for El Salvador  The Multiple Purpose Household Surveys for 2005 and 2009  Agricultural Censuses for 1971 and 2008 28 El Salvador Country Land Assessment World Bank  Urban Growth Management Initiative Phase II data collected from a cross-section of regional cities  Land Use Zoning Plan for the Metropolitan Area of San Salvador  Soil Quality Classification for all Land in El Salvador  A 2010 detailed ecosystem classification map for all of El Salvador  Coffee sector productivity data received from the Association of Private Producers  Land market prices in about 100 locations compiles by the Salvadorian Real Estate Association  Administrative boundaries of all municipalities  World Bank Development Indicators The Study also relied on key informant interviews from the public, private and civil society sectors including academia. Several workshops in June 2011, January and June 2012 held in San Salvador received considerable feedback on approach and findings that have been incorporated in this Report. Due to limitations in coverage of available data and resource constraints of this Study, the findings are neither comprehensive nor necessarily reflective of the totality of land use, land tenure and land market dynamism across the entire country. However, for the most part the Study maximizes the use of existing data and selectivity in the analysis is usually informed by strategic relevance. Structure of the Report This Report is structured into four Sections in addition to this Overview Chapter, the remainder of which is devoted to describing key institutional and policy variables related to land and tenure security in El Salvador. In Section I, Chapters 2 and 3 focus on land issues in the rural economy including rural inequality and poverty and the performance of the agrarian reform cooperatives. Section II first examines the urban questions of spatial expansion of cities including aspects of sustainability and natural disaster vulnerability in Chapter 4 and the inclusiveness of this urban expansion in Chapter 5. In Section III, Chapters 6 looks at the scale and distribution of deforestation trends over the past decade. Finally, Section IV takes a multi- sectoral view. Chapter 7 looks at recent trends in urban and rural land markets in three Departments where large datasets were available. Chapter 8 looks at the empirical findings of the earlier chapters from an integrated perspective of land use and public policy. And finally, Chapter 9 presents a brief summary of conclusions and policy options for addressing ongoing challenges. It includes a matrix of policy options for the consideration of the Government of El Salvador and civil society organizations in the country. 29 El Salvador Country Land Assessment World Bank 1.3 Agrarian Reform Attempts at agrarian reform in the 1960s and 1970s failed to reduce land inequalities. The landless population increased from 12 to 40 percent in these two decades. About 50 percent of the rural population owned less than two hectares of land, eight percent owned between two and 10 hectares, and the remaining two percent of the population owned large estates for the production of coffee, sugarcane, cotton, cattle grazing, and other agro-industrial activities. Facing increasing deprivation in rural areas, peasant organizations increased pressure for agrarian reform in the mid-1970s. They were supported by trade unions, the urban middle class and the Catholic Church. The Government responded by creating the Salvadoran Institute for Agrarian Reform (ISTA) in 1975, replacing the ineffective Institute of Rural Colonization. However, these efforts faced strong opposition from the landed elites. Strong support for agrarian reform materialized when a civil-military junta took power in 1979. But, by this time, the civil conflict had started and the 1980 Agrarian Reform was launched as a strategy to neutralize the insurgency. The Agrarian Reform of the 1980s went through three distinct stages. Phase I involved the expropriation of large estates to redistribute the land to agricultural workers in the form of cooperatives. The “Basic Law of Agrarian Reform”(Decree 153) and the “Law for the Execution of the First Phase of the Reform” (Decree 154) ordered the expropriation of all farms larger than 500 hectares (landowners were allowed to keep 100 to 150 hectares of land, depending on land quality). The beneficiaries were organized into cooperatives under collective land ownership. Cooperatives contracted a debt with the Government in exchange for the land and received credit and technical assistance for production. Most of the land expropriated in this first phase of the reform was pasture land or land left unused, representing about 15 percent of total farmland. Phase I was quickly implemented, it created 327 cooperatives covering approximately 232,674 hectares of expropriated lands for the benefit of 36,697 families (see Table 1.1). Phase II of the Reform aimed at expropriating farms between 100 and 500 hectares (representing approximately 23 percent of total farmland). This process was to affect the redistribution of some of the best agricultural land in the country suitable for export commodities such as coffee, sugar cane, and cotton (representing nearly 75 percent of the country’s export crops). However, this phase of the reform was never implemented due to the opposition of the affected landowners. The 1983 Constitution nullified Phase II of the Agrarian Reform by establishing a land holding ceiling of 245 hectares. Phase III was mandated by Decree 207 and became known as the “Land to the Tiller” program. It was managed by a special agency, the National Financier of Agricultural Lands (FINATA, now closed). This third phase of the Reform allowed tenants who farmed up to 7 hectares to claim legal title to rented lands. However, beneficiaries of this phase received plots of land which on average were less than two hectares. This phase distributed approximately 62,942 hectares of land to 85,227 tenant and sharecropper families. Decree 839 issued in 1987 30 El Salvador Country Land Assessment World Bank added a component of voluntary transfers of land. Through this modality, ISTA distributed an additional 10,152 hectares of land to 6,041 families. Table 1.1 Land Redistributed and Beneficiaries of the 1980s Agrarian Reform Beneficiaries & Phase I Phase III Voluntary Totals Hectares Transfers (Decrees 154 & (Decree 207) (Decree 839) 842) Benef. Families 36,697 42,489 6,041 85,227 Total benef. 194,494 259,183 36,850 490,527 Hectares 232,674 62,942 10,152 305,768 Hectares/family 5.86 1.63 1.81 3.47 Source: Adapted from the original table in Thiesenhusen (1995, p. 154). Land area was updated with information provided by ISTA in February 2011. Land issues figured prominently in the 1992 Peace Accords. After intense negotiations, the Government and the Frente Farabundo Martí para la Liberación Nacional (FMLN), the rebel group which became a political party, agreed to implement a Land Transfer Program (PTT) through a willing-seller/willing-buyer model for the benefit of approximately 47,500 ex- combatants, renters, and tenants. Beneficiaries were given a 30-year loan, at 6 percent annual interest rate, and a 4-year grace period. Beneficiaries were given the option to apply as individuals or as associations. In addition, organizations of landless peasants also benefited as part of a 1991 agreement. After seven years, the PTT distributed 3,305 properties to 36,100 beneficiaries covering an area of approximately 100,000 hectares. The Land Bank was closed in 1998 and the backlog of properties was transferred to ISTA. Overall, Cooperatives under the Agrarian Reform had difficulty becoming successful enterprises. Cooperatives were established to produce the same agricultural products grown prior to expropriation, which was mostly coffee and sugar cane. These cooperatives, however, had to rely on government subsidies for credit and technical assistance. Since the start, cooperatives had managerial constraints despite technical assistance programs. When the government phased out government subsidies in the early 1990s, the conditions for the cooperatives worsen. Most of them had trouble repaying their loans. In 1996, agrarian reform beneficiaries staged major protests that culminated in the cancellation of 70 percent of their debts. Individual Agrarian Reform beneficiaries also struggled in increasing their agricultural incomes and paying off the debt. The land plots transferred under PTT through a Land Bank were on average of lower agricultural value than those transferred during the 1980s programs, which undermined the agricultural potential of these new farms. Subsequently, USAID provided additional funds to cancel 100 percent of the debt of PTT beneficiaries. 31 El Salvador Country Land Assessment World Bank Gender and Land Property Rights Phases I and III of the Agrarian Reform did little to expand access to land for women. Only 11.7% and 10.5% of beneficiaries were women under Phases I and III, respectively. On the contrary, women experienced a net loss under Phase III as they accounted for 35.9% of landowners affected by expropriation. Phase I largely overlooked women because it redistributed land to permanent agricultural workers, while women tended to participate more in seasonal agriculture. The PTT after the Peace Accords prioritized land transfers to female ex-combatants. Among FMLN beneficiaries, the share of women who received land (29.1%) was almost equivalent to their share of the total population (26.2%). Overall, women represented 33.4% of PTT beneficiaries. Between 2001 and 2011, women received 24% of 33,800 titles legalized by ISTA. Currently, women represent about 19% of members in active cooperatives. 1.4 Legal and Institutional Framework for Land Issues in El Salvador The legal framework related to land issues in El Salvador is largely contained in eleven key instruments. These are: (i) The Constitution, (ii) the Civil Code, (iii) the Urban and Construction Law, (iv) the Development and Territorial Planning Law for the Metropolitan Area of San Salvador, (v) the Environment Law, (vi) the National Protected Areas Law, (vii) Cultural Patrimony Law, (viii) the Restructuring of the Real Estate and Mortgages Registry Law, (ix) the Cadastre Law, (x) the Urban Titling Law, and (xi) the Agrarian Transformation Institute Law (see Table 1.2). In addition, there are a number of other institutional decrees and regulations which have a direct impact on land tenure rights and uses. A key feature of the institutional framework for urban land is the Urban Development and Construction Law of 1951 which aims at regulating urban land markets and parcel subdivisions. The Law also established the Vice-ministry of Housing and Urban Development and municipalities’ competencies for urban planning, requirements for obtaining construction authorizations and appeal mechanisms. It further established the criteria for defining urban land, land subject to urbanization, and rural land in absence of a local plan serving such purpose. A Territorial Planning and Development Law (decree 644 of 2011) was approved in March 2011, but this law will only become effective in August 2012. The law aims to regulate territorial planning and development by establishing principles and institutions, at the national and local levels. It creates a National Council for Territorial Planning and Development, composed of six central government officials and three mayors. It also establishes Department level Councils and it mandates that at the local level, the law shall be implemented by municipal councils. The law also establishes the instruments for its implementation. Nationally, it mandates (i) a National Plan, (ii) regional strategies and plans, and (iii) special territorial plans. 32 El Salvador Country Land Assessment World Bank Locally, it mandates (i) municipal or micro-regional plans, (ii) urban and rural development plans, and (iii) partial plans. Table 1.2: Summary of Legislation related to Land Issues Legal Instrument Key Provisions Constitution (1983)  Recognizes and guarantees private property land rights with “social function” (Art. 103)  In rural areas, private property rights subject to a 245 ha. ceiling, except for cooperatives or peasant communities (Art. 105)  Ownership of rural lands by foreign nationals is only allowed in reciprocity (i.e., if Salvadorans are allowed ownership in the other country), except for land for industrial use (Art. 109)  Rural lands of agricultural potential which are not indispensable for State use must be transferred (with appropriate payment) to agrarian reform beneficiaries (Art. 104)  All other transfers of ownership or use rights from State lands must be approved by the Legislature (Art. 233) Civil Code (1859, as  Recognizes rights to possess, enjoy and dispose of property (Art.568) periodically amended)  Regulates the who (Art. 1599) and how of real estate sales (Art.1597, 1605)  Civil Code establishes the class of instruments that are registered (Art. 676) and in Article 686 and 687 institutes the titles that must register with the Registry. For transactions to be valid vis-à-vis third parties, they must be registered in CNR (Art. 667), similarly with mortgages (Art. 668) Urban and Construction  Regulates preparation and implementation of Urban and Rural Development Plans Law (1951, regulated  Designates Vice-ministry of Housing and Urban Development as responsible for only in 1973, as issuing national housing and urban development policy, and regulating plot amended in 1991) subdivisions Development and  Covers 14 municipalities in the AMSS, but can be expanded (Art. 4) Planning of  Designates the Planning Office of the AMSS (OPAMSS) as responsible for the Metropolitan Area of supervision, control, and approval of all urban development and construction in the San Salvador (AMSS) jurisdiction (Art. 45) Law (1993)  Mandates issuance of Master Urban Development Plan (PLAMADUR), approved in 1997 Environment Law  Key objective is to protect, conserve, and recover environment; sustainable use of (1998) natural resources; regulate environmental issues on public and private lands; and ensure compliance with international agreements (Art. 1)  Defines which plans, projects or activities require environmental impact assessment (Art. 21); in particular in section “L” related to urban projects, construction, subdivision of lots, etc. National Protected  Regulates de declaration of “natural protected area” over lands Areas Law (2005)  The National System of Protected Areas is composed of public (national government, municipalities, or autonomous entities) and private lands (Art. 9)  Protected areas on private lands to be managed by owners, maintaining their property rights (Art. 11), but these areas provide a public purpose (Art. 30), consistent with the constitutional definition of social purpose of private lands  Declared Protected Areas must be registered in CNR (Art. 12) Cultural Patrimony Law  Regulates recovery, research, conservation, protection, promotion, development, and (1993) valorization of cultural assets (Art.1)  Cultural assets can be public or private (Art 10), with maintenance of ownership rights.  Transfer of ownership or use rights must be according to regulation 33 El Salvador Country Land Assessment World Bank  Establishes restrictions on private rights, such as inability to oppose the recognition, research or recovery of cultural assets (Art. 26)  Cultural assets are exempt from Asset Taxes (Art. 53) Restructuring of Real  Mandates the registration of both the legal (Registry) and geographical (Cadastre) Estate and Mortgages aspects of land parcels, rural and urban, under a folio real system for registration of Registry Law (1986, as documents in the Registry of Real Property and Mortgages (Art. 5), related to Article amended) 48 of the Regulations of the Act.  Mandates registration of adjudication, transfers, cancelations, or modifications to real estate rights (Art. 61 of the Regulations), that is: o Ownership titles o Use, inheritance, mortgages, and other changes or transfers o Contracts (e.g., rentals) o Encumbrances and liens  Information contained in the registry is public  Act of registration is declaratory in nature ( declarativo), not adjudicatory (constitutivo), that is, a legitimate can exist even if it is not registered Cadastre Law (1984)  Mandates the accurate location of land parcels, its boundaries and size  When there is a difference between land titles that protect property and the physical results of the cadastre, both in terms of location and with respect to the area or linear measures, the cadastral record shall take precedence, except if there is a juidicial resolution signed to the contrary. Urban Titling Law  Allows untitled claimants (“owners”) of land to obtain a municipal (i.e., urban) title (1900, as repeatedly from the municipal government (Art. 9) amended) Agrarian  Designates ISTA as the agency responsible for implementing the national agrarian Transformation Institute transformation policy issued by the National Commission for Agrarian of El Salvador (ISTA) Transformation (Art. 2) Law (1975, as  In charge of transferring lands, and providing some complementary services, to amended) agrarian reform beneficiaries Territorial Management  Establishes a framework for implementing territorial management and development and Development Law that includes instances at the national, departmental, and municipal level. (2011)  Introduces a new set of instruments to guide territorial management and development at national, departmental, and municipal level. Municipalities have the option of forming a micro-region for planning purposes.  Creates systems to share territorial information, evaluate the impact of territorial- based initiatives, and foster citizen participation and consultation. Land Subdivision  Regulate the possession, commercialization, and transfer of land plots in Special Law (2012) subdivisions.  Establishes a transitory regime for the regularization of informal subdivisions. The Salvadoran legal framework provides for adequate recognition and protection of private and public lands. In El Salvador, virtually all land is in the private domain, with the public domain limited to designated Protected Areas and those used by public entities. As there are no legally recognized Indigenous Peoples in the country, there is no special tenure regime for this. Collective ownership is recognized under the Civil Code in the form of cooperatives and other joint ownership arrangements, but there are no communal lands or those under customary tenure. Private rural lands have a size ceiling of 245 hectares, except for those belonging to cooperatives. Titling of rural lands is the responsibility of the Agrarian Transformation Institute of El Salvador (ISTA), while urban lands are titled by municipalities. There is no real estate or land tax in El Salvador. 34 El Salvador Country Land Assessment World Bank Since the early 1990s, El Salvador has made considerable progress in updating its legal and institutional framework for land administration. Over the past two decades, the country has modernized many of its key land administration agencies (see Table 1.3). With the creation of the Centro Nacional de Registros (CNR) in 1994, El Salvador was a pioneer in Latin America in merging the legal registry with the geographical description of land parcels (urban and rural) under a single entity, known as folio real.. Today, CNR is one of the most sophisticated registries in the region, is entirely self-financed (in fact, it typically transfers surpluses to the Ministry of Finance), and with significant approval from the public for the efficiency and transparency of its services. Table 1.3: Key Land Administration Entities Entity Core Functions Centro Nacional de  Created in 1994, as a decentralized entity under the Ministry of Justice, Registros (CNR) headquartered in San Salvador and with 10 regional offices. Currently its relationship with the executive branch is through the Ministry of Economy.  In 1995, CNR absorbs the functions of both the General Directorate of Registries and the National Geographic Institute, and registers them under a folio real system (Art. 5 of 1986 Registry Law) Municipalities  Primary political-administrative unit of the country, with own territorial jurisdiction, and government autonomy  Issue urban land titles (Art. 9 of 1900 Urban Titling Law)  Approve urban development plans and land plot subdivisions, in accordance with 1951 Urban and Construction Law OPAMSS  Created in 1993  Responsible for urban development (e.g. building permits, subdivision of plots, etc.) of the area called "Gran San Salvador" which includes several municipalities. Ministry of  Responsible for regulating environmental matters Environment and  In particular, responsible for approving environmental impact assessments related to Natural Resources urban, construction, or land parcel subdivisions (Art. 21 of 1998 Environment Law) (MARN) Liberty and Progress  Created in 1991, as a decentralized entity under the Presidency, responsible for Institute (ILP) providing technical assistance for the effective delivery of public services and civic participation in government decisions  To promote, facilitate, and secure resources for land titling and registration Agrarian  Created in 1975, as an autonomous entity responsible for implementing the national Transformation Institute agrarian transformation policy of El Salvador (ISTA)  In charge of transferring lands, and providing some complementary services, to agrarian reform beneficiaries More than two-thirds of the country’s land parcels have been systematically surveyed. Between 1996 and 2005, CNR implemented a land administration program (with World Bank financing) covering six departments (Ahuachapan, La Libertad, La Paz, Santa Ana, San Salvador, and Sonsonate), equivalent to 60 percent of the country’s land parcels. A second phase of this program (with financing from the Bank for Central American Integration, BCIE) is currently under implementation in the departments of Chalatenango, Cuscatlan, and San Vicente. 35 El Salvador Country Land Assessment World Bank 1.5 Property Rights El Salvador enjoys a top-third ranking of 54 out of 183 nations in the 2012 ranking of the World Bank Group’s Doing Business Indicator for Registering Property. The number of procedures (5), time taken (31 days), and the average cost as a percentage of the value of the property (3.7%) are all as good as or better than OECD norms and the latter two are practically twice as good as the regional average. An alternative measure of the effectiveness of property rights institutions in El Salvador is the Heritage Foundation international ranking (see Figure 1.1). El Salvador’s performance on this indicator has been reasonable and steady except for 2005 when it was ranked as one of the best countries. 36 El Salvador Country Land Assessment World Bank Figure 1.1: Property Rights Ranking of Countries in Latin America and the Caribbean Property Rights Indicator (1995-2010) Venezuela Uruguay Trinidad and Tobago The Bahamas Suriname Saint Vincent and the Grenadines Saint Lucia Peru Paraguay Panama Nicaragua Jamaica Honduras 2010 Haiti Guyana 2005 Guatemala 2000 El Salvador 1995 Ecuador Dominican Republic Dominica Cuba Costa Rica Colombia Chile Brazil Bolivia Belize Barbados Argentina 0 10 20 30 40 50 60 70 80 90 100 Source: Compiled from data from the Heritage Foundation Website. Note: The Property Rights component is an assessment of the ability of individuals to accumulate private property, secured by clear laws that are fully enforced by the state. It measures the degree to which a country’s laws protect private property rights and the degree to which its government enforces those laws. It also assesses the likelihood that private property will be expropriated and analyzes the independence of the judiciary, the existence of corruption within the judiciary, and the ability of individuals and businesses to enforce contracts. The more certain the legal protection of property, the higher a country’s score; similarly, the greater the chances of government expropriation of property, the lower a country’s score. Countries that fall between two categories ma y receive an intermediate score. 1.6 Land Tenure Informality Challenges Despite the relative strength of some aspects of the institutional framework for land governance, significant challenges remain related to land tenure informality. As in most of 37 El Salvador Country Land Assessment World Bank Latin America, informality in access to land, use, and market transactions is quite common in El Salvador. Although not as severe as during the civil war in the 1980s, certain forms of land tenure insecurity continue in rural areas. Since 2000, ISTA has registered land invasions on 53 rural properties. There are several land regularization initiatives to address the issues of informality and land invasions in rural areas. Table 1.4 summarizes the main tenure informality issues in rural areas. Table 1.4: Main forms of tenure informality in rural areas Type of Informality Issue Occupied state lands  There are no clear legal mechanisms to address occupations in protected areas and mangroves. Pending requests to title parcels  Occupants are unable to regularize their properties because the redistributed as part of Phase III original beneficiary is no longer in possession or the title is still registered in the name of the former (expropriated) owner (legally these lots are vacant). There are no legal mechanisms to transfer ownership to current tenants. Properties in joint ownership  The process to obtain an individual title to a portion of land held in (proindiviso) which have been proindiviso is still complex and onerous. irregularly subdivided De facto subdivisions and sales of  Cooperatives lands can be transferred through public auction, prior cooperatives’ lands or illegal authorization of the State, which is a long and cumbersome process. constructions Pending requests to parcel out  Huge backlog of requests for individual titling (following 1990s legal cooperatives’ land to individual reforms) of lands under cooperative ownership. Beneficiaries still lack members a title to their portion of land. Incomplete assignment of parcels in  Some beneficiaries have not yet received land, and there is incomplete the Rural Solidarity and Landless information to finalize the distribution. Peasant Program Beneficiaries cannot afford notary’s  Costs for the elaboration of deeds and other documents are around fees and thus do not complete the US$ 80. registration process. Despite the long existence of the Urban Development and Construction Law, much informal urban development, subdivisions and sales have been taking place giving rise to at least three regularization programs for urban lands. There are programs implemented by ILP, those related to housing projects, and those directly carried out by private agents. Recently, a law of land subdivision was approved in February 2012 to regulate this market and establish a transitory framework to regularize the approximately 300,000 informal land plot subdivisions in urban areas. Although progress has been made in recent years, there are still some informal practices which yield tenure informality in urban areas. The most significant bottlenecks in regularizing urban parcels include mortgages which have not been cancelled in the Registry, unregistered subdivisions or amendments to existing parcels, and inaccurate liens on properties (see Table 1.5). 38 El Salvador Country Land Assessment World Bank Table 1.5: Main Forms of Tenure Informality in Urban Areas Type of informality Issue Marginal communities  Invasions of public or private lands, some with significant housing (tugurios) overcrowding. Illegal subdivisions of  The land is usually the property of subdivision companies rural lands for urban (empresas lotificadoras) or ex farmers who are members of such use (colonias piratas) companies. There are no legal mechanisms for occupants to regularize their lot and register it. Construction of  Buildings lacking construction permit in areas developed buildings or additions according to urban norms. without permit Abandoned houses  The majority of houses built by the Social Housing Fund (FSV). illegally occupied Lack of registration of  Transfers have not been registered or involve parcels in co- subsequent transfers ownership. 1.7 Institutions and Outcomes Having focused on the design of the institutional framework for land governance and some historical outcomes in this opening Chapter, the remainder of this Report provides insights from new empirical analyses aimed at understanding these and other outcomes in the rural economy over the last few decades as well as outcomes in the last decade with respect to urbanization and also deforestation. The findings of these empirical analyses can provide useful feedback for the continuous process of improving institutional design including governing policies and implementation arrangements so that they better respond to the realities of the Salvadorian society. 39 El Salvador Country Land Assessment World Bank SECTION I: LAND IN THE RURAL ECONOMY 40 El Salvador Country Land Assessment World Bank Chapter 2 Rural Land Tenure and Poverty Reduction 2.1 Overview of Chapter This Chapter asks the question ‘How effective have public policy interventions including the Agrarian Reform been in reducing rural inequality and tenure insecurity?’ It begins with a brief look at the changing structure of the Salvadorian economy followed by analyses of equity indicators including the poor and non-poor’s access to land. Security of Tenure and Property Rights are examined next including the pending regularization of claims for many agrarian reform beneficiaries. The Chapter concludes with a brief discussion. 2.2. The Changing Structure of the Economy The process of economic development is one of continuous redefinition of the roles of agriculture, manufacturing, and services. The global tendency is that as countries develop the share of agriculture to GDP declines and so does its contribution to economic growth. In El Salvador, agriculture value added as a percentage of GDP fell from 17.4 percent to 10.5 percent between 1990 and 2000, but then it rose again to 12.6 percent in 2010 (WDI 2012). This trend shows that while agriculture has declined, the sector’s contribution actually increased in the 2000s. In fact, compared to the average for the Latin American Region, agriculture still plays an important role in Central American economies (see Figure 2.1). However, agriculture continues to face competition in El Salvador. The manufacturing and service sectors have grown at a much faster rate, and the country is becoming increasingly urbanized. In the early 1990s, the share of the population living in urban areas surpassed the share living in rural areas. In 2010, 61 percent of the population resided in urban areas while 39 percent did in rural areas (WDI 2012). These changes are creating new demands on land in rural areas. There are, however significant discrepancies in official accounting of the extent of land held under agriculture (see Annex 1). Land policies can play an important role in this evolving context to ensure that the objectives of poverty reduction and economic growth are accomplished. It is estimated that about 22 percent of the economically active population works in agriculture (FAO 2012). Most of these people are expected to be among the 39 percent who still live in rural areas, and they appear to be affected by high levels of poverty. The poverty rate in rural areas is 46.5 percent compared to 33.3 percent in urban areas (WDI 2012). With the declining importance of agriculture, the types of investments that are necessary to target the rural poor are not clear. Land is a key asset in rural areas and understanding its dynamics is fundamental for informing policymaking. In particular it is important to understand the distribution of land tenure and level of security of property rights. Having equitable land tenure distribution and secure property rights have been identified as factors that can contribute to poverty reduction and economic growth. 41 El Salvador Country Land Assessment World Bank Figure 2.1: Agriculture Value Added (% of GDP) 30 25 LCR 20 El Salvador % of GDP Guatemala 15 Honduras 10 Nicaragua 5 Costa Rica Panama 0 Source: WDI, 2012. 2.3 Equitable Distribution of Land Tenure and Access to Land Land distribution has become more equitable in the last four decades with Table 2.1: Comparison of landless households in rural sector, 1971 and 2008 the Land Gini falling from 0.81 to 0.75 and the percentage of landless 1971 2008 Rural Population 2,309,037 2,408,944 rural households from 41% to 18%. Average household 5 5 The number of producers distributed by Estimated Number of size of total holdings in the 1971 and 461,807 481,789 Rural Households 2008 agricultural census was used to Farms in agricultural 270,869 395,588 calculate the Gini coefficient for land census distribution.7 Overall, the Land Gini fell Estimated Number of 190,938 86,201 from 0.81 to 0.75 in this period Landless indicating an improvement in the equity % of Landless 41.35% 17.89% Households of land distribution. In comparison, the Source: Comparison of III and IV Agricultural Census. average Land Gini for the Latin American region is 0.81 and for the world is 0.65 (Deininger and Olinto, 1999). The departments of La Union, San Vicente, and Ahuachapan had the largest reductions in the Land Gini coefficients. In contrast, the department of San Miguel experienced an increase of 11% in the Land Gini, but this department had the lowest coefficient in 1971 (0.68). During these four decades, however, the composition of the population structure changed 7 The Gini coefficient measures the ratio of the area between the Lorenz curve representing land distribution inequality and the equi-distribution line to the area of maximum concentration. A value of 1 represents perfect inequity whereas a value of 0 represents perfect equity. 42 El Salvador Country Land Assessment World Bank from predominately rural to predominately urban. Considering that the demand for agricultural land could have decreased in relative terms, the number of landless rural households in 1971 and 2008 is estimated using methodology suggested in Anriquez and Bonomi (2007). Under this set of assumptions described in Table 2.1, the percentage of landless households in El Salvador decreased from 41 to 18 percent between 1971 and 2008. Table 2.2: Land Redistribution Programs by Department Department Land Redistributed Distribution of Producers and Land Gini Total % of 1971 2008 % in (ha) Dept. Producer Land Producer Land Gini s Gini s Gini - Ahuachapán 33,496 25.0% 18,812 0.84 33,663 0.69 18.0% - Cabañas 6,913 3.0% 24,438 0.87 34,798 0.76 12.2% Chalatenango 23,420 6.0% 17,796 0.88 31,957 0.82 -6.8% Cuscatlán 14,472 8.0% 21,322 0.78 20,771 0.74 -4.8% La Libertad 58,895 34.0% 20,202 0.87 39,247 0.79 -9.6% La Paz 33,443 26.0% 13,937 0.82 24,959 0.83 1.1% - La Unión 37,623 18.0% 17,950 0.74 19,754 0.58 21.7% - Morazán 7,727 3.0% 15,115 0.85 25,636 0.73 14.2% San Miguel 39,390 14.0% 15,599 0.68 18,051 0.76 11.0% San Salvador 20,886 18.0% 13,081 0.81 20,221 0.74 -8.7% - San Vicente 36,404 16.0% 19,269 0.82 35,049 0.66 19.0% Santa Ana 32,429 15.0% 29,102 0.80 34,572 0.72 -9.6% Sonsonate 49,680 39.0% 19,310 0.72 23,980 0.71 -1.8% Usulután 88,102 32.0% 24,935 0.75 32,930 0.70 -6.9% National 482,882 - 270,868 0.81 395,588 0.75 -8.3% Source: (1) For land redistribution data see source in Table 2.3; (2) Number of producers in 1971 and 2008 was obtained from the III and IV Agricultural Census respectively; (3) Land Gini coefficient are the authors' calculations. Land redistribution programs under the Agrarian Reform and the Land Transfer Program (PTT) had an important contribution in reducing inequality in land tenure. Table 2.3 shows the number of hectares redistributed by the most important programs. Overall, these programs have redistributed 482,882 hectares of land or about 31% of estimated agricultural land area in the country. In terms of size, Phase I of the Agrarian Reform was the program that redistributed the largest quantity of land. The departments of Usulutan, La Libertad, and Sonsonate account for 41% of redistributed land. In relative terms, these three departments also have the largest percentage of redistributed land in their territories (over 30%). 43 El Salvador Country Land Assessment World Bank Despite the significant efforts in land redistribution, Table 2.3: Land Redistributed by the poor continue to have less access to agricultural Program land. The poor continue to have less access to agricultural land. The Multiple Purposes Household Program Area (ha) Survey (EHPM) datasets for 2005 and 2009 are used to Traditional Sector 58,313 compare the amount of land owned by the poor Phase I 232,674 Phase III - Decree compared to the non-poor. Figure 2.2 compares the 207 62,942 access to agricultural land for the poor and non-poor in PTT 103,289 2005 and 2009. The datasets permit to distinguish Decree 839 10,152 between rural and urban areas. In 2005, the poor had on FINATA average 0.99 hectares of land – about 1.6 hectares less Lotificaciones 134 than the average non-poor household. In 2009, the size Excedentes 13,908 of an average farm for the poor decreased to 0.84 '3 de Julio' Accords 1,470 hectares, but the average farm for the non-poor became Total: 482,882 Source: (1) PTT data was obtained from FUNDE smaller as well. Proximity of farmland to the urban (2009); (2) Data for the other programs was provided by ISTA's Information Division in center is another differentiating factor between the poor February 2011. and non-poor. Farmland located close to the urban center tends to be more valuable for its proximity to markets and the potential for urbanization. The differences in farm sizes between the poor and non-poor are larger for land located in urban areas compared to rural areas. Between 2005 and 2009, however, the average farm size for the non-poor decreased from 5.17 to 2.8 hectares while the average size for the poor was maintained. One possible explanation is that urbanization absorbed some of the farmland owned by non-poor households. The non-poor may be holding on to some of the farmland near urban center, for speculation. Nonetheless, the data shows that the farms owned by the non-poor cultivated 70 percent of their land and had less than 3 percent left fallow in 2005 indicating active production. Figure 2.2: Poor and Non-Poor's Access to Agricultural Land A. Access to Land in 2005 B. Access to Land in 2009 Source: Author’s calculations using EHPM 2005 and 2009 (DIGESTYC). 44 El Salvador Country Land Assessment World Bank The high fragmentation of agricultural land is one challenge to poverty reduction in the rural sector. Table 2.4 shows the distribution of producers by size of landholding in 1971 and 2008. In 1971, producers with less than 2 hectares of land made up 71% of the distribution and had access to 10.4% of total land. In 2008, the share of producers with less than 2 hectares increased to 85% with access to 23.8% of the land. These changes again show that land distribution has become more equitable. However, these changes have also led to more fragmentation of the agricultural land. The 2008 Agricultural Census indicates that about 36% of producers in the country develop their agricultural activities in holdings of less than 0.5 hectares. This fragmentation of landholdings raises a number of concerns for rural development. At such small sizes, the economic viability of the farm may be at risk. The relationship between the size of a farm and economic viability depends on the productivity of the land, the availability of public goods, technological packages, and agro-ecological conditions among others. This is of particular concern for the poor because they are more likely to have small plots of land. Table 2.4: Agrarian Structure, 1971 and 2008 Size of 1971 2008 Holdings (ha) No. of Land Cumulativ No. of Land Cumulativ Producer Area e Producer Area e s s Up to 0.49 61,257 18,425 1.3% 143,202 43,798 4.7% 0.50 - 0.99 71,207 51,862 4.8% 126,490 88,414 14.2% 1.00 - 1.99 59,063 81,039 10.4% 67,637 88,641 23.8% 2.00 - 2.99 25,357 61,245 14.6% 20,349 48,210 29.0% 3.00 - 3.99 9,905 34,361 17.0% 7,124 24,709 31.7% 4.00 - 4.99 8,152 36,378 19.5% 5,676 25,455 34.4% 5.00 - 9.99 15,598 110,472 27.1% 10,857 76,324 42.6% 10.00 - 19.99 9,164 126,975 35.9% 6,740 93,650 52.7% 160,41 20.00 - 49.99 6,986 215,455 50.7% 5,261 3 70.0% 100,89 50.00 - 99.99 2,238 154,164 61.3% 1,501 2 80.9% 100.00 - 199.99 1,103 152,056 71.8% 531 70,577 88.5% 200.00 - 499.99 636 190,821 84.9% 166 46,920 93.6% > 500 202 218,641 100.0% 54 59,840 100.0% 1,451,89 927,84 Total 270,868 4 395,588 3 Source: Comipled from III and IV Agricultural Census 45 El Salvador Country Land Assessment World Bank 2.4 Security of Land Tenure and Property Rights Overall, land tenure security is high for the population as a whole. The 2009 EHMP dataset was used to compare land tenure arrangements for agricultural land and perceptions of security between the poor and non-poor. Figure 2.3 shows that the poor are less likely to be the owners of the agricultural land on which they produce: 23 percent of the poor were landowners compared to 37 percent of the non-poor. For the poor, land rental is the most important modality to access agricultural land. The levels of deed registration between the poor and non-poor are similar. About 71 percent of the poor report having their land titles registered in CNR compared to 79 percent of the non-poor. In terms of the perception of tenure security, both the poor and non-poor believe that their property rights are secure (97 percent of the population in both groups).8 Figure 2.3: Access to Land and Tenure Security Source: Authors’ calculations using 2009 EHPM (DIGESTYC). The tenure security of agricultural lands is more at risk in areas contiguous to the San Salvador Metropolitan Area. Figure 2.4 shows the percentage of agricultural parcels that are registered in CNR and the percentage of the population that feels that they have secure land rights. The poor in the San Salvador Metropolitan Area have the lowest level of property registration for agricultural land (57 percent) in the sample. The widest gap in perception of land tenure security between the poor and non-poor is also found in the San Salvador Metropolitan Area. In addition, the Central I Region, which includes departments contiguous to the San Salvador Metropolitan Area (La Libertad, San Salvador, and La Paz) has the lowest level of perception of land tenure security for the poor and non-poor. This evidence suggests that agricultural land contiguous with the San Salvador Metropolitan Area is most at risk of weak property rights. The poor are especially at a disadvantage in this situation. 8 The EHPM dataset interprets land tenure security from the point of view of whether there is a risk that somebody else would have a claim on the property. Questions on land tenure security were not included in the 2005 survey. 46 El Salvador Country Land Assessment World Bank Figure 2.4: Land Tenure Security by Region Source: Authors’ calculations using 2005 and 2000 EHPM (DIGESTYC). The formalization of property rights of the land redistributed under the Agrarian Reform is still a pending issue. The agrarian reform efforts of the 1980s and 1990s redistributed large amounts of land. The formalization of property rights, however, is a process that has taken longer to be resolved. There were many anomalies in the ISTA titles. The cooperatives’ titles were held until the 30-year period of co-management was over or debt was repaid (as with many other aspects of the Agrarian Reform, there were no rules or systematic application of rules). The titles were listed in the ISTA President’s special volume, instead of the notaries’ protocol as priv ate conveyances must be. For this reason, these titles were not subject to the same inspection and quality control of the Supreme Court as the notaries’ protocols. Similarly, cooperatives could get their provisional title only after their period of co- management was over with a 30-year repayment period starting then. Once the titles were granted, beneficiaries often did not know that they had to register them in the Land Registry or they did not receive support to complete the process. The distribution of land to groups of individual beneficiaries under the modality of joint titles (proindiviso) has made legalization more difficult. This modality was used to transfer land to beneficiaries under Phase III and the PTT. The transfer of a single property to a group of beneficiaries improved the rate at which land was redistributed, but delayed the formalization of individual property rights. A similar situation occurs with members of the cooperatives that want to obtain title to their individual plots. ISTA is still pending to legalize individual titles to over 170,000 landholdings. Table 2.5 shows the number of landholdings that are pending legalization to individual beneficiaries. 9 The Traditional Sector is the program that has the largest number of holdings pending legalization. This program includes the properties that were transferred by ISTA between 1975 and 1980, prior to Phase I. Decrees 747 and 719 gave cooperatives that have fully paid their debt the option of requesting titles on their individual plots. There are over 12,000 requests for individual titles. 9 The variable of property refers to original landholding that was expropriated or purchased. The values for area are not available for the Traditional Sector and Phase III. 47 El Salvador Country Land Assessment World Bank These decrees also allowed cooperatives that have not yet fully paid their debt the option of cancelling the original land adjudication from ISTA to the Cooperative and instead transfer it directly to the individual members. There are over 33,000 of these cases pending legalization. Phase III involved the transfer of land in the land-to-the-tiller program. There are still about 12,000 cases remaining for legalization from the cases transferred from FINATA to ISTA. The Rural Solidarity and Landless Peasants program started in 2001 to redistribute land to former policemen who were excluded from the original PTT. There are about 35,800 cases pending legalization for this program. ISTA also has 62 properties that it needs to transfer to the State because they involve forestry areas.10 Once these properties are transferred to the State, MARN is supposed to assume responsibility for their conservation. The Institute of Liberty and Progress (ILP) participated in the legalization of 36,272 properties under the PTT, and there are no pending adjudications under this program. Table 2.5: Landholdings pending individual titling in ISTA Area Program Properties Holdings (ha) Traditional Sector - 80,045 - Phase I - Individual Plots 87 12,241 21,034 Phase I - Individual Plots (reprocessed) 293 33,579 37,232 Phase III 149 12,153 - Rural Solidarity and Landless Peasants 154 35,850 9,603 Pending Transfer to the State - 62 12,542 Total 529 173,930 80,411 Source: (1) Information on the Traditional Sector and pending transfer to the State was provided by ISTA in March 2011; (2) The information for the other programs was provided by ISTA in February 2012. There are on-going efforts to finalize the legalization of properties, but the lack of reliable and updated property records is a main constraint. The Funes Administration (2009-2014) has set a goal of issuing over 90,000 titles to Agrarian Reform beneficiaries. Between June 2009 and March 2011, more than 13,000 titles were distributed (compared to 21,000 titles distributed between 2001 and 2008). This is a significant improvement, but the lack of reliable and updated land records in ISTA continues to be a significant constraint. This institution does not have a digital inventory to keep track of transferred properties and those pending legalization. For those cases pending legalization, many of the files are incomplete or have outdated information. Each of the programs has its particular set of difficulties: Phase I: Adjudication of individual titles to members of cooperatives is often delayed because the cadastral information possessed by ISTA and the cooperatives does not match the information registered in CNR or the field reality. The main reason for these 10 Article 115 of the ISTA Law establishes that the land use of forestry areas should be maintained. 48 El Salvador Country Land Assessment World Bank inconsistencies is that most of ISTA’s cadastral information was produced prior to the issuance of the CNR Law, which established technical specifications for land surveying. ISTA and the cooperatives have to conduct new land surveying that meets these specifications adding to the cost of legalization. Phase III: One particular challenge for the legalization of these properties is that the current occupants requesting a title are different from the original beneficiaries. Many of the original beneficiaries that received land abandoned the properties during the civil conflict. These holdings were subsequently settled by new occupants who then claim possession of the land. ISTA’s policy has been to legalize the holdings in favor of the current possessors, but they are legally unable to do so if the original beneficiary does not formally bequest his or her right to the property. Rural Solidarity and Landless Peasants: The most salient challenge for this program is actually completing the land allocation to beneficiaries. Under this program, beneficiaries received certificates valid for a parcel of land. On many occasions, these certificates either did not specify the land’s exact location or assigned a parcel located on a site far from the beneficiary’s place of residence (sometimes in another department ). These factors have prevented beneficiaries from taking possession of the land, if such land is actually available. Updating the inventory of properties and the cadastral information in ISTA’s land records is an important investment to improve land governance. The systematization of existing land records in a digital inventory would provide ISTA with a tool to better track land transactions. A digital inventory would also improve transparency and reduce transaction costs by making the information more easily available to other institutions and the general public. Furthermore, ISTA could invest in a more systematic scheme to update the cadastral information of records that do not comply with the current technical specifications instead of trying to solve each case individually. These activities do not only require substantial investments in technology and personnel, but also an appropriate level of institutional coordination with CNR. In particular, the CNR can only register documents that meet the legal requirements to maintain legal security. 2.5 General Discussion El Salvador has made substantial gains in improving the equity of land distribution and strengthening property rights, but there are still some challenges to consolidate these gains. The Government of El Salvador implemented an extensive Agrarian Reform program in the 1980s and 1990s to correct the skewed distribution in land tenure that prevailed in the country. Since the mid-1990s, the country has invested in land administration initiatives to improve cadastral and registry services and increase land tenure security. These policies have delivered results: the country has a more equitable land distribution and security of tenure is relative high across the country. Nonetheless, the country now faces the challenge of fragmentation in 49 El Salvador Country Land Assessment World Bank agricultural land and there are still some sectors that have weak property rights, most notably the properties redistributed by the Agrarian Reform. A more equitable land distribution makes the benefits of land use conversion accessible to a larger segment of the population. Over the last four decades there has been a significant improvement towards a more equitable land tenure distribution. In a context in which the role of agriculture in the economy is changing, there are new demands on land for urbanization, industry, tourism, or simply as investment opportunities. There is evidence that the largest increase in land prices occurs when land passes from agriculture to other uses. Having a more equitable distribution of agricultural land allows a larger segment of the population to share these economic gains. The current fragmentation of agricultural land indicates that rural households are depending on land in different ways than they did in the past. A World Bank study conducted in 1997 estimated that the poor would need to have farms of at least 5 hectares to bring them to up to the extreme poverty line and 12.6 hectares to surpass poverty altogether if they were completely reliant on agriculture.11 Using this benchmark even in an approximate way indicates that rural households long term economic prospects are tied to income earning opportunities related to off-farm supplementary employment, urbanization and remittance transfers12. In this context, government policies should not discourage any of these activities which may be seen as rational choices of rural households. 11 These estimations were calculated assuming an elasticity of land to income of 0.4 percent, an average farm size of 2 hectares, a poverty line of US$360 per annum, and an extreme poverty line of US$180 per annum. The elasticity was calculated from a GLS model regressing household income on land, physical capital, human capital, and labor. 12 This is not completely unlike the situation in China today whereby average farm size is 0.6 ha although there rental of farm machinery is helping to raise productivity and government is encouraging consolidation of farms - the latter being a less feasible proposition in El Salvador. 50 El Salvador Country Land Assessment World Bank Chapter 3 Land and the Agrarian Reform Cooperatives 3.1 Overview of Chapter This Chapter explores what has happened to the land assets entrusted to Agrarian Reform Cooperatives under the last iterations of the Agrarian Reform. It is divided into two parts. The first looks at the amount of land resources that they control and the productivity with which they use those resources while the second part examines their performance in land market transactions using public auctions. Assessing the current performance of the cooperatives is a relevant policy question for a land assessment considering the significant amount of land that they control. In a land scarce country such as El Salvador, the land held under this sector is an important resource for development whether it is maintained as agricultural land or it changes to another use. Considering that cooperatives originated from a policy intervention, analyzing their performance is also relevant to understand the mechanics of social engineering experiments involving land. 3.2 Context of the Agrarian Reform Cooperatives From the onset, Agrarian Reform Cooperatives faced many challenges to make their agricultural enterprises productive. Agrarian Reform Cooperatives were established with the intention of creating agricultural enterprises owned by the former waged workers in the expropriated estates. Government policy encouraged the continuation of commercial agriculture in these lands, particularly coffee and sugar cane. Cooperative members had little experience in management, and even though they received technical assistance in this area, they struggled with organizational matters. During the 1980s the government provided credit, inputs, and technical assistance. Cooperative members received wages for their labor in the collective enterprise. Government assistance faded in the early 1990s. No longer able to pay agricultural wages, the labor contribution of its members dropped. Cooperatives also struggled to access credit. Without capital, many of these cooperatives went bankrupt and were unable to meet their financial obligations. A series of reforms were introduced in the 1990s to help cooperatives become financial solvent again. These reforms included debt relief and elimination of restrictions on land sales. Land held under Cooperatives had many uses besides collective production. Since the creation of cooperatives, individual members were able to use part of the land for their own agricultural production. Cooperatives did not assign specific plots to individual members but rather they were allowed to produce on whatever land was available. This situation created a lack of incentives to make medium or long term investments on the land. In the 1990s the legal framework was changed and the cooperatives were able to allocate individual parcels. Some cooperatives also rented out land to outsiders. 51 El Salvador Country Land Assessment World Bank The amount of land held under Cooperatives has decreased overtime. A significant share of cooperative enterprises is no longer under production or has sold land to external agents. As cooperatives struggled to obtain credit, technical assistance, and inputs, many of the cooperatives stop producing. Some of them were able to rent out the land to external agents. A significant trend has been the subdivision of cooperative land into individual plots. One study has suggested that land sales are motivated to pay off the debts or acquire capital to continue investing in the share of land that they still maintain (Baumeister, 1999). The demand to convert land to non- agricultural uses, particularly urbanization, has also added pressure for land sales. 3.3 Analysis of Land Use and Agricultural Productivity This Section compares the performance of the Agrarian Reform Cooperatives with other rural enterprises in a comparable group. The analysis uses data from the 2008 Agricultural Census by landholding. Agrarian Reform Cooperatives were identified by matching MAG’s registry of active cooperatives with the name of cooperatives in the Agricultural Census. Commercial enterprises, as supposed to individual producers, were selected as the group for comparison with the Agrarian Reform Cooperatives. This group includes cooperatives, corporations, government entities, and NGOs. T-tests were performed to compare the differences in means between the two groups. Since the variables of analysis are sensible to the scale of operations, t-tests were also performed using total land size of holding as a control variable. The sample of Agrarian Reform Cooperatives considered for this analysis may not be representative of the universe of cooperatives in this sector. Table 3.1 has the distribution of Agrarian Reform Cooperatives according to three data sources. The first column indicates the cooperatives that are registered with ISTA as having received land from the Agrarian Reform. The second column indicates the cooperatives that are registered as still active in MAG as of February 2011. The third column indicates the cooperatives that are identified in the 2008 Agricultural Census. The first point to note is that only 270 of 327 cooperatives (83%) that received land from ISTA appear as still having active membership. The departments of San Miguel, San Vicente, and Usulutan lost the largest number of cooperatives. The second point is that only 127 of 270 cooperatives (47%) with active membership are identified in the 2008 Agricultural Census. 52 El Salvador Country Land Assessment World Bank There are many possible reasons why 53% of Table 3.1: Registry of Agrarian Reform the cooperative were not included in the Cooperatives by Source Department ISTA MAG Census Agricultural Census: (i) cooperatives could Registry Registry Registry have been in one of the urban segments not sampled; (ii) respondents refused to answer, Ahuachapán 29 27 9 Cabañas 2 2 0 were absent at the time of the interview, or had Chalatenango 9 7 1 incomplete questionnaires; (iii) cooperatives Cuscatlán 3 2 2 were not identified in the initial list of producers La Libertad 57 54 40 and; (iv) cooperatives were abandoned or La Paz 46 42 12 without production. It is not possible to La Unión 17 12 2 determine the weight that each of these possible Morazán 2 2 1 explanations has. Therefore, one of the San Miguel 26 14 5 constraining factors of this analysis is that the San Salvador 13 11 7 comparison between Agrarian Reform San Vicente 23 13 0 Cooperatives and Other Commercial Enterprises Santa Ana 23 19 8 is only done for the subset that appears in the Sonsonate 34 31 22 census, which may not be representative of the Usulután 43 34 18 universe of cooperatives. Total 327 270 127 The sample considered for the analysis includes 135 landholdings under Agrarian Reform Cooperatives and 1,207 landholdings under Other Commercial Enterprises. Table 3.2 shows the distribution by department of producers and holdings of these two groups. The department of La Libertad has the largest number of producers and holdings for both groups. For the other departments, there also appears to be a trend that the larger the number of Commercial Enterprises, the larger the number of Agrarian Reform Cooperatives. Table 3.2: Distribution of Agrarian Reform Cooperatives and Comparison Group by Department Department AG Cooperatives Other Commercial Producers Holdings Producers Holdings Ahuachapán 9 10 68 72 Cabañas 0 0 5 5 Chalatenango 1 1 142 147 Cuscatlán 2 2 22 22 La Libertad 40 42 176 264 La Paz 12 12 30 32 La Unión 2 2 4 4 Morazán 1 1 6 6 San Miguel 5 5 44 47 San Salvador 7 9 152 229 53 El Salvador Country Land Assessment World Bank San Vicente 0 0 9 15 Santa Ana 8 10 92 105 Sonsonate 22 23 142 191 Usulután 18 18 63 68 Total 127 135 955 1,207 Source: Compiled from IV Agricultural Census Agrarian Reform Cooperatives have more land Figure 3.1: Box Plot of Total Land available and are more likely to be landowners Source: Compiled from IV Agricultural Census Total Land (ha) than Other Commercial Enterprises. Table 3.3 2,000 shows that the average land size of a holding for the Agrarian Reform Cooperatives is 328 hectares 1,500 compared to 73 hectares for Other Commercial Land total 1,000 Enterprises. Figure 3.1 shows the distribution of the 25th, 50th, and 75th percentiles of land distribution in a 500 box plot display, along with the outside values. This figure also shows that although there are some 0 Other Commercial Agrarian Reform Commercial Enterprises with large holdings, the Agrarian Reform’s holdings tend to be larger in size than those in the comparison group. Agrarian Reform Cooperatives also produce almost exclusively on land owned by them (98%). Other Commercial Enterprises also develop most of their activities on land owned by them (83%), but they access land through other types of tenure arrangements as well. Agrarian Reform Cooperatives have less Figure 3.2: Average distribution of land by use land under cultivation than Other Distribution of Land Use in Holdings Distribution of Land Use in Holdings Other Commercial Agrarian Reform Commercial Enterprises. Figure 3.2 shows the distribution of land use in percentage 14% 3% 9% 2% terms for both groups. Agrarian Reform 8% 12% Cooperatives on average cultivate 63% of 3% 5% 2% the land surface compared to 68% among 2% 68% 5% 2% 63% Other Commercial Enterprises. If land size is controlled, the results show that Agrarian Cultivation Permanent Pasture Reform Cooperative on average cultivate Seasonal Pasture Fallow Forest Unsuitable 10% less of their holdings than the Other comparison group. This difference is Graphs by AG Cooperatives and Other Commercial Entreprises significant at the 5%-level. On the other Source: Compiled from IV Agricultural Census hand, Agrarian Reform Cooperatives have more surface area under temporary pastures than the comparison group; this difference is significant with and without controls. For Commercial Enterprises the second most important land use category is ‘other uses’ which represents 14% of their total holdings. This category includes surface areas under water bodies and installations. The difference, however, is not statistically significant. Also, there are no statistically significant 54 El Salvador Country Land Assessment World Bank differences between the two groups for permanent pastures, fallow land, forests, or unsuitable areas for cultivation. Sugarcane and Coffee are the two most Figure 3.3: Distribution of cultivated area by important crops for both Agrarian crop Reform Cooperatives and Other Distribution of Cultivated Area Distribution of Cultivated Area Other Commercial Agrarian Reform Commercial Enterprises, but coffee is cultivated by a larger number of 2%2% 3% 2% 4% 3% producers. Figure 3.3 shows that total crop area dedicated to sugarcane is 49% among 47% 49% Agrarian Reform Cooperatives and 47% 46% 42% among Other Commercial Enterprises. Coffee comes in a close second with 42% and 46% of land under cultivation Sugarcane Coffee Basic Grains Agro-Industrial respectively. However, the share of holdings Fruits and Vegetables that cultivate coffee is larger than for Graphs by AG Cooperatives and Other Commercial Entreprises sugarcane. Coffee is cultivated in 54% of Source: Compiled from IV Agricultural Census Agrarian Reform Cooperatives and 43% of Other Commercial Enterprises. On the other hand, sugarcane is only cultivated in 33% and 18% of holdings respectively. With respect to differences between the two groups, Table 3.3 shows that on average the holdings of Agrarian Reform Cooperatives dedicate less land to the cultivation of basic grains, fruits, and vegetables, but more to agro-industrial crops, compared to Other Commercial Enterprises. Table 3.3: Summary Statistics - Cultivated Land % of Cultivated AR Cooperatives Other Commercial Diff+ Diff++ Land Obs. Mean Obs. Mean Sugarcane 45 0.86 220 0.92 -0.061 -0.031 Coffee 73 0.87 515 0.94 -0.0625* -0.010 - Basic Grains 119 0.04 968 0.11 0.0710** -0.028 Agro-industrial 119 0.07 968 0.03 0.0446** 0.0478** Fruits and Vegetables 119 0.03 968 0.15 -0.127** -0.0607* Source: Author’s analysis of data from IV Agricultural Census Notes: Surface values are for areas harvested for grain and seeds; (+) Mean value of other commercial subtracted from mean value of reform cooperatives; (++) Difference in (+) controlling for size of holding; (*) Difference is significant at the 5% level; (**) Difference is significant at the 10% level. Agrarian Reform Cooperatives have higher productivity in the production of maize and sugarcane than Other Commercial Enterprises. Table 3.4 compares the average yields for basic grains (maize, sorghum, beans, and rice) and sugarcane. For basic grains, Agrarian Reform Cooperatives appear to be more productive in maize and beans and less productive in sorghum 55 El Salvador Country Land Assessment World Bank and rice, but only the difference for maize yields is statistically significant. This difference, however, becomes insignificant when land size is included. For sugarcane, Agrarian Reform Cooperatives have a higher yield (100.4 ton/ha) than Other Commercial Enterprises (94.3 ton/ha). This difference is significant at the 5%-level, but insignificant when land size is included. These results suggest that some of the productivity gains perceived in maize and sugarcane for Agrarian Reform Cooperatives may be due to economies of scale. A graphical inspection of the data in figures 3.4 and 3.5 however does not show a clear relation between land size and yields. Table 3.4: Yields in Basic Grains and Sugarcane % of Cultivated AR Cooperatives Other Commercial Diff+ Diff++ Land Obs. Mean Obs. Mean Maize Surface Area (ha) 20 12.5 208 3.0 9.441** 7.461* Production (qq) 21 2,269.8 208 260.2 2,010** 2,103** Yields (qq/ha) 16 81.5 152 54.3 27.27** 16.100 Sorghum Surface Area (ha) 21 0.7 208 0.5 0.143 -0.244 Production (qq) 21 22.9 208 24.6 -1.696 -22.850 Yields (qq/ha) 3 36.2 26 39.6 -3.338 -3.665 Beans Surface Area (ha) 20 3.0 207 0.5 2.506** 1.907* Production (qq) 21 153.3 208 20.2 133.2** 106.6* Yields (qq/ha) 6 26.8 46 21.2 5.565 5.643 Rice Surface Area (ha) 21 1.1 208 0.6 0.419 0.025 Production (qq) 21 128.6 208 108.5 20.070 -64.280 Yields (qq/ha) 2 115.1 11 124.0 -8.878 -44.840 Sugar Cane Surface Area (ha) 45 249.6 220 119.2 130.3** -84.08** Production (ton) 45 25,044.0 220 11,211.3 13,833** -5,607* Yields (ton/ha) 45 100.4 219 94.3 6.063* 6.076 Source: Author’s analysis of Agricultural Census (2008) Notes: Surface values are for areas harvested for grain and seeds; (+) Mean value of other commercial subtracted from mean value of reform cooperatives; (++) Difference in (+) controlling for size of holding; (*) Difference is significant at the 5% level; (**) Difference is significant at the 10% level. 56 El Salvador Country Land Assessment World Bank Figure 3.4: Yields in Sugarcane by Figure 3.5: Yields in Maize by Farm Farm Size Size Yields in Sugarcane Yields in Maize 200 150 150 100 ton/ha qq/ha 100 50 50 0 0 500 1000 1500 2000 0 500 1000 1500 2000 Farm Size (Ha) Farm Size (Ha) AR Cooperative Other Commercial AR Cooperative Other Commercial Agrarian Reform Cooperatives and Other Commercial Enterprises have similar access to machinery, inputs, and credit once the size of their land holdings is taken into account. Table 3.6 shows the percentages of holdings that have access to different types of machinery, inputs, and credit. Agrarian Reform Cooperatives have more access to machinery, inputs, and credit than Other Commercial Enterprises. In particular, the difference in access to tractors, harvesting equipment, fertilizer use, and use of credit is statistically significant. These differences, however, become insignificant when land size is included, which suggests that the scale of operations may be the factor correlated with higher use. Also, when land size is taken into account, the results show that Agrarian Reform Cooperatives use 11.9% less spraying equipment than the comparison group. Nonetheless, the overall comparison shows that both groups have similar access to resources once land size is taken into account. Figure 3.6: Machinery, Inputs, and Credit % of Holdings: AR Cooperatives Other Commercial Diff+ Diff++ Obs. Mean Obs. Mean Machinery Tractor 135 0.41 1207 0.26 0.151** 0.007 Irrigation 135 0.19 1207 0.17 0.016 -0.055 Spraying 135 0.61 1207 0.59 0.013 -0.119* Harvesting 135 0.21 1207 0.14 0.0715* -0.010 Inputs and Practices for Agroindustrial Crops Improved seed 135 0.05 1207 0.03 0.018 0.009 Fertilizer 135 0.33 1207 0.20 0.134** 0.001 Integrated Pest Mgmt. 135 0.10 1207 0.07 0.028 -0.037 Soil Conservation 135 0.07 1207 0.05 0.019 -0.026 Credit Applied for Credit 135 0.50 1207 0.36 0.147** 0.089 Received Credit 68 0.99 430 1.00 -0.010 -0.008 Source: Author’s analysis of data from IV Agricultural Census 57 El Salvador Country Land Assessment World Bank Notes: (+) Mean value of other commercial substracted from mean value of reform cooperatives; (++) Difference in (+) controlling for size of holding; (*) Difference is significant at the 5% level; (**) Difference is significant at the 10% level. 3.4 Land Transactions in Cooperatives of the Agrarian Reform Stringent restrictions were placed on land owned by the Agrarian Reform Cooperatives to avoid fragmentation and maintain economies of scale. When Phase I started, cooperatives were not allowed to sell, rent, or subdivide their lands. These restrictions were progressively lifted during the 1990s, when the country adopted a policy aimed at liberalizing and reactivating the reformed sector (Programa Promoción de la Reactivación Económica y Social). This change was brought about through the replacement of the cooperative production scheme with alternative models of organization where beneficiaries had the opportunity to choose the tenure form and production system that most suited their interests. The liberalization process included several steps. Decree 747 was the first step toward liberalizing the cooperative sector by giving cooperative members options to their form of association. Members of cooperatives could decide if they wanted to retain the cooperative model or individualize it. It also allowed cooperatives to parcel out lots to its members for housing and production. The decree further established rules on the rights and obligations of beneficiaries, including the right to sell, donate, or rent the land that they had obtained. However, such right was limited in that agrarian reform lands could only be sold, rented and donated to other landless beneficiaries for a total land area of up to 7 hectares per beneficiary. Finally, the decree still imposed restrictions on the subdivision of parcels. Decree 719 made further important changes towards liberalization, in addition to reaffirming the right of beneficiaries to freely choose between individual and collective land tenure regimes. Specifically, Decree 719 allowed cooperatives to (i) sell portions of land to its members for housing, and (ii) sell or rent parcels or lots to non-members, after approval by the Extraordinary General Assembly. Decree 719 also allowed cooperatives to use the proceeds from the sale to reduce arrears to the financial system and state agencies13. The selling of cooperative land to non-members requires a non-judicial public auction. The sale must be approved by two thirds of the cooperative members through an Extraordinary General Assembly meeting, which requires the participation of representatives of several public entities including the Attorney General and the Ministry of Agriculture. There are limitations for the establishment of the sale price, which cannot be lower than the base price as evaluated by a financial expert. The notice of public auction has to be approved by several entities, including the ISTA delegates, the Ministry of Agriculture, the Attorney General, and all members of the 13 In 1998, the Government condoned 85% of the debt accrued by land reform beneficiaries. 58 El Salvador Country Land Assessment World Bank cooperative’s Supervisory Board. The notice has then to be published in two of the main newspapers at different times with an interval of at least 7 days. A dataset of 78 public auctions was compiled to analyze the performance of this mechanism. The Ministry of Agriculture is required to keep records of all the public auctions. These records were digitized to examine trends in public auctions and observe how successful the cooperatives were in selling the plots they offered. There were 78 public auctions between 1997 and 2006. After this period, there were only two public auctions registered in 2011 and 2012. The department of La Paz had the largest number of public auctions, followed by the departments of La Libertad and San Miguel. In total, cooperatives offered 6,135 hectares of land out of which 5,293 hectares were sold. In 53 out of the 78 cases, cooperatives were able to sell all the land that was offered. The average size of plots successfully auctioned was 68 hectares and the average sale price per hectare was US$8,459 (in 2005 dollars). For some cases, the sale price in the auction was significantly lower than the average sale price for a rural property in the same municipality. Since land market data for the rural sector was available for the departments of Ahuachapan, Santa Ana, and Sonsonate from CNR, average land sale prices in 2005 were compared to the auction sale prices by municipality. Twenty municipalities were matched: 3 in Ahuachapan, 10 in San Ana, and 7 in Sonsonate. In 8 cases, the auction price was higher than the average sale price in the municipality. For this subgroup, on average the auction prices were 62 percent higher than the sale prices. On the other hand, in 12 cases the auction price was lower than the average sale price in the municipality. There is a large variance in the price differentials. Average sale prices range between 45 to 970 percent higher than the auction prices.14 It is important to note the existence of these extreme differences in the auction prices and average sale prices. However, it is difficult to make a generalized conclusion because sale price are sensitive to a wide range of factors, such as location and soil quality that were not available for analysis. Table 3.5: Public Auctions in Cooperatives of the Agrarian Reform Number Total Land Cooperatives that Average Total Land Department of Offered Sold all Offered Price per Sold (Ha) Auctions (Ha) Land Ha. (USD) Ahuachapán 3 244 242 2 4,996 Chalatenango 2 82 82 2 2,999 La Libertad 12 927 729 9 13,042 La Paz 14 766 726 6 7,421 La Unión 2 114 103 0 2,962 San Miguel 12 1,081 785 6 7,291 14 The most extreme differentials are in the municipality of El Congo in Santa Ana where the average land price was US$17,708 per hectare compared to auction prices of US$1,655, US$1,659, and US$4,351 per hectare. 59 El Salvador Country Land Assessment World Bank San Salvador 4 176 176 4 27,213 San Vicente 1 128 128 1 25,765 Santa Ana 10 816 643 7 5,100 Sonsonate 9 759 759 9 6,129 Usulután 9 1,041 918 7 4,915 Total 78 6,135 5,293 53 8,459 Source: Author’s analysis of dataset created from data obtained from Division of Agrarian Policy, MAG. Notes: Prices are standarized to 2005 dollars Public auctions attracted little competition: there was only one bidder in 55 out of the 78 cases. The conditions for the cooperatives were more unfavorable in the cases where only one bidder was present. In these cases, the average price per hectare at the time of sale was actually lower than the price per hectare in the reserved price. Instead, in cases where multiple bidders were present, the price per hectare during the sale was 46 percent higher than the price per hectare at the reserved price. More competition in the biddings seems to have improved the terms for the cooperatives. Figure 3.7: Reserved and Sale Prices by Number of Bidders Public Auctions in Cooperatives Reserved and Sale Prices 10,000 9779.42 9,000 Price per Hectare 8,000 7906.91 7178 7,000 6251.11 6,000 One Bidder Multiple Bidders Average Reserved Price Average Sale Price In most cases, cooperatives auctioned land that was used for the cultivation of sugarcane, basic grains, or coffee. Land with agricultural potential was sold at higher prices than land where no crops were grown. In particular, prices for land used to grow fruit trees and coffee were on average higher than for any other land use. Unfortunately, it is not possible to say from the data what was the intended purpose for the land at the time of the auction and what has happened to the land since then. 60 El Salvador Country Land Assessment World Bank Figure 3.8: Land Use of Plot Prior to Public Auction Types of Land Use Sesame 1 Fruit trees 1 Coconut 2 Henequen 3 Forest 5 No crops 6 Pasture 13 Coffee 18 Basic grains 26 Sugar cane 32 0 5 10 15 20 25 30 35 Number of Plots Figure 3.9: Sale Price by Type of Prior Land Use Public Auctions in Cooperatives Price by Prior Use no crops 2317.12 other 7157.48 basic grains 7392.42 sugar cane 8447 pasture 8600.11 forest 9877.8 coffee 11887.2 fruit trees 16353.7 0 5,000 10,000 15,000 20,000 Average Sale Price per Hectare 3.5 General Discussion Land held by cooperatives of the Agrarian Reform is an important resource for the country’s development. Phase I redistributed 232,674 hectares of land to cooperatives which are equivalent to 15 percent of estimated agricultural land in El Salvador. This is a relatively 61 El Salvador Country Land Assessment World Bank large amount of land under this type of control. The development potential of this land is also intrinsically linked to the original objectives of the Agrarian Reform. The cooperatives were created as a social experiment to pass control of the land from a few landowners to the agricultural workers in the haciendas. The decision not to subdivide the land was based on the economic rationale of maintaining economies of scale to improve productivity. Since early on, however, the cooperatives struggled to maintain productivity and many defaulted on their loans. Policy interventions in the 1990s relieved the sector’s indebtedness and lifted some of the constraints on land transactions and modes of association. One of these reforms allowed cooperatives to participate openly in land markets by offering land in public auctions. Agrarian Reform Cooperatives identified in the 2008 Agricultural Census have a performance equivalent to that of the comparison group, but it is not possible to make inferences for the rest of the cooperatives. Both groups are dedicated primarily to sugarcane and coffee production. They have similar levels of productivity and access to machinery, inputs, and credit. There are differences too. Agrarian Reform Cooperatives do not only have larger holdings but they are also more likely to own them. This group shows slightly less area under cultivation in percentage terms. This situation is an opportunity for expanding agricultural production since they are the landowners. However, the sample of Agrarian Reform Cooperatives may not be representative of the universe. As explained, only 127 cooperatives were identified in the 2008 Agricultural Census out of 270 registered as active in MAG. Despite the legal reforms of the 1990s, public auctions are not an effective mechanism for cooperatives to participate in land market transactions. Cooperatives have not reported using this mechanism since 2006. Also, most of the auctions involved a single bidder. In these cases, the cooperatives obtained uncompetitive prices. Meanwhile, cooperatives continue to use informal mechanisms to sell land to external agents. Two informal mechanisms include the adjudication of a plot to an individual member who in turn sells the land to an external agent and entering into a fictitious collateralized credit obligation where the land is foreclosed. The extent to which these informal mechanisms are used is unknown, but they must be common practice considering the low frequency of public auctions in the past years. This type of informal subdivision may be one of the main sources of new land for urbanization. These informal practices, however, complicate regularization because these lands are still classified as rural. There can be many possible reasons why over 50% of the cooperatives were not identified in the census. A common held view is that many cooperatives are not producing and the land is idle. In some cases, cooperatives could have already sold their productive land to external agents. As observed above, cooperatives have sold over 5,000 hectares in public auctions but many more hectares could have been transferred through informal mechanisms. Yet, probably the most notable trend in this land market is the large demand for subdivision. Since the legal reforms in the 1990s, thousands of cooperative members have requested individual land rights to their housing and agricultural plots. It is possible that the 2008 Agricultural Census could have 62 El Salvador Country Land Assessment World Bank accounted for these cooperative members as individual producers, and this could be one of the reasons why so many cooperatives were not identified as such. This reflects the new reality of the Agrarian Reform Cooperatives, and programs aiming to improve the performance of this sector need to account both for group and individual producers. In the case of communal production, lessons could be learned from the cooperatives that have continued to operate successfully to improve the performance of those that had stopped producing. For individual plots, cooperatives would need to evolve from communal enterprises to service providers that can offer assistance to improve the production, transformation, and marketing of products grown by their members. 63 El Salvador Country Land Assessment World Bank SECTION II: LAND FOR URBAN EXPANSION 64 El Salvador Country Land Assessment World Bank Chapter 4 Urban Expansion – Extent, Tradeoffs and Disaster-Vulnerability 4.1 Overview of Chapter This Chapter asks the question ‘how has land governance in El Salvador responded to the challenges of urban land supply in the last decade?’ It looks at how San Salvador and the two next largest urban areas, Santa Ana and San Miguel have fared especially in relation to the efficiency and sustainability of urban land supply. Among the metrics that the research generates are the scale of urban expansion (percentage change in the built up areas over a ten year period), and the extent to which this expansion has infringed topographically challenging territory (steep slopes) and former agricultural land of good soil quality. In San Salvador, the expansion’s footprint is also compared with land use planning zones in the master plans as well as with hot spots for applications for planning permission. The Chapter begins with a brief overview of urbanization trends in the region and concludes with a general discussion about the appropriateness of the institutional framework for regulating land development. 4.2 Urbanization trends In 2010, El Salvador’s population was 61% urban, up from about 38% in 1960. While this is more urbanized that many other Central American neighbors such as Honduras, Nicaragua and Guatemala, it is still way short of the Latin America and Caribbean regional average of close to 80% (see Figure 4.1). Based on the urbanization trend in El Salvador over the last 5 decades and the existing gap to the regional average, it is safe to assume that El Salvador will continue to urbanize over the next few decades. Given that research has shown that population and GDP per capita growth are the main determinants of expanding city footprints15, we can also safely predict a growing demand for land around El Salvador’s cities in the coming decades 16 The annual rate of population increase in the country’s capital region, the AMSS, during the 1990s was also more rapid than in many other cities of the region including Mexico City and Guadalajara in Mexico, Sao Paulo and Ribeirao Preto in Brazil and Valledupar in Colombia. Like most cities around the world, San Salvador became less dense over this decade as its footprint grew somewhat faster than its population. However perhaps due to the relatively scarcity of land, the annual percentage increase in built-up area per person in San Salvador was only marginal compared to many other Latin American cities during this period (see Figure 4.2). 15 See the Dynamics of Global Urban Expansion, World Bank 2005. 16 El Salvador’s GDP per capita in current US$ terms climbed steadily between 1990 and 2010 . 65 El Salvador Country Land Assessment World Bank Figure 4.1: Urbanization Trends for El Salvador and its Neighbors Source: WDI, 2012. Figure 4.2: Annual Percentage Change in Built-up Area per Person in Select Latin American Cities during the 1990s Annual % Change in Builtup Area per Person 6.00% 5.00% 4.00% 3.00% 2.00% 1.00% 0.00% -1.00% -2.00% -3.00% Source: Based on data from World Bank 2005 4.3 The Extent of Urban Expansion 2001-2010 This Section measures land use change associated with urbanization in the first decade of this century (2001-1010). Change detection analyses were done of land use in and around the country’s three largest cities/metropolitan areas: San Salvador (AMSS), Santa Ana and San Miguel based on Landsat satellite images. Higher resolution visual inspection of Google Earth 66 El Salvador Country Land Assessment World Bank images for 5-6 Areas of Interest in each city where land use change was most significant was also used. More details of the Google Earth observations in the Areas of Interest of each city can be found in Annex 2. In the first decade of this century, the AMSS ’s urban foo tprint (including the spatially contiguous city of Soyapango) grew by approximately 30% from about 115 square km to over 148 square km. Figure 4.3 shows the spatial distribution of this growth. San Salvador’s growth appeared mostly along the extreme outskirts of the city. In some cases , particularly in the north east but also in parts of the south, the growth appeared to replace forests. Most of the growth appears to have been preceded by infrastructure, at least roads. The rate of expansion was comparable to global rates in the 1990s. Figure 4.3: Urban Growth in San Salvador Metropolitan Area (AMSS) 2001-2010 Very significantly, two satellite municipalities of the AMSS, Colon and San Juan Opico both in the Department of La Libertad went from being completely undeveloped to 26 square kilometers of urban development in this same period. 67 El Salvador Country Land Assessment World Bank Figure 4.4: Location of Satellite Municipalities of Colon and San Juan Opico In the first decade of this century, Santa Ana’s17 urban footprint grew very substantially by approximately 120% from approximately 17 square km to over 37 square km. Figure 4.5 shows the spatial distribution of this growth. Most growth appeared on the fringes of the city. In one part of the south-west, significant growth has occurred at the expense of forests and sometimes agriculture. Some of the expansion into forests is non-contiguous with other built development following a more inefficient leapfrogging pattern. Other parts of the expansion appear informal. Figure 4.5: Urban Expansion of Santa Ana between 2001 and 2010 17 Santa Ana is the second largest city in El Salvador, located 64 kilometers west of San Salvador, the capital city. Santa Ana serves both as the capital of the department of Santa Ana and as the municipal seat for the surrounding municipality of the same name. 68 El Salvador Country Land Assessment World Bank In the first decade of this century, San Miguel’s18 urban footprint grew by approximately 60% from approximately 24 square km to over 38 square km. Figure 4.6 shows the spatial distribution of this growth. San Miguel has experienced growth along its outer edges. Mostly, it appears that houses have replaced agriculture rather than forests. 19 Figure 4.6: Urban Expansion of San Miguel between 2001 and 2010 4.4 New Settlement on Steep Slopes – Natural Disaster Vulnerability This Section assesses the extent to which urban spatial expansion is occurring on steep slopes and therefore on potentially vulnerable locations viz a viz natural disasters or areas with significant environmental impact. The urban growth areas are overlaid on a Digital Elevation Model of the country to facilitate the analyses. In all three cities there is a significant amount of growth that is occurring on steep slopes (depicted by red in Figures 4.7 and 4.8) 20. This is especially so in San Salvador, its satellite Colon, and Santa Ana with San Miguel being in much more low-lying terrain. In San Salvador these slopes are sometimes more than 30 degrees and in Santa Ana they are as steep as 28 degrees By comparison built slopes in Colon and San Juan Opico are never more than 9 degrees. 18 San Miguel is located 138 km east of San Salvador. It is also the capital of the department of San Miguel and a municipality. 19 One thing to keep in mind- March of 2004 is the earliest Google Earth image for this area. Since the change detection from 2001-2010 picked up significant urban growth in these 5 AOIs, which in some cases isn’t apparent by the Google Earth imagery, it’s possible that a lot of the growth occurred between 2001 and 2004. 20 The methodology does not allow an exact area or percentage of urbanized land that occurs at certain slopes to be calculated. 69 El Salvador Country Land Assessment World Bank Figure 4.7: Urban Expansion 2000-2010 by Slope in San Salvador and Satellite Municipalities Figure 4.8: Urban Expansion 2000-2010 by Slope in Santa Ana and San Miguel In Santa Ana almost 90% of the urban growth occurred at elevations above 600m, whereas for San Salvador the corresponding proportion was a little more than half. By contrast almost 90% of growth in San Miguel was at elevations under 200m. See Table 4.1. Table 4.1: Percentage of Urban Growth 2000-2010 by Elevation Elevation San Salvador Santa Ana San Miguel 31-99 0.0 0.0 42.3 100-199 0.0 0.0 47.9 200-299 0.0 0.0 6.1 300-399 0.2 0.0 3.3 400-499m 22.1 0.0 0.4 500-599m 17.0 6.7 0.0 600-699m 21.6 25.8 0.0 700-799m 15.6 62.9 0.0 800-899m 10.5 4.1 0.0 900-999m 11.2 0.2 0.0 1000- 1114m 1.7 0.4 0.0 70 El Salvador Country Land Assessment World Bank 4.5 New Settlement on viable agricultural land This Section assesses the extent to which urban spatial expansion is occurring on viable agricultural land and therefore on potentially important land for the nation’s food production strategy. To accomplish the assessment the 2009 soil classification map for El Salvador was overlaid with the areas of urban growth. Six (6) square kilometers of San Salvador’s urban growth in the last decade occurred on relatively good agricultural lands (class II). This is about one fifth (18%) of the city’s expansion. Almost 3 square kilometers (about 10%) of the expansion in the AMSS two satellite municipalities of Colon and San Juan Opico occurred on such land. It is also reported that much of the urban growth in these two municipalities displaced shade coffee. Figure 4.9: Area by Soil Type that became urban in San Salvador between 2000 and2010 Area (in sq. km) of soil types that became urban land 12 10 8 6 4 2 0 Clase I Clase II Clase IIIClase IV Clase V Clase VI Clase Clase Urbana VII VIII Figure 4.10: Area by Soil Type that became urban in the Satellite Municipalities of Colon and San Juan Opico - 2000-2010 Area (in sq km) of soil types that became urban land 16 14 12 10 8 6 4 2 0 Clase I Clase II Clase III Clase IV Clase V Clase VI Clase Clase Urbana VII VIII 71 El Salvador Country Land Assessment World Bank Almost 10 square kilometers of Santa Ana’s urban growth occurred on good agricultural land (Classes I and II). This is just short of half (48%) of all of the city’s urban expansion in the last decade. Figure 4.11: Area by Soil Type that became urban in Santa Ana between 2001 and 2010 Area (in sq. km) of soil types that became urban land 10 8 6 4 2 0 Clase I Clase II Clase III Clase IV Clase V Clase VI Clase Clase Urbana VII VIII Approximately 4.5 square kilometers of San Miguel’s urban growth occurred on good agricultural land (Classes I and II). This is one quarter of all of San Miguel’s urban expansion in the last decade. Figure 4.12: Area by Soil Type that became urban in San Miguel between 2000-2010 Area (in sq.km) of soil types that became urban land 8 6 4 2 0 Clase I Clase II Clase III Clase IV Clase V Clase VI Clase Clase Urbana VII VIII Summary Given its comparatively high ratio of population to land area, the opportunity cost of agricultural production is expected to be quite high in El Salvador. Market pressures to convert agricultural land to more productive uses are expected to be strong. The analysis of urban growth patterns in the last decade for the country’s three main cities supports this 72 El Salvador Country Land Assessment World Bank hypothesis. A significant part of the urban expansion of the last decade has occurred on land with good agricultural potential (Classes I or II soils). This proportion was greatest for Santa Ana (almost one half), then San Miguel (one quarter) and least for San Salvador (about one fifth). This trend raises questions about the effectiveness of local land use planning mechanisms in all cities but in secondary cities in particular. 4.6 Development Control in AMSS This Section examines two sets of data which give further insight into the effectiveness of Development Control mechanisms and implementation in the AMSS. The first is a spatial analysis of the degree to which urban expansion in the AMSS over the decade of 2000-2010 complied with land use planning/zoning. The second is the universe of applications for change of land use over the three year period 2008-2010. 4.6.1 Settlement compatibility with land use zoning The vast majority of urban expansion in the AMSS between 2000 and 2010 occurred in accordance with the Master Plan although urban expansion did infringe upon risk-prone areas that were zoned for maximum protection especially in the south and occurred upon land zoned for agricultural development in the north east. See Figure 4.13. The exact percentages of San Salvador’s total area in 2010 that fell into the respective zoning categories are shown in Table 4.2. It shows that almost 14% of the city’s built footprint by 2010 had occurred in areas not zoned for urban use. Of this about 4% was in the Maximum Protection Zone, and another 4% was in the agricultural development zone. 5% was on land at the periphery with no zoning. 73 El Salvador Country Land Assessment World Bank Figure 4.13: Overlay of Urban Expansion in AMSS 2000-2010 with Land Use Planning Zones Table 4.2: Percentage of Urban Development in SS 2010 by Land Use Zone Zone Area (Square km) Percentage of Total Area AU (urban potential) 123.8 83.6 CIL 0.9 0.6 DA (agricultural development) 5.6 3.8 DR1 0.4 0.3 DR2 2.7 1.8 ED 0.1 0.1 MP (maximum protection) 6.0 4.0 74 El Salvador Country Land Assessment World Bank PT 0.0 0.0 ZA 0.6 0.4 ZARG 0.1 0.1 ZP 0.6 0.4 Growth outside any zone 7.4 5.0 Total 148.0 100.0 4.6.2. Analysis of land use change applications In total the 14 municipalities comprising the AMSS received 2615 land use change applications in the 3 years spanning 2008-2010. Table 4.3 shows the distribution of those applications first by municipality and then by outcome. These changes were not necessarily across broad land use categories such as residential, commercial, industrial and agricultural. Rather more often than not the changes applied for related to an intensification of land use within the same category e.g. change from single family residential to condominium residential. By far the most dynamic area of land use change applications has been the core municipality of San Salvador, accounting for almost two thirds of all applications. The only other municipalities of note are Santa Tecla, Mejicanos and Soyapango which together accounted for another 20% of all applications. On average about one third (32%) of applications received favorable rather than conditional outcomes. This proportion was more or less the same for the municipality of San Salvador. It is interesting to observe that despite much of the growth of the AMSS over the last decade being on the periphery and therefore in municipalities other than San Salvador municipality, formal applications for land use change development permission are relatively few in the outlying areas of the metropolitan area, at least in the last three years. It may be that much of this development occurred in the earlier part of the decade (2001-2007) but it seems that this is unlikely to be the total explanation especially in municipalities with significant land use dynamism (see Figure 4.1) such as Apoya in the north and Soyapango and San Martin in the west. Rather a more feasible explanation is that a significant part of the urban expansion in these outlying municipalities may be bypassing the formal development permission process. Table 4.3: Spatial Distribution of Land Use Change Development Permission Applications in AMSS 2008-2010 Municipality Number of Applications Number Approved without cond’ ANTIGUO CUSCATLAN 117 58 APOPA 61 32 AYUTUXTEPEQUE 11 8 CIUDAD DELGADO 57 18 CUSCATANCINGO 32 19 75 El Salvador Country Land Assessment World Bank ILOPANGO 30 10 MEJICANOS 95 33 NEJAPA 18 10 SAN MARCOS 45 16 SAN MARTIN 29 13 SAN SALVADOR 1663 507 SANTA TECLA 361 78 SOYAPANGO 83 22 TONACATEPEQUE 13 2 TOTAL 2615 826 4.7 General Discussion: Responsiveness of Institutional Framework for Urban Land Development The initial findings suggest that rather than adopt policies that were either resistant or indifferent to urbanization, the Government of El Salvador adopted a more proactive approach to facilitate the demographic shift that accompanied the economic structure transitions especially in San Salvador. Notably the umbrella planning agency OPAMSS, which is financed through building and subdivision permit fees, coordinates policies and regulations and streamlines procedures for obtaining permits in the metropolitan area (for all municipalities integrating the metropolitan area). Urban development in the country’s capital has been better regulated than in secondary cities. Feedback from private sector developers and other real estate professionals is that this coordinating mechanism in San Salvador has provided a predictable and fairly efficient framework for land development. The average time for approval of permits by OPAMSS ranges from 8 to 12 months. Also the process for the approval of low income housing subdivisions is regulated by special legislation, which requires only minimum urban planning requirements (Decree 708 of 1992). A regional comparison of typical time it takes to obtain all building permits to convert land from rural to urban use saw San Salvador rating very favorably at 2.5 months on average in 2006 (see Figure 4.14). Long times usually mean lesser elasticity in land supply leading to less affordable prices and more informal development. Shorter times for regulatory approvals can however be a double-edged sword. They may imply bureaucratic short cuts or lesser scrutiny leading to gaps in oversight that can contribute to less sustainable land use conversion. 76 El Salvador Country Land Assessment World Bank Figure 4.14: Regional Comparison (2006) of Typical Time it takes to obtain all permits to convert land from rural to urban use Source: Based on Urban Growth Management Initiative, Phase 2 Data, 2006 Although El Salvador ranks in the bottom third at 144 in the World Bank Group’s Doing Business 2012 Survey for Construction Permitting, the land-related elements of this process in San Salvador are not responsible for this unenviable placement. In particular, the time to receive inspection by the Office of Planning of Metropolitan Area of San Salvador (OPAMSS) is recorded as a single day and is cost-free. Also the time to obtain zoning or road construction revisions is 21 days at an average cost of $316, which is a mere 5% of the total permitting cost. The low ranking appears to be mostly as a consequence of the cost as a percentage of per capita income (168%) versus an OECD average of 45%. Total time taken (157 days) as an indicator of implementation arrangements efficiency is very similar to the OECD average (152 days) and considerably better than the regional average of 221 days. 77 El Salvador Country Land Assessment World Bank On the other hand, urban planning and development control outside of San Salvador is much weaker and construction permitting processes can reportedly take up to two years. Without the required permits, subdivided plots cannot be registered at CNR or used as collateral for loans in commercial banks. There are few direct links or coordination between the Vice- Ministry of Housing and Urban Development and the municipalities of San Miguel, Santa Ana, and those in the San Andres valley. These operate fairly autonomously, although occasionally they request technical assistance from the Vice-ministry. The satellite municipalities of Colon and San Juan Opico in La Libertad have experienced very significant urban growth that is directly related to their strong functional and economic linkages to AMSS. However, they do not fall under the jurisdiction of OPAMSS and Development Control is comparatively weaker in these municipalities and relatively uncoordinated with decisions in San Salvador. Since the early 2000s, the Government has been developing instruments for territorial development aimed at promoting the adequate use and development of the territory but with little impact. The elaboration of the National Plan for Territorial Planning and Development (PNODT), under the leadership of the Vice-ministry of Housing and Urban Development in the Ministry of Public Works, and the Ministry of Environment and Natural Resources started in 2001 through an open and participatory process, which was completed in 2004. At the same time, the Vice-ministry adopted a policy of decentralization in the area of urban planning and development and promoted the elaboration of local plans (at the municipal, micro-regional, and regional level) within the framework established by the national and regional plans. Specifically, Territorial Development Plans (PDT) were prepared for each of the country’s 14 Departments with the participation of the municipalities of each Department and other local stakeholders. However, the lack of coordination between the central and local/municipal levels has created a proliferation of local plans, which are not compatible with the PNODT and thus cannot be implemented or suffer significant delays in implementation. The main obstacles for initiating the process for the approval of urban projects and delays in the issuance of authorizations include:  the parcel or lot to be developed lacks a registered title, which is an impediment for initiating the approval process;  the subdivision to which the lot to be developed belongs has not been regularized, which is also an impediment for initiating the approval process  proliferation of requirements and procedural steps and lack of coordination or overlaps of responsibilities among entities involved in the process;  discretionality in the interpretation of applicable norms;  long delays for obtaining sectoral authorizations (e.g. from MARN, which can take up to one year) or the issuance of the final permit (additional year or more). Despite the creation of OPAMSS, the process for obtaining authorizations and permits also in the case of the San Salvador Metropolitan Area remain somewhat complex, with delays due mainly to: 78 El Salvador Country Land Assessment World Bank  the requirement for both OPAMSS and the relevant municipality to approve the subdivision;  the duplication of requirements and discrepancy among criteria used by OPAMSS and the municipalities of the metropolitan area (OPAMSS can approve but not impose sanctions, which can only be enforced by municipalities);  the fact that a significant percentage of subdivision in the metropolitan areas are located in disaster prone-risk areas, which impedes the issuance of any development authorization. Efforts to simplify the process for the approval of subdivisions and urban projects for low income housing created the One Stop-shop Window Program (Programa de Ventanilla Unica), the objective of which was to significantly reduce the paperwork, steps and costs to obtain a permit. However, institutional and legal bottlenecks remain, making the process still onerous and complex. A significant finding is that the rate of urban expansion in areas outside of contiguous San Salvador is far outstripping that of San Salvador. In the period 2000 to 2010, the municipalities of Colon and San Juan Opico west of San Salvador (AMSS) grew from practically no urban development to 26 square km of built space. Santa Ana more than doubled and San Miguel grew by more than 50%. Although a significant degree of the expansion in these areas appears to have been preceded by infrastructure (at least roads) and comprised infilling (densification), it is not apparent that sufficient attention was paid to environmental and soil potential issues. For example, in Santa Ana, urban expansion infringed on a significant amount (9 square kilometers) of Class II agricultural land. By comparison the equivalent figure in San Miguel was 4 square kilometers and in San Salvador was 6 square kilometers. Some forested lands also clearly gave way to urban expansion. Most of the expansion occurred within a few municipalities so inter-municipal coordination was less relevant than in San Salvador. The Urban Development and Construction Law of 1951 aims at regulating urban land markets and parcel subdivisions, however despite its existence, much informal urban development, subdivisions and sales have been taking place. These have been creating problems for the municipalities involved and the Vice-ministry of Housing and Urban Development, the lead agency responsible for overseeing compliance with this law. The majority of municipalities have their own ordinances to regulate urban land markets, but these are not necessarily in accordance with National or Regional Plans, as mandated by the Law. The Law, its regulations (from 1973, as amended in 1991), and the guidance provided by the Vice- Ministry provide a mostly adequate framework for regulating these markets; reducing the need for individual ordinances at the municipal level. 79 El Salvador Country Land Assessment World Bank Chapter 5 Urban Expansion and Inclusion 5.1 Overview of Chapter In this Chapter the extent to which urban spatial expansion in El Salvador has been occurring in an inclusive way is explored. The emphasis is on socio-economic inclusiveness assessed from two main perspectives. The first part of the Chapter examines the location and extent of precarious urban settlements in AMSS and its two main satellite municipalities (Colon and San Juan Opico), Santa Ana and San Miguel. The second part looks at land market prices in relation to the income distribution of the populations of these same cities in order to gauge affordability. The Chapter concludes with a discussion centered around the regulatory framework for land development as a contributor to the cost of servicing urban land. 5.2 Urban Expansion and Broader Socio-Economic Vulnerability This first Section builds upon a UNDP 2010 dataset and Study on precarious urban settlements in all of El Salvador. The objective of that Study was to assess the location and level of poverty and social exclusion in what they call asentamientos urbanos precarious (AUP) – which are basically marginalized areas in urban centers with low access to services. The methodology of this study had two parts: (1) assessing the primary locations of AUPs in El Salvador urban areas; and (2) assessing the level of social exclusion in each of these areas and the highest concentration of poverty within these areas in order for this model to serve as a foundation for designing poverty alleviation programs in urban areas. The methodology the UNDP Study used for identifying the main AUPs has 2 steps: first, to locate the main urban areas characterized by housing constructed with unstable materials and low access to basic services; and second, to characterize each manzana (territorial subdivision) by a level of poverty. Designation as an AUP does not necessarily mean the settlements are informal from a land tenure perspective as that layer of information was not available. The UNDP study used data from the 2008 population census (Direccion General de Estadistica y Censos) and housing census (Ministerio de Economia), as well as FUNDASAL and UNDP 2009 studies to locate AUPs in all 14 departments. This Report builds on the UNDP Study by introducing slope data (a key indicator of physical vulnerability) and by focusing on the subset of precarious settlements in the areas of urban expansion between 2001 and 2010. Firstly the location of all these settlements in the three main cities is overlaid with the Digital Elevation Model (DEM) for the country to get a sense of the extent to which precarious urban settlements in these cities are located on steep slopes. This is then repeated for only the areas of urban growth between 2001 and 2010 as identified in the previous Chapter. This latter analysis allowed quantification of the extent to which urban growth in the first decade of this century was characterized by growth in precarious 80 El Salvador Country Land Assessment World Bank settlements as well as by varying degree of vulnerability in such settlements. The Results are shown in Figures 5.1 through 5.4 and in Tables 5.1 through 5.4. In the main cities, more than a third and sometimes close to a half of all urban growth occurred in precarious settlements, with a majority being in settlements of extreme or high vulnerability in Santa Ana and San Miguel. Almost half (45%) of AMSS’s and Santa Ana’s new urban land growth in the first decade of this century occurred in precarious settlements. In AMSS about half of this precarious growth (44%) was in settlements of extreme or high vulnerability whereas in Santa Ana the equivalent proportion was two thirds (66%). For San Miguel more than a third of its new urban land growth in the first decade of this century occurred in precarious settlements with almost three quarters (72%) occurring in areas of high or extreme vulnerability. In the two main satellite municipalities of the AMSS, Colon and San Juan Opico, more than a third (38%) of the urban growth in the last decade occurred in precarious settlements with more than a third of this (37%) in settlements of extreme or high vulnerability. Figure 5.1: Location of Precarious Settlements in AMSS in Relation to Slope Table 5.1: Proportion of new urban growth 2001 to 2010 in the AMSS that was in Precarious Settlements Category of Precarious Settlement Area of all new urban growth Percentage of all new urban (2001-2010)in square km growth (2001-2010) Extreme 1.2 4% High 5.1 16% Moderate 5.0 15% Low 3.5 11% Total 14.8 45% 81 El Salvador Country Land Assessment World Bank Figure 5.2: Location of all Precarious Table 5.2: Proportion of new urban growth Settlements in AMSS in the municipalities 2001 to 2010 in the municipalities of Colon of Colon and San Juan Opico and San Juan Opico that was in Precarious Settlements Category of Area of all Percentage of Precarious new urban all new urban Settlement growth (2001- growth (2001- 2010)in 2010) square km Extreme 1.3 5 High 2.3 9 Moderate 5.9 23 Low 0.2 1 Total 9.7 38 Figure 5.3: Location of Precarious Settlements Santa Ana in Relation to Slope 82 El Salvador Country Land Assessment World Bank Table 5.3: Proportion of new urban growth 2001 to 2010 in Santa Ana that was in Precarious Settlements Category of Precarious Settlement Area of all new urban growth Percentage of all new urban (2001-2010)square km growth (2001-2010) Extreme 1.5 8% High 4.5 22% Moderate 2.2 11% Low 0.8 4% Total 9.0 45% Figure 5.4: Location of all Precarious Settlements San Miguel in Relation to Slope Figure 5.5: Proportion of new urban growth 2001 to 2010 in San Miguel that was in Precarious Settlements Category of Precarious Settlement Area of all new urban growth Percentage of all new urban (2001-2010)in square km growth (2001-2010) Extreme 0.8 5% High 2.9 21% Moderate 1.0 7% Low 0.4 3% Total 5.0 36% 5.3 Urban Land Prices and Affordability This Section takes a more direct look at the affordability of urban land by comparing the cost of a typical small urban plot of 200 square meters in the three main cities with the income distribution of the population in the Departments in which those cities are located. The land prices considered are at the lower end of prices for formal private sector land 83 El Salvador Country Land Assessment World Bank subdivisions sourced from a listing of prices for 2009/2010 obtained from the Salvadorian Real Estate Association (See Table 5.5 and Figure 5.6). They are therefore considered to be a good indication of market prices in that period. The income distribution data is sourced from the Annual Household Surveys (Encuestra de Hogares) of the corresponding year, 2009 and is disaggregated by Department to provide a better match for the geographic areas in which the sub-divisions are located. Figure 5.6: Spot Land Prices 2009/2010 Source: Salvadorian Real Estate Association 84 El Salvador Country Land Assessment World Bank Figure 5.7: Urban land prices in San Salvador Metropolitan Area (US$ per vara cuadrada)21 Source: Compiled by Study Team using Market Data supplied by the Salvadorian Real Estate Association Table 5.4: Average Monthly Incomes and Percentage of Population Living in Poverty 2005 and 2009 Santa Santa San San AMSS AMSS Ana Ana Miguel Miguel 2005 2009 2005 2009 2005 2009 Average Income (US$) 575 773 336 490 Average Income Per Capita (US$) 172 190 101 120 % of Poor 0.24 0.33 0.38 0.48 % of Extreme Poor 0.06 0.08 0.14 0.16 Source: Compiled by Study Team from data in the Annual Household Surveys (EHPM) Based on these data, a typical small plot at the low end of the market is relatively affordable in the San Salvador metropolitan area (AMSS), moderately affordable in Santa Ana and largely unaffordable in San Miguel. These results are based on a working definition of affordability being that the cost of the land does not exceed three times annual household income22. Table 5.5 summarizes the results. They show that in San Salvador (AMSS), one quarter of households in the corresponding Department would find land prices at the lower end of the formal market unaffordable. In Santa Ana this proportion grows to 40% and in San Miguel 21 1 vara cuadrada is equivalent to 0.7 square meters. 22 This is based in part on the 5th Annual Demographia International Housing Affordability Survey Report (Cox W. and Pavletich H.) which deems Median House Price to Median Household Income ratios of above 4 as being seriously or severely unaffordable. A ratio of 3 was used in this El Salvador Study because the prices analyzed were land prices only, not housing and land prices combined as used in the Demographia International Survey. Also the Demographia International Survey only considers data from metropolitan markets in six highly developed countries. 85 El Salvador Country Land Assessment World Bank it is much higher at 80%23. If this analysis assumes that the cost of land as a proportion of the total cost of land and housing was less than three-quarters (eg, half), affordability of urbanized land in each of these cities would be significantly lower. Table 5.5: Land Affordability in 2009 San Salvador (AMSS) Santa Ana San Miguel Typical small plot size 200m2 200m2 200m2 Cost of a typical small plot $11,400 $8,600 $25,700 Proportion of the 25% 40% 80% population in the Department for whom a typical small plot at the lower end of the market is unaffordable Source: Based on original analysis of income distribution data from the Annual Household Survey (EHPM 2009) and land market data obtained from the Salvadorian Real Estate Association 5.4 General Discussion: Inclusion and the Regulatory Framework for Land Development The degree of socio-economic accessibility of El Salvador’s main cities is a lot less even than the relatively regular physical pattern of land development that appears to have characterized much of their spatial expansion in the last decade. This Chapter found significant socio-economic vulnerability in new settlements in the main cities, accounting for between one third and a half of all urban growth between 1998 and 2008. These settlements would be typically lacking one or more basic urban utility service as well as have houses constructed from unstable materials These trends are a cause for public policy concern as they suggest that a significant part of El Salvador’s urban expansion is occurring with considerable socio-economic vulnerability. Comparisons of income distributions and urban land prices at the lower end of the markets in the three main cities also exposed significant affordability challenges for populations in secondary cities. These segments of unaffordability are likely underestimates given that the migrant population targeting the country’s main cities is likely to be coming from multiple Departments, many with lower income distributions than the host Departments of the cities. The picture is likely to be further complicated by the degree of accuracy to which remittances have been factored into incomes. 23 Notably the San Miguel analysis was made on the basis of very few available sub-division prices, however local stakeholders perceived the reported prices to be realistic estimates for that sub-market. 86 El Salvador Country Land Assessment World Bank An apparent correlation between the location of precarious settlements and steep slopes in some parts of the AMSS and Santa Ana was also noted, exposing a double-edged vulnerability for some. This correlation is also a cause for further public policy concern as it implies that the vulnerability of households located in such settlements may run deeper than socio-economic variables indicate and extend to a susceptibility to outright loss of life and or assets due to sudden onset natural disasters that are associated with living on steep slopes in tropical climates. Still, by at least one measure, three quarters of the population of San Salvador Department can afford a typical small plot in the tail end of the metropolitan area which suggests that some aspect of the regulatory framework for land development may be conducive to a relatively elastic supply of land for low-income households. Conduciveness in this respect would mean not over-prescribing the standards that developers have to comply with in order to respond to effective demand. In this regard, a key regulation, the minimum plot size, is realistically prescribed in El Salvador. The Regulations of the Law of Development of AMSS permit three types of residential sub-divisions: Social Interest Settlements, Popular Settlements and Residential Settlements. In San Salvador, formal plots can be as small as 75 square meters in Social Interest or Popular housing settlements and 250 square meters in the next category of Residential Settlements. Subdivisions in Santa Ana and San Miguel are guided by the ‘Reglamento de la Ley de Urbanismo y Construcción’ which allow five categories of sub-divisions starting with category D1 of plots up 100 square meters and category D2 for plots between 100 and 200 meters square. These are modest by international standards and imply a degree of realism on the part of the Government in not setting this important regulatory parameter at a level way above the means of the population. Table 5.6: Minimum Plot Sizes in San Salvador (AMSS) by Settlement Type Settlement Lot (M2) Density (persons/hectare) Social Interest 75 800 Popular 75 800 Residential 125 400 Residential 250 200 Residential 500 100 Residential 1000 50 87 El Salvador Country Land Assessment World Bank A regional comparison of another key regulatory provision, the minimum road widths in urban subdivisions also shows San Salvador in a favorable light. Minimum road width in urban subdivisions around the fringe of the city was comparatively narrow according to a 2006 Survey making for lower costs to developers (Figure 5.8). Normally high public space requirements, including road width requirements, translate into less saleable land in the subdivision and thereby materialize as higher per unit costs of land for consumers. Figur 5.8: Regional Comparison (2006) of Minimum Road Width (m) in a new legal subdivision on the urban fringe Source: Based on Urban Growth Management Initiative, Phase 2 Data, 2006 In summary, although ‘slums’ as they are known in much of Latin America and the rest of the developing world do not exist in significant measure in El Salvador, a substantial part of the recent Salvadorian urbanization storyline is one of socio-economic vulnerability. The appearance of formality whereby rational road networks appear to precede significant settlement is somewhat deceiving as other data suggests that many basic services are missing in these new settlements which are also characterized by non-durable housing. 88 El Salvador Country Land Assessment World Bank Informality offers entry into a system at different price points24 and the mixed storyline of this Chapter is consistent with the prevalence of informal residential subdivisions that reportedly number over 300 nationally. Given the affordability indicators observed in this Chapter, it is likely that significant segments of El Salvador’s urban population are gaining access to urban spaces by entering at price points below those on offer in the formal market i.e. by purchasing plots in the informal sector, most commonly in informal sub-divisions. Given that several key regulatory measures for land development in El Salvador are realistically prescribed, low incomes may continue to be the binding constraint sustaining informal markets for some time. Applications for development permission after the fact of land conversion are reportedly common. The efficiency of the regularization process 25 that follows, if it follows, will be a key determinant of how long these vulnerable segments of the population remain in situations of compromised welfare. 24 This conceptualization of informality was discussed in February 2012 at a Seminar on ‘The Political Economy of Green Growth’ at the World Bank, Washington D.C. 25 The most significant bottlenecks in regularizing urban parcels include mortgages which have not been cancelled in the Registry, unregistered subdivisions or amendments to existing parcels, and inaccurate liens on properties. 89 El Salvador Country Land Assessment World Bank SECTION III: LAND USE CHANGE AND DEFORESTATION 90 El Salvador Country Land Assessment World Bank Chapter 6 Land Use and Deforestation 6.1 Overview of Chapter In this Chapter the findings of original analyses of land use in relation to deforestation are presented. Firstly, it reports on the scale, rate and geographic distribution of deforestation. Secondly, it investigates possible correlations between deforestation and various factors that are commonly thought to play a contributory role. It ends with a short discussion about deforestation and public policy. Methodologically, change detections of natural forest cover to other forms of land use were conducted comparing the Ministry of Environment’s very detailed Ecosystems Map based on 1998 satellite images and ground-truthing with a newly created, and equally detailed baseline map for 1998 and an update for 2011. Further details of the methodology can be found in Annex3. 6.2 Scale, Rate and Distribution of Deforestation During the period 1998 - 2008 a total of about 44,000 hectares of natural forests were lost at an average annual deforestation rate of 1.5%. Maps 7.1 and 7.2 visualize the loss in natural vegetation cover between 1998 – 2008 and 2008 – 2011 respectively. Map 6.1:Changes in the physiognomical structure of the ecosystems in El Salvador between 1998 and 2008 91 El Salvador Country Land Assessment World Bank Green: Natural ecosystems year 2008; Red: Lost natural ecosystems since 1998; Yellow: Other within the Conservation Areas; Grey: Other outside the Conservation Areas. The categories “Otros” - other – refer to any land use other than (semi -) natural ecosystems. During the period 2008 - 2011, only 1,334 ha of natural forest were lost at a much reduced average annual deforestation rate of 0.17%. While the rate in this second period is significantly lower, the fact that there were only three points of measurement in 13 years does not preclude the possibility that the trend may have started earlier than 2008. Map 6.2: Changes in the physiognomical structure of the ecosystems in El Salvador between 2008 and 2011 Green: Natural ecosystems year 2011; Red: Lost natural ecosystems since 2008; Yellow: “Other within the Conservation Areas; Grey: “” Other” outside the Conservation Areas. These deforestation rates are mostly consistent with the estimates made in the FAO Global Forest Resources Assessment of 2010 but there is one important difference. In the FAO Report, the average annual deforestation rate between 2000 and 2005 was 1.43% and for 2005 to 2010 was -1.47%26. So in aggregate, over the first decade of this century, the 26 In the FAO Report, the estimated forest cover in 2000 and 2010 was 332,000 ha and 287,000 ha respectively compared to calculations of 311,000 ha and 270,000 ha in 1998 and 2008 in this Study. This translates to roughly equal annual rates of deforestation between the two Studies. 92 El Salvador Country Land Assessment World Bank FAO Report estimated deforestation to occur at an annual average rate of around 1.4% which is highly consistent with the rate of 1.5% observed in this Study between 1998 and 2008. However in the absence of an intermediate sampling point between 2005 and 2010 in the FAO Study, the impression is created that deforestation continued at a steady rate in the latter half of the decade. This is in comparison to a finding in this Study of a drastically reduced rate at least since 2008. As this Study is based on a much more detailed matching of local forests than was done for the FAO global Study, the significant drop in rate is a reliable measure. However, by comparing the two Studies it can be deduced that the sharp decline in annual deforestation rates likely started after 2005 but no later than 2008. The conversion of forests was mostly in lowland tropical deciduous broad-leaved forest, on well drained soils. The changes are summarized in Figure 7.1, however, Annex 4 contains a more detailed listing of the changes from natural vegetation cover to productive land use for each ecosystem type. In that Annex, the changes of natural forests to other land uses are separated from non-forest natural vegetation types. For other users of the data, the changes from other ecosystem types to productive land use are also presented there. Figure 6.1: Vegetation Cover for Natural Ecosystem for the Years 1998, 2008 and 2011 See Annex 4 For Key to ecosystem codes Most of the lost forests were converted to annual crops and pastures for what appears to be subsistence farming landscape. Only field verification during the growing season can make this assessment with confidence including the identification of the specific crops. However, it is likely that in relatively steep hilly terrain, the deforestation has been carried 93 El Salvador Country Land Assessment World Bank out to create grazing land, whereas in more level terrain, the forest may have been cleared for the preparation of agricultural fields. During both periods there was some clear-cutting of mangroves for the creation shrimp farms, but the sizes of the losses were not all that large with 91 and 32 ha respectively, for each period. It is also difficult to distinguish a gradual conversion of natural forest to small scale shade coffee plantations, however, in spite of specific attention in the intensive coffee growing areas, no such conversion was identified. It is possible the coffee growing certification may have a positive effect on protecting the last remaining stands of natural forests. At the same time, it is important to realize that this lack of identification may be due to poor visual distinction between the two land use types in satellite images. Deforestation was greatest in the north-west, central and south-east regions of the country. It was almost non-existent in the South-West and only minor in the North-East. In the North-East, deforestation for farming purposes has been less intense, probably because the land is steep and growing of crops including coffee in remote areas appears to be less attractive. That said, there are both abandoned and freshly planted coffee groves in the North-East. Tables 6.1 and 6.2 show the municipalities with deforestation rates greater than 50% and between 25% and 50%, respectively. There were eight municipalities in the first category and twelve in the latter. Municipalities with baseline forests in 1998 of less than 1000ha were excluded from this categorization. Table 6.1: Municipalities with Deforestation Rates Greater than 50% between 1998 and 2008 Municipality Change_ha % Change Victoria -3907 -95% Texistepeque -1704 -76% Poloros -1500 -74% Nueva Concepcion -1147 -63% San Ildefonso -1063 -55% San Isidro -1250 -54% -1341 -53% San Antonio 94 El Salvador Country Land Assessment World Bank Pajonal Tenancingo -1085 -53% Table 6.2: Municipalities with Deforestation Rates Greater than of 25 and 50% between 1998 and 2008 Municipality Change_ha % Change Dulce Nombre de Maria -1414 -49% San Esteban Catarina -932 -41% La Union -1347 -39% Nuevo Eden de San Juan -1680 -38% Santa Ana -588 -32% El Sauce -932 -30% Intipuca -1039 -29% San Dionisio -619 -28% Dolores -568 -25% Sensuntepeque -636 -25% Concepcion Batres -1034 -24% Metapan -3231 -22% The degree of thinning of the forests through selective logging of mature trees and the collection of firewood, the main cooking fuel for rural El Salvador, is not known. What can be seen on satellite imagery are changes from forest to crops or pastures. This harvesting of wood, in many cases keeps the forests in a very early pioneer stage of secondary forest. From an environmental point of view, this is worrisome, as young forests have relatively low biodiversity value as compared to mature forests. 95 El Salvador Country Land Assessment World Bank 6.3 Deforestation and Public Policies in the Agrarian Sector The causes of deforestation and of the decrease in average annual deforestation rates between the periods 1998 – 2008 and 2008 – 2011 can't be defined without further knowledge of the socio-economic processes that took place in rural El Salvador. Neither this Study nor the fieldwork for producing the Ministry of Environment’s Eco Systems Map focused on investigating possible causes. In the absence of such a focus, what is presented in this Section is a preliminary analysis of two potential factors identified by local stakeholders as areas of public policy in the agrarian sector that may have some relationship to deforestation. The potential role of land redistribution programs, subsidized distribution of seed and fertilizers and agrarian structure were explored by performing correlation analyses but none of the associated variables was found to have high value or statistically significant correlation. In these analyses, the deforestation variable is the percentage change of forest cover between 1998 and 2008 by municipality. Two variables were included to measure the effect of land redistribution programs on deforestation. One variable measures the amount of land redistributed in the municipality by various programs associated with the Agrarian Reform (PhaseI, Phase III, Decree 839, 3 of July Accords). Land distributed under the PTT by department is included in a separate variable. Another potential factor contributing to deforestation is participation in government-sponsored seed programs, which could encourage the expansion of agriculture into forested areas. The potential effect of this type of program is measured by the amount of land cultivated with improved seeds for growing maize, both total area in the municipality and average percentage of total cultivated area. The production yields in basic grains and sugarcane are also included in the analysis to examine the effect of agricultural productivity. Lastly, a number of variables are included to capture the effects of the agrarian structure; these variables include farm land and number of producers, distinguishing those in farms with less than 2 hectares. The correlation matrix in Table 7.3 shows that none of the variables has a high value or statistically significant correlation coefficient. Table 6.3: Summary Results of Correlation Analysis of Deforestation N Mean Min. Max. Corr. Deforestation (% of forest change) 173 -0.11 -1.00 0.58 - Agrarian Reform (ha) 173 1,867.86 0.00 24,851.00 -0.025 PTT (ha) 173 11,494.75 2,455.90 37,982.40 -0.016 Improved maize seed (ha) 173 253.41 1.92 1,837.69 -0.042 Improved maize seed (% of 173 0.74 0.16 1.00 -0.053 96 El Salvador Country Land Assessment World Bank cultivated) Sugar cane (ton/ha) 108 92.01 48.65 157.87 0.097 Maize yields (qq/ha) 173 54.00 26.92 80.10 -0.059 Sorghum yields (qq/ha) 166 29.82 16.71 66.77 -0.026 Bean yields (qq/ha) 173 16.88 2.86 25.93 0.064 Rice yields (qq/ha) 80 107.31 14.31 201.89 -0.094 Agricultural land in farms < 2 ha 173 1,012.34 26.99 5,379.03 -0.009 (ha) Agricultural land in farms < 2 ha (% 173 0.30 0.01 0.80 -0.052 of total land) Total agricultural land (ha) 173 4,523.65 180.76 29,047.86 -0.022 Producers with < 2 ha 173 311.39 18.00 1,464.00 0.036 Producers with < 2 ha (% of total 173 0.73 0.14 0.95 -0.038 farms) Total producers 173 432.15 42.00 2,218.00 0.023 It is important to emphasize that a systematic Study of the causes of deforestation was not undertaken as part of this evaluation. However, as part of it preparations under the REDD agenda, the Government of El Salvador is currently conducting such an investigation. 6.4 General Discussion Broadly, all ecosystems in El Salvador have undergone rather severe human interventions. Of the 19 identified terrestrial ecosystems, 12 are forests, most of which are secondary forest and the few stands that many Salvadorian biologists consider primary forests have at least undergone significant selective logging. Therefore, all forest ecosystems have been classified with the extension “secondary or intervened”. From a biodiversity point of view, this means that all forests are under stress and they are less rich in biodiversity than the original forests have been. From a REDD+ point of view, all forests have potential for carbon sequestration. 97 El Salvador Country Land Assessment World Bank It is noteworthy that a significant part of the deforestation occurred within formally demarcated Conservation Areas. This was shown in Maps 7.1 and 7.2 which visualized the “Áreas de Conservación”, that were developed and delineated in 2004 as part of the “Plan Nacional de Ordenamiento y Desarrollo Territorial (PNODT)” - the “National Plan for territorial zoning and development”. These were meant as zoning units, supposedly where certain regulation would favor protected areas and biological corridors. However, although defined in the Law on Protected Areas, no further regulation has been defined legally. Also in practice, no specific land use practices or conservation oriented management has been promoted by the Government outside the legally acquired protected areas. Finally, during the periods of the evaluations, there was no change towards recovery from productive land use to natural ecosystems. All changes were conversions form natural ecosystems to crops, grazing and other productive land use. This is a cause for public policy concern as there is a perceived need to restore some previously degraded or lost eco- systems to create a more sustainable balance between different land uses in a land- constrained country. 98 El Salvador Country Land Assessment World Bank SECTION IV: LAND IN EL SALVADOR – MULTI-SECTORAL PERSPECTIVES 99 El Salvador Country Land Assessment World Bank Chapter 7 Land Market Trends 7.1 Overview of Chapter In this Chapter, land market data are analyzed to identify trends and land use dynamism in the first decade of this century. It begins with an analysis of transaction trends in both urban and rural parts of municipalities over the period 2001-2010. This is followed by a more detailed comparison of trends in the cities of Santa Ana and Sonsonate and two contiguous municipalities. The Chapter concludes with a short general discussion. A CNR dataset of all registered transactions for the departments of Ahuachapan, Santa Ana, and Sonsonate is used. These locations were selected because they are the only Departments where CNR is running a computerized registry-cadastr platform. The dataset includes the number of transactions recorded in the registry for various types of requests (sales, bequests, donations, mortgages, collateralized credit obligations) as well as their average size and price for properties in urban and rural land markets. CNR determined whether the properties were urban or rural before starting the cadastral field survey in 2001 based on their proximity to the original urban perimeter and size. The data covers 41 municipalities for the years 2001, 2005, and 2010. 7.2 Land Market Trends in Ahuachapan, Santa Ana, and Sonsonate- 2001-2010 Formal land transactions increased significantly between 2001 and 2005, the period that coincides with the implementation of the Land Administration Project. The Bank-financed Land Administration Project (Phase I of the Program) financed systematic cadastral surveying and deeds registration between 1997 and 2005. It is expected that the investments carried out in modernizing the property rights systems would have a positive effect on incentivizing formal land markets by reducing transaction costs. Table 7.1 shows the number, size, and price of four types of transactions recorded in CNR. The volume of registered transactions increased significantly between 2001 and 2005. Some municipalities went from having a handful of registered transactions to hundreds of them. In very few municipalities, the actual number of transactions dropped. During this period, total registered sales increased by 77%, mortgages by 33%, bequests by 113% and donations by 189%.27 These trends were similar for urban and rural transactions. The total number of registered transactions peaked in 2005, and by 2010 it had declined for mortgages, bequests, and donations. The number of sales continued to increase by 2.2% between 2005 and 2010. These results suggest that the Land Administration Project had a positive impact 27 Only the total volume of collateralized credit obligations decreased 46% during this period (not shown in the table). 100 El Salvador Country Land Assessment World Bank on increasing formal transactions, particularly among bequests and donations, which often occurr among family members and are not formalized. Formal land transactions increased more in rural than urban land markets between 2001 and 2010. The number of transactions relative to the total number of properties in each municipality is recorded as a percentage in Table 7.1. In 2001, the percentage of properties sold was 2.1% in urban land markets and 1.7% in rural land markets. In 2005, the percentage of properties sold was higher in rural areas (3.6%) than urban areas (3.1%). In 2010, the percentages of properties sold decreased for both areas, but rural land markets maintained the highest level. Mortgages, bequests, and donations represent a small percentage of the total number of properties, but these transactions also experienced more growth in rural than urban areas between 2001 and 2005. Considering that this period coincides with the Land Administration Project, these trends suggests a greater reduction of transaction costs in rural than urban areas in relative terms. Table 7.1: Land Market Transactions in the Departments of Ahuachapan, Santa Ana, and Sonsonate – 2001, 2005 and 2010 Type of Year Urban Markets Rural Markets Transaction Total Mean Mean % of Total Mean Mean % of Num. Size Price Properties Num. Size Price Properties Sale 2001 2,279 1,654 21.81 2.12% 8,092 26,760 1.31 1.71% 2005 3,899 845 22.58 3.07% 14,508 23,768 1.65 3.62% 2010 3,587 1,969 24.89 2.65% 15,228 16,636 1.26 3.13% Mortgage 2001 2,681 4,195 41.73 1.02% 5,919 73,029 4.04 0.60% 2005 3,597 827 43.42 1.18% 7,911 36,286 4.79 0.85% 2010 2,804 3,986 56.71 0.99% 5,871 27,759 4.08 0.61% Bequest 2001 234 234 7,030 0.24% 570 74,819 0.00 0.14% 2005 510 510 3,037 0.38% 1,203 38,101 0.03 0.32% 2010 363 363 21,028 0.32% 1,184 27,430 0.53 0.29% Donation 2001 134 134 942 0.10% 402 28,432 1.38 0.06% 2005 355 355 508 0.22% 1,196 24,468 2.47 0.52% 2010 368 368 2,985 0.25% 807 34,385 1.37 0.19% Source: Author’s analyses of data obtained from CNR Notes: (1) Total number is the sum of all transactions in that year; (2) Mean size is the average size by municipality in square meters; (3) Mean price is the average price per square meter by municipality in US$. Prices for land sales remained relatively stable between 2001 and 2010. Figures 7.1 show the distribution of the 25th, 50th, and 75th percentiles of land sale prices per square meter in a box plot display, along with the outside values, for urban and rural areas respectively. Prices were adjusted by the Consumer Price Index using 2005 as the base year. In both urban and rural markets the cut-off points for the 25th, 50th, and 75th percentiles show little variation between 101 El Salvador Country Land Assessment World Bank 2001 and 2010. However, the areas contiguous to the cities of Santa Ana and Sonsonate show different trends. Figure 7.1: Box Plot of Land Sale Prices in the Departments of Ahuachapan, Santa Ana, and Sonsonate – 2001, 2005 and 2010 Average Price per Sq. Meter in Urban Land Markets Average Price per Sq.Meter in Rural Land Markets Sales Sales 150 30 Average Real Price per Sq.Meter 100 20 50 10 0 0 2001 2005 2010 2001 2005 2010 The average size of properties being sold has rebounded in urban land markets and decreased in rural land markets between 2001 and 2010. Figures 7.2 show the box plot display for average land sizes in urban and rural areas. In urban areas, the average size of a land plot in a sale decreased from 1,722 to 875 square meters between 2001 and 2004, but then increased again to 2,059. One outlier is the municipality of Texistepeque in the department of Santa Ana which had the largest average plot size recorded in urban areas in 2010 (36,021 square meters). On the other hand, rural properties being sold are shrinking in size. Average sizes have decreased from 28,047 to 24,865 square meters between 2001 and 2005, and further to 17,272 square meters in 2010. Even the outliers show a decreasing trend for rural properties. This trend in rural land markets confirm the fragmentation of landholdings that was observed in the comparison of the 1971 and 2008 agricultural census data. 102 El Salvador Country Land Assessment World Bank Figure 7.2: Box plot display of average land sizes in sales in the Departments of Ahuachapan, Santa Ana, and Sonsonate – 2001, 2005 and 2010 Average Size in Urban Land Markets Average Size in Rural Land Markets Sales Sales 40,000 200000 30,000 150000 20,000 size size 100000 10,000 50,000 0 0 2001 2005 2010 2001 2005 2010 7.3 Land Market Trends in the Cities of Santa Ana and Sonsonate and Contiguous Municipalities Land prices in rural areas contiguous to the cities of Santa Ana and Sonsonate behaved differently from prices in other municipalities. Table 7.2 displays the land market data for the municipalities of Santa Ana and San Sebastian Salitrillo in the department of Santa Ana, and for the municipalities of Sonsonate and Sonzacate in the department of Sonsonate. The municipality of Santa Ana had an average price in urban areas of US$55 per square meter in 2001. The price in urban areas grew to US$63 and $US81 per square meter in 2005 and 2010 respectively. Mean price in the rural areas of this municipality fell in 2005 and 2010 compared to their base in 2001. In contrast, the nearby municipality of San Sebastian Salitrillo experienced a significant increase in land sales of rural properties from 18 in 2001 to 773 in 2005. Average prices in rural areas also increased substantially from US$1.53 per square meter in 2001 to US$7.5 and US$13.4 per meter in 2005 and 2010 respectively. In the department of Sonsonate, the municipality of Sonzacate, which is contiguous to the city of Sonsonate, had the highest recorded value for rural properties in 2001 and 2005, and the second highest in 2010. These trends suggest that there is high dynamism in rural land markets near large cities, and that the level of development that is occurring on these lands is perceived to be of high value. 103 El Salvador Country Land Assessment World Bank Table 7.2: Land market data for the municipalities of Santa Ana, San Sebastian, Salitrillo Sonsonate and Sonzacate Municipality Yea Urban Markets Rural Markets r Tota Tota Mea % of Mea % of l Mean l Mean n Propertie n Propertie Num Price Num Size Size s Price s . . Santa Ana 200 1,62 1 663 342 55.47 1.62% 5 2,790 11.78 2.43% 200 1,32 2,44 5 4 276 62.93 3.23% 7 4,320 2.70 3.65% 201 1,24 1,80 0 6 242 81.00 3.04% 4 7,649 1.30 2.69% San Sebastian Salitrillo 200 1 0 . . 0.00% 18 3,468 1.53 0.23% 200 114.0 5 92 95 3 1.72% 773 2,112 7.51 9.69% 201 123.0 0 77 77 0 1.44% 273 747 13.37 3.42% Sonsonate 200 24,89 1 386 316 44.43 3.36% 392 9 1.33 2.12% 200 10,38 5 392 284 67.37 3.41% 724 8 0.82 3.92% 201 18,16 0 383 209 66.06 3.33% 552 8 0.68 2.99% Sonzacate 200 1 77 362 46.91 2.24% 92 522 23.07 1.66% 200 5 125 228 68.26 3.64% 290 423 31.47 5.23% 201 0 101 203 71.19 2.94% 141 757 8.32 2.54% Source: Author’s analysis of data obtained from CNR 7.4 General Discussion In relatively free markets, prices and volume of transactions are indicators of dynamism as they normally capitalize information about public policy including projected infrastructure expansion as well as planned private sector investments. Steep rises in prices often indicate relative scarcity usually due to inelastic supply. Whereas the analysis in previous Chapters, especially Chapter 4, made some reference to prevailing land market prices in specific urban subdivisions, this Chapter draws on much more comprehensive datasets of land market 104 El Salvador Country Land Assessment World Bank transactions although the spatial specificity of those transactions is limited to a classification of urban or rural at the municipal level. Land markets in the departments of Ahuachapan, Santa Ana, and Sonsonate experienced more dynamism in the rural than the urban sector. The number of registered transactions increased the most in rural areas compared to urban ones. Prices of land sales in rural areas also show an upward trend, with notable outliers in places contiguous to large urban areas. The dynamism in the municipality of San Sebastian Salitrillo confirms the pattern of spatial expansion of the city of Santa Ana which physically grew into this municipality during the period in which this land market dynamism was observed (see Figure 4.5). Furthermore, rural properties being sold continue to decrease in size suggesting greater demand for the land. In some cases the demand is driven by urbanization, but the trend is more general. 105 El Salvador Country Land Assessment World Bank Chapter 8 An Integrated Perspective of Land Use and Public Policy 8.1 Overview of this Chapter This Chapter looks at the Study’s empirical findings from a more integrated, cross -sectoral perspective so that their implications for development are better understood. In doing so, the reference framework is the way in which land ideally ought to feature in the development process. This includes that land functions as a resource that changes use and form efficiently in response to opportunities for greater productivity created by market forces including changes in demographic and economic structures. It also includes being a resource that is accessible to the vast majority of the population and one which affords holders of rights sufficient security of tenure to make productive and welfare-enhancing investments. Finally, as a finite resource, land uses also need to be sustainable. The Chapter is structured in three parts. The first section identifies and discusses areas in which there appears to be congruence between public policy and outcomes on the ground. Next the focus is placed on areas where this congruence is lacking. Finally, a brief assessment is made of the existing institutions that are meant to mediate conflicts and promote efficiency in land use in the national interest. 8.2 Where are public policy objectives being achieved? 8.2.1 Elasticity in conversion of rural and peri-urban land to urban use The rate of conversion of rural and peri-urban land to urban use in El Salvador has been relatively elastic in the last decade. Most of the urban physical expansion of the last decade appears to have been preceded by infrastructure, especially roads. The most intense areas of expansion were mostly characterized by systematic patterns of rights of way, implying a degree of formality. San Salvador and the secondary cities do not have the extent of poorly serviced squatter settlements and land invasions observed in many other metropolitan areas of Latin American and the rest of the globe. This suggests that land supply, at least in quantitative terms, is mostly keeping up with the demographic transition of urbanization, although much of it may be through informal sub-divisions. The elasticity of supply is also reflected in land prices. Available data showed moderately growing or stable real urban and rural land prices over the last decade. This may be partly the result of deflationary pressures after rapid increases in land prices in the 1990s. Consistent with this, other indicators show relative affordability of land in relation to annual incomes in San Salvador and Santa Ana, though less so in San Miguel. Private sector real estate developers also acknowledge the relative efficiency of transaction costs for development permitting under OPAMSS in AMSS. By these indicators, outcomes on the ground and public policy ambitions and mechanisms for inclusive urbanization, are somewhat congruent. 106 El Salvador Country Land Assessment World Bank 8.2.2 Effectiveness of institutional mechanisms for mediating competing demands for peri-urban land in contiguous AMSS In the last decade, land use zoning and development control as mechanisms for mediating the competing demands for land in peri-urban spaces in the contiguous metropolitan area of San Salvador have mostly worked. With few exceptions, notably conversion of some land zoned for agriculture in the north-east and some infringement of land zoned for maximum protection in the south and north-west, most of the growth in AMSS has occurred in the ‘urban zones’. This may be an indicator of the effectiveness of the coordinated approach to land use planning and development control through OPAMSS and the fourteen municipalities that comprise AMSS. However, there are two important caveats. Firstly, in the municipalities of Colon and San Juan Opico in the neighboring Department of La Libertad, an almost equivalent amount of urban expansion (26 square kilometers) has occurred in the last decade as in all of the AMSS, and there the level of development control has been weaker. Although this area effectively serves as a satellite to the AMSS due to its strong economic and functional connections, land use zoning and development control are not coordinated through OPAMSS, but rather through the Office of Planning for the San Andrés Valley. Secondly, it appears that zoning did not adequately exclude steep slopes around AMSS from urbanization. Although only a small proportion of AMSS’s expansion has been in ‘no-build’ zones, a much larger proportion has been on slopes close to or in excess of 30 degrees. 8.2.3 Reduction in rural land-based inequality There is significantly more equality in the distribution of land ownership and access in rural El Salvador today than in the past. The combination of active policies and market forces has dramatically changed access to land. Inequality in rural land ownership has dropped very significantly in key departments where the Agrarian Reform was implemented. In rural areas, the proportion of landless rural households dropped by more than half from 41 percent in 1971 to just 18 percent in 2008. Noting that the agrarian reform did not actively create micro parcels, the rapid fragmentation of rural parcels has also been driven by market forces, partly as a result of declining profitability in farming and the need for rural homesteads stimulated by large influxes of remittances from abroad. Today, a quarter of rural lands are occupied by homesteads of less than 2 hectares, as compared with just 10% in 1971. 8.2.4 Stemming of deforestation rates in recent years At least since 2008, the average annual rate at which agricultural land use has been displacing forests has dropped significantly. The average annual rate of deforestation in the period 2008-2011 was 0.17%, one ninth of what it averaged between 1998 and 2008 when most of the cleared forests appear to have been replaced by annual crops or pasture. While this outcome does not account for forest thinning through selective logging, it is consistent with expressed public policy to pursue a development trajectory that is more environmentally sustainable. However, the marked decrease in deforestation rates may be attributable to factors other than land-related public policy in the field, since there has been no significant shift in the 107 El Salvador Country Land Assessment World Bank regulatory framework or implementation capacity related to environmental management in recent years. 8.3 Where are public policy goals clashing? 8.3.1 Accessible urban settlement locations and disaster vulnerability due to topography In the last decade, land use zoning and development control mechanisms for directing urban settlement away from some areas vulnerable to suffering or exacerbating natural disaster damage have not worked well. A significant amount of urban expansion around AMSS and Santa Ana in particular has occurred on slopes close to or greater than 30% which are particularly susceptible to land slippage after heavy rains such as during tropical storms. Even the 2001 earthquakes triggered landslides that caused many fatalities on settlements close to or on steep slopes, particularly in the Santa Tecla municipality of AMSS where deforestation associated with residential land development had been significant. Notably urban expansion in the last decade in parts of San Salvador and Santa Ana has also displaced forests which may have had key ecological and watershed functions that mitigate disaster risks. This pattern of urban expansion indicates that while the rate of urban land supply may be keeping pace with demand, the sustainability of the created housing solutions is still in questionThis inconsistency may reflect an inadequate incorporation of natural disaster vulnerability criteria into land use planning and development control mechanisms. It also suggests that regulatory provisions and land prices are not creating sufficiently strong incentives to encourage vertical and more compact development. 8.3.2 Urban settlement expansion on scarce, good agricultural potential soils Land Use Zoning and Development Control mechanisms in the last decade did not steer urban expansion away from relatively scarce soils of good agricultural potential. Rather, a significant proportion of the expansion of the main cities in the last decade occurred on Classes I and II soils. Moreover, this proportion is greater outside of San Salvador reaching almost 50% in Santa Ana. This raises questions about the effectiveness of these mechanisms in all cities but in secondary cities in particular. It also raises similar concerns about the insufficiently strong incentives for vertical development. The gradual decline of the agricultural sector in El Salvador over the past two decades has increased the opportunity cost of leaving agricultural lands under cultivation, particularly in areas close to the country’s main cities. The high proportions of urban expansion into good agricultural lands in the last decade around the three main cities attest to this. The Government needs to consider whether the observed pattern of urban expansion at the expense of good agricultural land is an acceptable trend for future urban growth. Where the alternative would be to favor a leap-frogging, non-contiguous pattern of urban expansion, the associated higher per-capita costs of providing infrastructure services to new urban households would quite likely outweigh the benefits of preserving these lands for agricultural production. However, if this is not the case, the Government needs to find a way to direct the future expansion of its cities so that the inevitable expansion that will accompany the changing structure of the economy will infringe less upon good quality soils. 108 El Salvador Country Land Assessment World Bank 8.3.3 Low income settlement and concentration of the urban poor A large proportion of the supply of affordable shelter in the last decade has occurred in socio-economically precarious settlements, implying another qualitative deficit in the relatively strong urban land supply-side elasticity story. Almost half of AMSS’s and Santa Ana’s new urban land growth in the first decade of this century occurred in precarious settlements. For San Miguel and the municipalities of Colon and San Juan Opico, satellites of the AMSS, it was more than a third. In all these cases, a high proportion of this vulnerability was rated as high or extreme. These trends are a cause for public policy concern as they suggest that a significant part of El Salvador’s urban expansion is occurring with considerable socio -economic vulnerability including only partial infrastructure services and in many cases, irregular land tenure28. Spatial concentrations of the urban poor are also notorious for their associations with urban crime and violence, another major public policy concern in Central America. 8.3.4 More egalitarian land tenure structure and agricultural production The more egalitarian land tenure structure created by the Agrarian Reform has not resulted in enhanced agricultural productivity. Despite having more land available and being more likely to be landowners as well as enjoying similar access to machinery, inputs, and credit, once the size of their land holdings are taken into account, agrarian reform cooperatives have no better yields than other Commercial Enterprises. Yields in sugarcane, coffee, basic grains, fruits and vegetables are no better or worse in the Agrarian Reform Cooperatives when farm sizes are matched. Agrarian Reform cooperatives use land less intensively than non-Reform cooperatives. When land size is controlled, Agrarian Reform cooperatives on average cultivate 10% less in their holdings than other commercial enterprises. Moreover, if the universe of active Agrarian Reform cooperatives is in fact now only about half of what it originally was as suggested by the 2008 Agricultural Census, then the contrast in land use intensity between Agrarian Reform cooperatives and other commercial enterprises would be much more stark. This finding is of particular concern because it is the result of an explicit public policy instrument. The fact that public policy also severely restricts formal exit routes for transfer out of Agrarian Reform cooperativs (through supervised auctions and other means) further inhibits market-induced land use changes that would make for more efficient and potentially more sustainable land use. These findings suggest that three decades later, the public policy ambitions embodied in the Agrarian Reform, while successful in promoting broader access to land, has not led to a noticeable boost in agricultural production. This underscores the difficulty in using land policy instruments as effective mechanisms for reviving the agricultural sector today. The stagnation and revival of the agricultural sector should be the subject of a separate study. 28 The ILP estimates that a large majority of the precarious settlements identified by UNDP have irregular land tenure. 109 El Salvador Country Land Assessment World Bank 8.3.5 Fragmentation of agricultural parcels, agricultural production and rural poverty reduction The high degree of fragmentation of rural plots that exists today is inconsistent with public policy aspirations to eliminate rural poverty predominantly through agricultural income. Although land is significantly more accessible today than three decades ago, El Salvador’s poor have considerably less agricultural land than the non-poor. In the rural sector, the average size of a farm for the poor is less than one hectare. And unlike the non-poor, the average size of farms managed by the poor shows little variation across regions in 2005 and 2009. Bringing these farmers out of poverty based on agricultural incomes alone would require increasing their access to land several fold. Under the conditions of high population density and market trends experienced in El Salvador, this approach is not feasible. Non-farm income has to play a growing role in reducing poverty in rural areas. But again, this analysis should be the subject of an agricultural sector assessment that reviews all policies, not just those related to access to land and land use, that affect the profitability of agricultural activities. 8.3.6 Agricultural expansion and deforestation Land governance provisions did not protect forests and the environmental services that they provide when markets and other influences favored agricultural expansion . The average annual deforestation rate of 1.5% between 1998 and 2008 appeared to be mostly on account of agricultural expansion for annual crops and pasture. However, the precise causes of deforestation are difficult to identify. A main line of thought has been that the return to the countryside by certain populations after the Peace Accords in the early 1990s led to an agricultural expansion which resulted in increased deforestation. Another has been that active promotion of agricultural expansion over the years through free or subsidized distribution of seeds and fertilizers by the Ministry of Agriculture may have inadvertently encouraged deforestation. However, neither theory was substantiated by correlation analysis done at the municipal level. The net result has been that in a ten year period, 44,000 ha of forests were lost in a country where co-dependent ecosystems exist in a very finite geographic space. The arising tension even affects the choice of locations and crops that may be promoted through agricultural expansion drives some of which exert considerable strain on limited water resources. Meanwhile with deforestation, settlements, infrastructure assets and crops became more vulnerable to landslides as evidenced by the very costly damage inflicted by recent tropical storms. 8.3.7 Expanding land access and formal security of tenure The perceived high transaction costs of formal land conversion and subdivision created a barrier to efficient, market-based responses to the changing nature of demand for land during the civil war. The Institute of Liberty and Progress (ILP) estimates that there are about 300,000 plots that are the result of informal subdivisions where the lack of formal registration means that residents lack formally recognized documentation of their property rights. In many cases, these land use conversions were a response to changing circumstances whereby during the civil war migration to urban areas increased and many farms were subdivided as they went out of 110 El Salvador Country Land Assessment World Bank production. This in turn limits the ability of the current residents to transfer these lands in formal markets as well as to leverage their real-estate assets to raise capital including mortgage finance. By contrast, departments of the country that experienced a rationalization of property rights and land administration systems, were seen to experience growth in almost all kinds of land market transactions including transfers, mortgages, bequests and donations. Much of the agrarian reform land also cannot be efficiently transferred to alternative uses due to land tenure and registration irregularities. The legal framework prevented the registration of titles under the beneficiaries’ name until the debt was fully repaid. Similarly, cooperatives could get their provisional title only after their period of co-management was over with a 30-year repayment period starting then. Also the Agrarian Reform adjudications were kept separate from regular registration procedures which created a series of anomalies. Finally, the transfer of land under the modality of joint titles (proindivisos) further complicated matters for beneficiaries under Phase III and the PTT and for members of the cooperatives who want to obtain title to their individual plots. 8.4 Institutions for mediating public policy conflicts and promoting land use efficiency 8.4.1 Local Land Use Planning and Development Control Land Use Zoning and Development Control are instruments with the potential for reconciling the land use imperatives associated with different development and sectoral agendas however they are insufficiently deployed in El Salvador. For example, these instruments can reconcile the need for urban settlement expansion with the importance of preserving the environmental functions of forested land, the maintenance of rare high potential soils in agriculture and the minimization of natural disaster risks. Although Master Plans and local plans do exist, there is little evidence in El Salvador of these being the result of a wide and substantive participatory process engaging the perspectives of diverse stakeholders. The consequences are two-fold. Firstly, not all pertinent information is factored into designating particular spatial zones for specific land uses. And secondly, buy-in for the formal plan is often lacking among key stakeholders whose participation is necessary for successful Plan implementation. The Government’s recent CAPRA (Central American Probabilistic Risk Assessment) initiative in conjunction with the Global Facility for Disaster Reduction and Recovery (GFDRR) is a step in the right direction when it comes to reconciliation of land use with natural disaster risk. Another limitation is that there is little evidence of land use planning and development control being dynamic and well-resourced processes outside of the AMSS. Human resource capacity for these activities at the municipal level and in secondary cities and towns as well as reliance on dynamic information systems such as remote sensing and land market pricing datasets is very low. The National Land Use and Development Plan (PNODT) is an initial attempt at using integrated land use planning as an instrument of mediating public policy conflicts, however, as noted earlier, there is insufficient coordination between central and lcoal government planning. 111 El Salvador Country Land Assessment World Bank 8.4.2. Land Tenure Regularization Initiatives Informality offers entry into a system at different price points29 and various affordability indicators suggest that significant segments of El Salvador’s urban population are gaining access to urban spaces by entering at price points below those on offer in the formal market. Given that several key regulatory measures for land development in El Salvador are realistically prescribed, low incomes may continue to be the binding constraint sustaining informal markets for some time. The efficiency of the regularization process30 that follows, if it follows, will be a key determinant of how long these vulnerable segments of the population remain in situations of compromised welfare. Land tenure regularization is an ex-post mechanism for reconciling land use conflicts where informal occupation or use is later deemed suitable to continue but progress has been slow. El Salvador has a large regularization challenge and a multiplicity of institutional arrangements that are only gradually and partly addressing it. For example, ISTA is yet to legalize individual titles to over 170,000 landholdings that were part of the agrarian reform. On- going efforts are hampered by the lack of reliable and updated property records and inadequate coordination between itself and CNR. In particular, the CNR can only register documents that meet the legal requirements to maintain legal security. ILP is mediating the regularization of settlements on informal sub-divisions in which there are approximately 300,000 plots. This is done in partnership with OPAMSS in AMSS but is less straightforward and subject to varying processes and standards outside of AMSS. The proposed Bill for Land Tenure Regularization aims to address these challenges in a more systematic way. 8.4.3 Land Information Systems and Governance In democratic societies such as El Salvador, the ultimate avenue for informed decision making and for balancing competing interests over the use of land is through the empowerment of the government and the citizenry with reliable information about the use, value and rights attached to land. If citizens and non-governmental institutions have ready access to such information, government authorities at all levels and private sector interests who seek to influence those authorities, will likely be held to a higher standard of accountability. The consultative processes and debates that ensues from such empowerment would serve as a threshing-floor for competing developmental plans for specific locations much as ‘local land use planning hearings’ do in the United States. El Salvador has started the process of improving the reliability, accessibility and comprehensiveness of land information but has some distance to go. Over the past two decades, the country has modernized many of its key land administration agencies and the CNR which merged the legal registry with the geographical description of land parcels (urban and 29 This conceptualization of informality was discussed in February 2012 at a seminar on ‘The Political Economy of Green Growth’ at the World Bank, Washington D.C. 30 The most significant bottlenecks in regularizing urban parcels include mortgages which have not been cancelled in the Registry, unregistered subdivisions or amendments to existing parcels, and inaccurate liens on properties. 112 El Salvador Country Land Assessment World Bank rural) now enjoys significant approval from the public for the efficiency and transparency of its services. More than two-thirds of the country’s land parcels have been systematically surveyed. Still, as this Study found, the accessibility of land information in an integrated platform to support decision making and public inquiry is limited. Many of the adverse developmental outcomes observed in this Study can be attributed in part to inadequate reconciliation and analysis of existing data sources and weak inter-Agency coordination within the Government. A proposal of the Ministry of Environment and Natural Resources to create and maintain an integrated land use platform with remotely sourced information as well as field datais a step towards filling this gap. However, further steps are needed to incorporate data from public policy decisions (such as development approvals or refusals, land allocations, regulatory exemptions) and market-sourced land value data from real estate agencies and banks. 8.4.4 Property Taxation Property taxation is an instrument capable of incentivizing more efficient use of land but remains non-existent in El Salvador. With land values low in relation to incomes and land uses and land vacancy not taxed, there is little incentive to maximize the return from land. . Urban municipalities lacking this source of revenue also remain under-resourced to finance or co-fund infrastructure and services expansion to accommodate population growth. 8.4.5 The new Law for Territorial Planning and Development The new Law of Territorial Planning and Development provides a common framework for land use across the country’s territory but its implementation will be challenging. It is viewed as a mechanism for managing the competing demands of various sectors across the national terrain. It may also potentially serve as a check on the current tendency of individual municipalities to adopt too narrow and too short-term a development perspective when vetting development applications in their jurisdiction. This potential may induce resistance from some municipalities on the grounds that it counters the broader constitutional and public policy provisions for decentralization and subsidiarity. At the most basic level, there are several key outstanding actions that affect implementation of this law. The Ministry of Finance has not yet allocated the resources for its implementation. The National Council has not yet been established, nor has the Presidency designated its representatives to the Council. Moreover, the national and department-level plans mandated by the law have not been prepared. The implementation will require a skillful balancing of central government cross-sectoral and cross-departmental perspectives with the logical delegation of implementation to inadequately-resourced local level municipalities. It is unrealistic to expect more than 260 municipalities to each have adequate human resource and information systems capacity for rational, evidence-based implementation. So some institutional formulation will be required to allow contiguous municipalities to pool such resources, perhaps at the Departmental level. This institutional arrangement would probably not directly mirror the model of the umbrella planning agency of OPAMSS utilized for the AMSS as there are no large contiguous built-up areas that span multiple municipalities outside of the AMSS. However, there are some contiguous protected areas and ecosystems that require a coordinated approach to sustainable management. 113 El Salvador Country Land Assessment World Bank Chapter 9 Policy Options 9.1 Key Findings This Study presents an analysis of land in El Salvador from a broad development perspective. It assesses the alignment of land use and land market trends with the developmental challenges faced by the country. Specifically, the Study gauges the effectiveness of land sector institutions and policies in facilitating developmental outcomes in agricultural development, urbanization, and environmental management. The Study notes that the evolution in access and use of land resources, urban and rural, is the result of a combination of policy-induced changes, shifting demographics, and market forces. The major conclusions from the Study are related to (i) the access and affordability of land both urban and rural, (ii) the formalization of land rights, (iii) the rural to urban transition in land use and related dynamics of urbanization, and (iv) the consequences of these changes on the rates of deforestation in the country. Public policy in El Salvador has been fairly effective over the past three decades in meeting some key development challenges. In particular, the Study finds that access to and affordability of land, rural and urban, has significantly improved. Similarly, land use zoning and development control mechanisms have worked fairly well in mediating competing demands for land in peri- urban areas in the metropolitan area of San Salvador. And finally, the country has advanced considerably in the systematic surveying (but less so in titling) of land parcels, urban and rural, in several regions of the country which has resulted in increased overall tenure security. On the other hand, the country’s policies and institutions have not been as effective in dealing with certain critical issues. The Study finds that there continue to be a significant backlog of unresolved titling issues of land parcels affected by the various phases of the agrarian reform, some dating back several decades, which limits the scope for market transfers towards more efficient and sustainable land uses. Also, the Study’s spatial analysis shows that some urban settlements have taken place in areas which are vulnerable to suffering or exacerbating natural disaster damage, or which were at the expense of scarce good-quality agricultural lands. And finally, it is noted that although the rate of deforestation has slowed down in the last few years, in the late 1990s and early 2000s, the country lost about 44,000 ha of forests due to agricultural expansion. 9.2 Policy Options The Study presents a series of policy options for consideration by the Government of El Salvador and the country’s civil society organizations in order to address the key challenges related to land tenure, land use, and territorial planning. The analyses of land use and land market dynamics presented in this Study while useful as a base for decision making is in and of itself insufficient for generating specific recommendations. Each of the options presented in Table 9.1 below should be subject to technical feasibility, cost-benefit, institutional, and political economy analyses, as well as public debate before they can be effectively converted 114 El Salvador Country Land Assessment World Bank into specific recommendations. For each subject area, the table presents policy options to address the concern. It also identifies likely trade-offs and possible challenges to implementation. The options presented here should be considered individually and in conjunction with other possible measures. They are not to be interpreted as alternatives, but rather as thematic public policy areas subject to further analysis and consideration. The matrix of policy options encompasses several policy instruments, including national legal/regulatory changes, taxation, municipal level planning measures, and increased civil society and private sector engagement. Some policy options may require legislative action. For example, there are already ongoing discussions to pass legislation to allow the titling of current landholders on lands which were titled to absentee agrarian reform beneficiaries, or to streamline the procedures for regularizing land rights of beneficiaries with joint titles (proindiviso), or to address the issues of informal plots in urban areas. All these initiatives deserve careful consideration and debate by their key stakeholders. Land taxation also deserves careful consideration. Although in theory land taxes are among the least distortionary, in practice they are difficult to implement due to various factors. To be effective, tax rates have to be high enough to change behavior (e.g. from vacant to more intensive land uses, or from less to more valuable activities) and justify the administrative costs of collecting it. Unless the local benefits of the investments financed by land taxes are clearly visible to taxpayers and are linked to increased land values, strong opposition (or evasion) is to be expected. Also, land taxes require accurate and timely information on ownership and use. Nevertheless, land taxation can be a powerful policy instrument which El Salvador has historically underutilized, and now deserves careful consideration, particularly in the context of increased decentralization. There are several instruments for urban planning and development which can be considered to address the rapid urbanization challenges faced by El Salvador. These include raising the permitted ratio of total built space to land area; increasing development control policing of zones of “maximum protection” and introducing stiffer penalties (substantial fines and or demolition for new construction) in such zones; expediting the regularization of land titles to vacant land within the existing city perimeters; and introducing a “vacant” land tax to discourage speculative withholding of well-located urban lands from the market. Rapidly growing secondary cities need to strengthen their municipal planning capacity. This can be achieved through training and deployment of technical staff, as well as investments in land use and land market monitoring systems. One option to consider is the use of a “Guided Land Development” approach for the expansion of secondary cities. This would entail further analysis of land market and land use trends to anticipate the future growth of the cities and then proactively setting aside rights of way for their systematic and more sustainable expansion. A second option is “Land Readjustment/Pooling” for acquiring ‘rights of way’ from private land owners without placing a financial burden on municipalities. Eminent Domain would be another policy option for acquiring such ‘rights of way’ but this tends to be costlier to government and litigation prone. Increased engagement of civil society organizations and the private sector in planning and service provision must be considered as a key element of any package of policy measures. Organized rural community and urban neighborhood associations can be low-cost and effective 115 El Salvador Country Land Assessment World Bank mechanisms to gather and update land tenure information, resolve boundary disputes, prioritize investments, and monitor compliance with regulations. Similarly, engagement of private sector agents in facilitating land market transactions (rural and urban) is critical. Careful consideration must be given to certification programs (e.g., real estate agents, surveyors, etc.), differentiated tax incentives for land development and subdivisions (e.g., to move away from vulnerable areas or to encourage mixed-income area developments), and for financing of infrastructure. Table 9.1: Matrix of Policy Options Public Policy Policy Option Tradeoffs Possible Challenges Concern to Implementation Backlog of untitled Adopt temporary measures for May encourage Requires careful analysis agrarian reform lands systematic regularization of emergence of old of legal and regulatory lands by area, rather than case disputes over rights options by case Limitations on Streamline the procedures for Some rural lands may Elimination of restrictions transferability of agrarian regularizing land rights exit agriculture may require legislative reform lands and those of altogether action residents of informal urban Remove obstacles to voluntary subdivisions direct transfers from Land may again cooperatives to individuals become concentrated holders, and from absentee under a few producers owners to current holders Provision of technical assistance to agrarian reform beneficiaries based on new technologies and market opportunities to facilitate better informed decision making about selling or retention of agrarian reform lands Encroachment of urban Direct future urban expansion Possibly higher Costs of enforcement expansion onto good away from areas with high infrastructure costs quality agricultural land potential soils Some high quality areas And slower are already partially commercial developed development Increased commuter times Insufficient land use Create a shared pool of Loss of some Inter-municipal planning and development qualified technical personnel municipal autonomy competition for resources control capacity in and land information analysis and talent secondary cities and capability for contiguous satellite municipalities municipalities Not clear to which existing tier of government institutions such a shared resource can be attached Most contiguous spatial 116 El Salvador Country Land Assessment World Bank development is not inter- institutional Strengthen land information If undue emphasis is Capacity to use modern systems for decision making placed on highest systems will require including reliance on land accuracy data, it may strengthening market and spatial trends as delay crucial well as disaster risk mapping investments in to guide urban planning and infrastructure and prioritize infrastructure other public goods investments Decentralization of Introduce a property taxation Reduced financial Land taxes are unpopular responsibilities for land system which returns all or control from the development / service most collections to central government Land values are not expansion to weaker municipalities systematically recorded municipalities Administrative costs can be unjustifiably high Rapid horizontal Raise the permitted ratio of Increased urban Costs of enhancing expansion of cities often total built space to land area congestion, traffic, infrastructure and services onto steep slopes and unless adequately capacity to support higher agriculturally viable land addressed density population Cultural aversion to high rise living including residual fears from past earthquakes Increase enforcement in Restricts the High costs of enforcement, maximum protection zones development space, particularly for the poor, possibly driving land if affordable alternatives prices upward are not available Expedite regularization of land Discourages vertical Ownership status of lands titles to vacant land within city growth may be in conflict perimeters Introduce a vacant land tax in May encourage Administrative cost may urban areas cosmetic development be high to avoid tax Spatial concentration of Introduce incentives to May discourage some Requires broad and low-income housing promote mixed-income sub- commercial genuine political divisions such as tax breaks; development commitment for effective market guarantees; implementation contributions of public land Encourage greater participation May distract attention Requires sustained by civil associations to lobby away from increased encouragement from for improved services integration of central and municipal neighborhoods governments Multiple land use Use “Guided Land May increase costs of Limited capacities for conflicts associated with Development”31 to provide a development and consultation, unguided spatial expansion framework for more systematic infrastructure implementation and of secondary cities and sustainable expansion enforcement 31 Guided Land Development is a policy that involves the identification and setting aside of rights of way around the periphery of a city to create a basis for more systematic urban expansion. More Details can be found in Memo to the Mayor, World Bank 2011. 117 El Salvador Country Land Assessment World Bank Use “Land Readjustment”32 to Land prices within Legislative framework for provide a framework for more land readjustment this tool may require systematic and sustainable zones may rise strengthening expansion although they may decrease in the broader area due to more elastic supply 32 Land Readjustment or Land Pooling is a policy tool whereby land owners voluntarily contribute part of their land holdings to accommodate the rights of way for servicing their land while often also creating surplus vacant plots. The landowners receive a smaller but more valuable, better serviced plot and government’s land acquisition costs are significantly reduced. More Details can be found in Memo to the Mayor, World Bank 2011. 118 El Salvador Country Land Assessment World Bank Bibliography Anriquez, Gustavo, and Genny Bonomi. 2007. “Long-Term Farming and Rural Demographic Trends,” Background paper for the World Development Report 2008. Baumeister, Eduardo. 1999. “Las Iniciativas Campesinas y la Sostenibilidad de los Resultados de la Reforma Agraria en El Salvador, Nicaragua y Honduras.” United Nations Research Institute for Social Development Discussion Paper No. 105. Deininger, Klaus and Pedro Olinto. 1999. “Asset Distribution, Inequality, and Growth.” Washington, DC: World Bank Policy Research Working Paper Series 2375. Direccion General de Estadistica y Censos. 1971. Tercer Censo Nacional Agropecuario. San Salvador: Ministerio de Economia. Direccion General de Estadistica y Censos. 2005. Encuesta de Hogares de Propositos Multiples 2005. San Salvador: Ministerio de Economia. Direccion General de Estadistica y Censos. 2008. Cuarto Censo Nacional Agropecuario. San Salvador. San Salvador: Ministerio de Economia. Direccion General de Estadistica y Censos. 2009. Encuesta de Hogares de Propositos Multiples 2009. San Salvador: Ministerio de Economia. World Bank. 1997. “El Salvador Rural Development Study.” Washington, DC: World Bank Report No. 16253-ES. World Bank. 2011. World Development Indicators 2011. Washington, DC: World Bank. World Bank, 2005. The Dynamics of Global Urban Expansion. Washington. D.C: World Bank. 119 El Salvador Country Land Assessment World Bank Annex 1: Discrepancies in Estimates of Land Under Agriculture Comparison between the 1971 and 2008 agricultural census data show that the amount of cultivated land increased between these two years. However, the amount of total land held under farms, referred from now on as agricultural land, decreased by 36 percent. Have these lands been transferred out of agriculture? Other sources of data to estimate the size of agricultural land suggest that the agricultural sector is larger than the self-reported figure in the 2008 Agricultural Census. Figure 1 shows the surface area of agricultural land according to the estimates in the World Development Indicators. This graph shows that the surface area under agricultural use has been increasing since the 1960s.The estimated area for 2008 was 1,552,000 hectares, which is 624,000 more than the self-reported figure in the 2008 census. Another source of data to estimate the size of agricultural land is an ecosystem map prepared in 2008 based on the classified supervision of Landsat satellite images. This map shows that there were 1,713,709 hectares of land under agricultural use in 2008. Figure 1: Agricultural Land Source: World Development Indicators (2011). There is large discrepancy between the self-reported figure of agricultural land in the 2008 Agricultural Census and the other sources of data for agricultural land. Has this land really gone out of the agricultural sector or is it still available to expand agriculture? A significant amount of agricultural land was left fallow as a result of the 12-year civil conflict. Since most of the emigrants did not return, it is possible that a significant share of that land has not been restored back to agricultural use. However, considering El Salvador’s high population density, it is possible that much of this land has already been urbanized in a low-intensity pattern that is not well-captured by supervised classification of satellite imagery. In order to explore the question of whether this could be land still available for agricultural production, we compare the gap between the calculation of agricultural land in the ecosystem map and the self-reported figure in the census with data on soil quality. Table 1 shows the estimates of agricultural land by data source and the discrepancy. At the national level, there is 120 El Salvador Country Land Assessment World Bank an estimated gap of 50 percent of agricultural land that could arguably be used to expand agriculture. At the department level, Chalatenango, Cuscatlan, and Morazan have the largest gaps (66, 64 and 61 percent respectively). For most departments the gap between potential agricultural land and land under production is between 40 and 50 percent. There are 19 municipalities for which the amount of land reported in the 2008 Agricultural Census is actually larger than the estimated area for agricultural land use in the ecosystem map. These observations were not included in the calculation of the gap in agricultural land. Table 1: Estimates of Agricultural Land and Potential Gap Number of 2008 2008 Gap in % of Municipalities Agricultural Agricultural Agricultural Ecosystem Census (ha) Land (ha) Land Map (ha) Underutilized (a) (b) (c) (c/a) National 252 1,669,491 841,320 828,171 50% Ahuachapan 12 101710.9 56996.27 44,715 44% Cabañas 9 98253.82 50487.25 47,767 49% Chalatenango 33 157620.6 54141.84 103,479 66% Cuscatlan 16 60025.52 21514.15 38,511 64% La Libertad 22 135057.1 70225.79 64,831 48% La Paz 22 104898.8 53711.52 51,187 49% La Union 18 156218.9 89903.28 66,316 42% Morazan 26 121688.8 48064.15 73,625 61% San Miguel 20 173433.4 79050.19 94,383 54% San Salvador 19 52287.16 29206.42 23,081 44% San Vicente 13 90830.25 46791.01 44,039 48% Santa Ana 13 176035.2 96876.67 79,159 45% Sonsonate 16 99611.93 68728.96 30,883 31% Usulutan 23 141818.8 75622.98 66,196 47% 121 El Salvador Country Land Assessment World Bank Information on soil quality is used to examine the correlation that exists between the size of the gap in agricultural land and the agrological characteristics of the municipalities. Data on soil quality is gathered from an agrological map prepared by MAG in 1997. Soils are classified between Class I and Class VIII according to the United States Department of Agriculture Classification system. Soils under Class I and Class II have from slight to moderate restrictions for intensive cultivation; Class III and Class IV have from severe to very severe limitations for cultivation; Class V and Class VI are impractical for cultivation; and Class VII and Class VIII soils are restricted to forest, wildlife, or in some cases grazing. Soils under Class I and Class II are classified as good soil for agricultural practices; soils between Class III and Class VI are classified as poor soils for agricultural practices, and soils under Class VII and Class VIII are classified as soils for forestry use. Table 2 shows the distribution of soil quality by department. For the country, about 6.5 percent of land is classified as good soil, 42 percent as poor soil, and 50 percent as forestry use. For the departments of Cabanas, Chalatenango, La union, and Morazan, the percent of surface area under forestry use is larger than 70 percent. The department with the highest percentage of land classified as good soil is La Paz with 22 percent, followed by Usulutan and San Salvador with 15 and 11 percent respectively. Table 2: Agricultural Land Underutilized and Soil Quality Agricultural Good Poor Soil Forest Land Soil (% (% of (% of Underutilized of Land) Land) Land) (% of Land) National 0.44 0.07 0.42 0.50 Ahuachapan 0.44 0.06 0.52 0.42 Cabañas 0.49 0.00 0.21 0.79 Chalatenango 0.66 0.01 0.27 0.71 Cuscatlan 0.64 0.03 0.58 0.38 La Libertad 0.39 0.07 0.47 0.46 La Paz 0.49 0.22 0.55 0.22 La Union 0.42 0.00 0.28 0.71 122 El Salvador Country Land Assessment World Bank Morazan 0.61 0.00 0.13 0.87 San Miguel 0.54 0.04 0.47 0.47 San Salvador -0.11 0.11 0.52 0.29 San Vicente 0.47 0.09 0.48 0.43 Santa Ana 0.45 0.04 0.34 0.61 Sonsonate 0.24 0.09 0.61 0.29 Usulutan 0.46 0.15 0.49 0.35 A correlation analysis is performed to examine the relationship between the estimated gap in agricultural land and soil quality. The size of the gap in agricultural land is negatively correlated with the percentage of land classified as either good or poor soil and positively correlated with the percentage of land classified as forestry use. Table 3: Correlation between Gap in Agricultural Land and Soil Quality Gap in Agricultural Land Corr. P-Value % under Good Soil -0.1997* 0.001 % under Poor Soil -0.2135* 0.000 % under Forest 0.3385* 0.000 These correlations suggest that most of the land that is not reported in the agricultural census could be land that is not suitable for intensive agricultural production. In 1971, some of these areas could have been treated as fallow land in large farms. With higher population density, however, it seems that these lands may indeed have been transferring out of the agricultural sector. 123 El Salvador Country Land Assessment World Bank Annex 2: Details of Urban Expansion Analyses San Salvador In the first decade of this century, the AMSS ’s urban footprint (including the spatially contiguous city of Soyapango) grew by approximately 30% from about 115 square km to over 148 square km. Six Areas of Interest (AOIs) where this growth was concentrated were selected for further study through visual inspection of high resolution before and after Google Earth images. These AOIs are shown in the Figure 1 and numbered from 1 to 5 in anticlockwise fashion starting with the most northerly AOI. Figure 1: Urban Growth in San Salvador Metropolitan Area (AMSS) 2001-2010 124 El Salvador Country Land Assessment World Bank Table 1: Summary of Patterns of Observed Land Use Changes in San Salvador Location Summary of Land Use Changes AOI 1 AOI #1 is at the northern edge of the city, just south of the ….. In 2003, hardly any structures/buildings existed. By 2010, a significant number of buildings appeared on the landscape. AOI 2 No drastic change in this AOI (includes the northern suburb of Apopa)- urban area generally became denser. AOI 3 Not easy to pick up differences- earliest available image is only 5 years ago. Because the change detection picked up a lot of growth, it is likely that a lot of these subdivisions/ houses appeared in the early 2000’s. AOI 4 AOI #4 is further along the outskirts of San Salvador- a lot more visible growth in the later part of the decade (post-2005) than with AOI #3, which is closer to the city center. A lot of this development expanded into forests. Growth is clearly formal. AOI 5 AOI #5 is at the extreme southwestern edge of the city- including the town of Nuevo Cuscatlan. Growth here has been formal and mostly new subdivisons of housing. Sample close-ups show well-planned and laid out housing developments, mostly replacing forested areas. AOI6 AOI #6 is at the southern edge of the city. Again, growth here is formal and well planned-out. The imagery in 2002 shows the first stages of development (inside the red boxes), so we can’t be certain what these new buildings are replacing- but it was most likely forests, judging by the surrounding area. 125 El Salvador Country Land Assessment World Bank Figure 2: Sample close-ups (in AOI #5 which is at the extreme southwestern edge of the city- including the town of Nuevo Cuscatlan) show well-planned and laid out housing developments, mostly replacing forested areas. 126 El Salvador Country Land Assessment World Bank Santa Ana Figure 3: Urban Expansion of Santa Ana between 2001 and 2010 In the first decade of this century, Santa Anna’s urban footprint grew very substantially by approximately 120% from approximately 17 square km to over 37 square km. Five Areas of Interest (AOIs) where this growth was concentrated were selected for further study through visual inspection of high resolution before and after Google Earth images. These AOIs are shown in the Figure 3 and numbered from 1 to 5 in anticlockwise fashion starting with the AOI in the North-Western corner. Table 2: Summary of Patterns of Observed Land Use Changes in Santa Ana Location Summary of Land Use Changes AOI 1 AOI #1 is a northwestern suburb of Santa Ana- the town of Ranchador. Growth here appears formal. AOI 2 Cloudy images in Google earth- can’t go back any earlier than December 2004. Significant growth in the center of the AOI- looks like it replaced fields. Some areas at the southeastern corner replaced forests. Many new houses/buildings popping up. 127 El Salvador Country Land Assessment World Bank Seems to be replacing mostly forested area. Forests also being cleared for agriculture in some places. AOI 3 Significant growth in the northwest corner of the AOI. Looks like housing grew into land that had previous been forests. A few areas were agriculture. Some of the expansion into forests is non-contiguous with other built development following a more inefficient leapfrogging pattern. While some of the buildings that appear between 2003 and 2010 follow established roads, a lot of them in some parts of the AOI appear haphazard and are possibly informal. AOI 4 AOI #4 is at the far southeastern edge of the city. Growth appears formal as the city expands outward. Housing/buildings appear to become denser in this sample area from AOI #4. AOI 5 AOI #5 is at the eastern edge of Santa Ana. As with the other AOI’s, growth here appears formal and follows the same pattern of expansion. The built-up landscape becomes denser. Figure 4: Santa Ana: AOI 3: Encroachment of Forests and Leapfrogged Development Pattern April 2003 February 2010 128 El Salvador Country Land Assessment World Bank San Miguel In the first decade of this century, San Miguel’s urban footprint grew by approximately 60% from approximately 24 square km to over 38 square km. Figure 5 shows the spatial distribution of this growth. Five Areas of Interest (AOIs) where this growth was concentrated were selected for further study through visual inspection of high resolution before and after Google Earth images. These AOIs are shown in the Figure 5 and numbered from 1 to 5 in anticlockwise fashion starting with the most northerly AOI. Figure 5: Urban Expansion of San Miguel between 200\1 and 2010 129 El Salvador Country Land Assessment World Bank Table 3: Summary of Patterns of Observed Land Use Changes in San Miguel Location Summary of Land Use Changes AOI 1 Significant, planned (formal) growth in the western half of the AOI. The image from 2004 clearly shows agricultural land (fields) in place before the development occurred. This development is clearly formal- the infrastructure was in place before the development grew. AOI 2 Not a lot of visible changes, mostly filling in along the road that cuts through the center of the AOI. Most changes have come in the form of individual dwellings that dot the landscape. Generally looks to be along established roads, so likely to be formal development. AOI 3 Again, most of the growth in this area seems to be houses/dwellings becoming more numerous (dense). The new construction follows the established roads, so the infrastructure came first. The land being replaced with development appears to be scrub/bush. Nothing heavily forested is disappearing, but it doesn’t appear to be farmland either. AOI 4 Again, not a lot of dramatic changes- simply areas becoming more dense and new buildings dotting the landscape. AOI 5 Not very many visible changes, other than a few additional buildings that appear. Figure 6: San Miguel: AOI 1 March 2004 November 2009 130 El Salvador Country Land Assessment World Bank San Miguel has experienced growth along its outer edges. Mostly, it looks like houses that have replaced agriculture area. Doesn’t look like a lot of forests have been replaced. One thing to keep in mind- March of 2004 is the earliest Google Earth image for this area. Since the change detection from 2001-2010 picked up significant urban growth in these 5 AOIs, which in some cases isn’t apparent by the Google Earth imagery, it’s possible that a lot of the growth occurred between 2001 and 2004. 131 El Salvador Country Land Assessment World Bank Annex 3: Methodological Details for Deforestation Analysis From 1998 to 2001, the World Bank financed the production of the Ecosystems Map of Central America, with individual mapping teams of national scientists for each country. In El Salvador, this resulted in the document “Mapeo de la Vegetación Natural de los Ecosistemas Terrestres y Acuáticos de Centro América, Capítulo El Salvador” with the associated GIS file. The El Salvador GIS component was primarily based on 1998 Landsat images. In 2011 the Ministry of Environment contracted the study “Estudio de Racionalización y Priorización del Sistema de Áreas Naturales Protegidas de la República de El Salvador” to the World Institute for conservation and Environment (WICE). This study included the updating of the ecosystems map that was produced under aforementioned regional mapping project. The map update was prepared on the basis of satellite Aster Scenes of 2008 and Google Earth. Satellite images habitually used for ecosystems mapping purposes typically have a resolution of 8 to 30 m2, and taken simultaneously in different colour (spectral) combinations. This allows for rather sophisticated reconstructions of what the colours represent, but in reality, they don't visualize individual structures, like trees and shrubs. Google Earth, however, does visualize such structures and given its current availability, a much more detailed map could be made than the one carried out by the scientists at the turn of the century. As aforementioned update was written and made available in 2011, its date refers to the year of its digital publication, but the data are from 2008. Since 2011, the World Bank has been working on the “El Salvador Land Sector Assessment” study to assess the alignment of the current state of the land sector in El Salvador with the developmental challenges faced by the country. The assessment covers a broad range of issues affecting land use and land markets in both urban and rural areas. Part of the methodology includes temporal analysis of land use changes through change detection of spatial images. The World Bank had planned on using the ecosystems map update to carry out such temporal analysis on forest cover. However, the new map had been been made with a much better distinction between coffee plantations from (semi-) natural forests, and polygons could be recognized and delineated that were not found previously. As a result, the “2011 map”, although called an update, in reality is a completely new map, with more and different classes and many more finely digitized polygons than mapped in 2001. Moreover, the 2000 map was not produced for change detection purposes and was based on limited access to imagery at the time, while Google Earth was not yet available. 132 El Salvador Country Land Assessment World Bank Given the great level of detail carried out in 2011, and the different approach in 2000 with less detail and different classes, it was not possible to compare the original (2001) map with the recent one. Therefore, the World Bank hired the re-digitization of the ecosystems map for the 1998 situation from a variety of images. In an effort to provide an even broader use of the work performed for environmental purposes, WICE also carried out a change detection for the period 2008 – 2011, using its own financial resources. All the information was combined in one document in Spanish, with an extensive summary for the World Bank and other English speaking users. Table 1: Landsat scenes used for the 1998 re-digitization and 2012 ecosystems map update for their respective years and month taken. Landsat scenes used for the digitization of the original 2001 ecosystems map: Path 18; Row 50: Months: March 1994 & Dec. 1998 Path 18; Row 51: Month: Dec. 1998 Path 19; Row 50: Months: May 1992 & Dec. 1998 Path 19; Row 51: Month: Jan 1998 Landsat scenes used for the re-digitization based on 1998 scenes Path 18; Row 50: Months: Feb, Mar, May, Oct, Dec Path 18; Row 51: Months: Jan, April, May, Jul, Aug, Dec Path 19; Row 50: Months: Feb, Apr, May, Jun, Jul, Nov Path 19; Row 51: Months: Jan, March, Apr, May, Jul, Oct Landsat scenes used for digitizing the most current situation with s 201133 scenes Path 18; Row 50: Month: March 33 This was the selection of the most suitable available images in December 2011 during the execution of this study from the Global Land Cover Facility, of the University of Maryland, http://glcf.umiacs.umd.edu/data/ 133 El Salvador Country Land Assessment World Bank Path 18; Row 51: Month: May Path 19; Row 50: Month: April Path 19; Row 51: Month: April Just as for the 2001 map, the digitization of the 2011 and 2012 updates and the re- digitizing of the 1998 situation were done visually on screen, which is the most detailed procedure for ecosystems mapping. For the 2011 update Aster satellite images were used provided by CATHALAC. Figure 1: 2008 Aster scenes used for the 2011 update with their respective month taken. A change detection of natural forest cover to other forms of land use was conducted between the situation analysed during the update of the Map of Ecosystems in 2011 that has been produced on the basis of satellite images of respectively the 1998 and 2008 situations. We also performed a change detection between the situations 2008 - 2011 based on the latest available images, as well as for the entire 1998 – 2011 period. Because Google Earth is now available, it could also be used as reference for the analysis of the 1998 situation. For each scene, we acquired the five best available images in terms of cloud-free status during both the dry and rainy seasons. These were used to digitize an ecosystem map with the same ecosystem classes and the same level of detail as in the 2011 update. 134 El Salvador Country Land Assessment World Bank For the 2011 update, ground verification was done during 2 periods, first during the dry season in February 2010 and later in July - August 2010 during the wet season. In July – August a team of biologists travelled to a variety of locations to carry out “rapid ecological assessments”, which then were also used for further ground verification. Moreover, the biologists involved in the “Estudio de Racionalización y Priorización del Sistema de Áreas Naturales Protegidas de la República de El Salvador” are very familiar with the natural ecosystems of El Salvador, which is one of the reasons why such high level of detail could be achieved. For the redigitization of the 1998 situation, it turned out that a comparison between the 2008 and 1998 images greatly helped identifying the changes in the polygons. Knowing what existed in 2008, greatly helped to identify the 1998 ecosystems and adjust the 2008 polygons to the situation of the images of a decade earlier. For the 2011 situation, only one set of images was acquired, since Google Earth could be used for this most recent situation and a new map was digitized accordingly. As the three maps have all been digitized in the same fashion and share the same precise classification, the polygons of each year could be compared and changes could be calculated. 135 El Salvador Country Land Assessment World Bank Annex 4: Details of Eco-System Change detection in hectares for the periods 1998 – 2008 and 2008 – 2011 136 El Salvador Country Land Assessment World Bank Ecosystem Ecosystem ID Area in hectares Loss in hectares 1998 2008 2011 1998 - 2008 2008 - 1998 – 2011 2011 Tropical evergreen seasonal broad-leaved upper-montane forest, secondary IA2d(1) 2,331 2,331 2,331 0 0 or intervened 0 Tropical evergreen seasonal needle-leaved upper-montane forest, IA2d(2) 3,178 2,984 2,984 -195 0 secondary or intervened -195 Tropical evergreen seasonal needle-leaved lowland forest, secondary or IA2d(2)(a) 285 285 285 0 0 intervened 0 Tropical evergreen seasonal broad-leaved forest, secondary or intervened IA2e(1) 1,019 1,009 1,009 -10 0 -10 Tropical evergreen seasonal broad-leaved alluvial forest, occasionally IA2f(2)(a) 12,982 10,264 9,562 -2,718 -702 inundated, secondary or intervened -3,420 Tropical semi-deciduous broad-leaved well-drained lowland forest, IA3a(1)(a) 41,735 37,160 37,160 -4,575 0 secondary or intervened -4,575 Tropical semi-deciduous broad-leaved submontane forest, secondary or IA3b(1) 8,962 8,996 8,996 34 0 intervened 34 Tropical semi-deciduous mixed submontane forest, secondary or IA3b(1/2) 31,073 28,113 28,113 -2,960 0 intervened -2,960 Tropical semi-deciduous broad-leaved lower montane forest, secondary or IA3c(1) 1,666 1,882 1,882 216 0 intervened 216 Tropical semi-deciduous mixed lower montane forest, secondary or IA3c(1/2) 14,573 13,443 13,443 -1,130 0 intervened -1,130 137 El Salvador Country Land Assessment World Bank Pacific mangrove forest on clay, secondary or intervened IA5b(1) 38,566 38,475 38,443 -91 -32 -123 Tropical deciduous broad-leaved lowland forest, well-drained, secondary IB1a(1) 154,935 125,501 122,881 -29,434 -2,620 or intervened -32,054 Short grass low land well-drained savannah with evergreen VA2a(1)(1)(a) 683 683 683 0 0 broad-leaved trees, variety Curatella americana, secondary or intervened 0 Short grass low land well-drained savannah with small semi- VA2b(1) 5,943 3,004 3,004 -2,939 0 delicious broad-leaved trees, variaty Crescentia alata, secondary or intervened -2,939 Altimontane grassland or paramo VC2b 256 256 256 0 0 0 Scarcely vegetated lava flow VIAd 6,323 6,323 6,323 0 0 0 Scarcely vegetated tropical dune and beaches, intervened VIB1a(1) 2,282 2,282 2,282 0 0 0 Tropical coastal vegetation in successional transition on very recent VIB3a 150 150 150 0 0 sediments, moderately drained, secondary or intervened 0 Tropical freshwater reed-swamp formation VIIB1a 5,327 5,518 5,543 191 25 216 River segment of the Pacific littoral SA1a(3)(a) 3,545 3,545 3,545 0 0 0 Predominantly freshwater lake of the Pacific littoral plain SA1b(4)(a) 17,240 17,301 17,301 61 0 61 Estuary of the Pacific SA3c(1) 18,569 18,569 18,569 0 0 0 Reservoir SA1b(5)(b) 17,517 17,517 17,517 0 0 0 Marine coastal ecosystem with fine sediment bottom of the SA3(2)(a)(b) 183 183 183 0 0 Pacific 0 138 El Salvador Country Land Assessment World Bank Marine coastal ecosystem with sandy bottom of the Pacific SA3(2)(b)(b) 4,895 4,895 4,895 0 0 0 Marine coastal ecosystem with rocky bottom of the Pacific SA3(2)(c)(b) 2,941 2,941 2,941 0 0 0 Marine coastal ecosystem with rocky bottom of the Pacific SA3(2)(c)(1)(b) 1 1 1 0 0 with coral 0 Totals all ecosystems area in hectares 397,161 353,610 350,283 -43,550 -3,328 -46,878 Totals forested area in hectares 311,305 270,443 267,090 -40,863 -3,353 -44,216 139