Peru - An Agricultural Development Strategy

DRAFT CONFIDENTIAL Report No.: 17270 PE PERU An Agricultural Development Strategy Decision Draft December 17, 1997 This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization. FILE COPY GLOSSARY OF ACRONYMS ADEX Asociaci6n de Exportadores Exporters' Association AID Agency for International Development BAP Banco Agrarlo del Peri Agrarian Bank of Peru CGIAR Consultative Group for International Agricultural Research COFIDE Corporaci6n Financiera de Desarrollo Development Finance Corporation CONADE Corporaci6n Nacional de Desarrollo National Development Corporation CONASEV Comis16n Nacional Supervisora de Empresas y Valores National Supervisory Commission for Companies and Securities DS Decreto Supremo, Supreme Decree ECASA Empresa Comercializadora de Alimentos Food Marketing Enterprise ENCI Empresa Nacional de Comercializaci6n de Insunos Inputs Marketing Enterprise FEAS Proyecto de Fomento de la Transferencia de Tecnologia a las Comunidades Campesinas de la Sierra Development Project for Technology Transfer to Peasant Communities in the Sierra FONAVI Fondo Nacional de Vivienda National Housing Fund FONCODES Fondo Nacional de Compensaci6n y Desarrollo Social National Social Compensation and Development Fund FONDEAGRO Fondo para el Desarrollo del Agro Fund for Agricultural Development FRASA Fondo de Reactivaci6n Agraria y de Seguridad Alimentaria Fund for the Reactivation of Agriculture and Food Security GDP Gross National Product GRADE Grupo de Analisis para el Desarrollo Group of Analysis for Development GLOSSARY OF ACRONYMS (Cont.) IDB Banco Interamencano de Desarrollo Inter-American Development Bank INADE Instituto Nacional de Desarrollo National Institute for Development INDECOPI Instituto de Defensa de la Competencia y de la Propiedad Intelectual Institute for the Defense of Competition and Intellectual Property INIA Instituto de Investigaci6n Agraria National Institute for Agricultural Research INRENA Instituto Nacional de Recursos Naturales National Institute for Natural Resources LSMS Living Standards Measurement Survey Encuesta Nacional de Niveles de Vida (ENNIV) MAG Ministerio de Agricultura Ministry of Agriculture MIPRE Ministerio de la Presidencia Ministry of the Presidency NGOS Orgamzaciones No-Gubernamentales Non-Governmental Organization PETT Proyecto Especial de Titulaci6n de Teirras Special Project for Land Titling PRONAA Programa Nacional de Apoyo Alimenticio National Program for Food Assistance PRONAMACHCS Programa Nacional de Manejo de Cuencas Hidrogrificas y Conservaci6n de Suelos National Program for River Basin Management and Soil Conservation PRTR Programa Rural para Titulaci6n y Registro Rural Program for Titling and Registration SENASA Servicio Nacional de Sanidad Agraria National Agricultural Health Service UGECI Unidad de Gestion para la Cooperaci6n Internacional International Cooperation Management Unit UNDP United Nations Development Program WUA Junta de Usuarios de Agua Water User Association Preface This report is based mainly on the findings of a mission to Peru during November 3-10, 1997. The mission was led by Mateen Thobani (LCSPR), who prepared the section on water use and wrote the report based on the contribution of seven consultants, whose affiliation and primary areas of responsibility are listed below: Dale Adams, retired professor, Ohio State University (Rural Finance, Annex 3) Victor Agreda, GRADE (Annex 2) Javier Escobal, GRADE (Chapter 1, Marketing, Annex 1) Juan Jose Marthans, U. of Lima (Commercial Banking in Agriculture, Annex 3) Ramon Lopez, U. of Maryland (Research and Extension, Rural Poverty) Alberto Pasco-Font, GRADE (Land, Animal and Plant Health, Forestry) Martin Valdivia (Government Expenditures in Agriculture) Patricia Mendez and Patricia Chacon Holt helped in the document layout and distribution. The report was commissioned by the Country Management Unit (CMU) for Peru, whose Director is Isabel Guerrero. As the CMU anchor and Lead Specialist respectively for the Environmental and Social Sustainable Sector Management Unit, Pierre Werbrouck and Thomas Wiens oversaw the task. Alberto Valdes (RDV) and David Nielson (AFT) were peer reviewers for the study. TABLE OF CONTENTS EXECUTIVE SUM MARY ................................................................................................ i 1. THE STRUCTURE AND EVOLUTION OF PERUVIAN AGRICULTURE............ Agricultural Resource Base and Ownership.......................................................... ............................. .... R ecent Policies and Perform ance............................................................................................................ Constraints to Agricultural Growth ......................................................................................................... Government Expenditures Oriented to Agriculture......................................................... .............. .... . 6 Toward a Strategy for Agricultural Development................................................................................. ... 8 2. POLICIES TO STIMULATE AGRICULTURAL PRODUCTIVITY AND INVESTMENT ........................................................................................................... 10 Land Tenure and R egistration ................................................................................................................ 10 Irrigation and W ater Use........................................................................................................................ 12 F orestry.................................................................................................................................................15 Plant andAn m al H ealth........................................................................................................................ 17 R ural F inance ........................................................................................................................................ 19 R esearch and E xtension ......................................................................................................................... 23 M arketing..............................................................................................................................................27 3. COMBATING RURAL POVERTY: EVIDENCE AND POLICY...........................29 R ural Poverty Profile............................................................................................................................ 29 Poverty Alleviation Measures................................................................................................................. 30 R ecom m endations..................................................................................................................................32 ANNEXES Annex I Statistical Appendix Annex 2 Cases Studies of Recent Successful Small Agricultural Producers in Peri Annex 3 Rural Finance in Per-6 FIGURES Figure 1.1 Land Distribution Ginis For Selected Latin American Countries Figure 1.2 Real Exchange Rate And Agricultural Production 1988-1996 Figure 1.3 Real Farngate Prices For Selected Crops 1988-1996 TABLES Table 1.1 Poverty In Perii, 1985-1996 Table 1.2 Government Expenditures In Agriculture By Institution Table 1.3 Government Expenditures In Agriculture By Type Of Support Program Table 2.1 Extension Programs In Peru: Expenditures And Beneficiaries Table 3.1 Rural Poverty By Physical Region (1994) Executive Summary Overview 1. Following several years of decline, Peru's agriculture has grown at an average annual rate of 9.4 percent since 1992, even faster than GDP growth of 7.2 percent per year over the same period. Non-traditional agricultural exports over the period have grown by 24 percent annually and now exceed the value of traditional agricultural exports. The strong and continued recovery appears to be due to a stable macroeconomic climate, a decline in terrorism, normal weather conditions, increased investment in rural infrastructure, especially roads, and a more liberal production environment. Strikingly, the increased agricultural production has been almost entirely due to an increase in area cultivated rather than to an increase in yields, suggesting increased confidence in the sector and the potential for gains in productivity. While rural poverty has decreased substantially since its peak in 1991, it is still very high. Moreover, it is higher than it was during the mid-eighties. 2. There is tremendous potential for additional agricultural growth arising from increased agricultural productivity and investment. To realize this potential, the study finds that the government will need to implement policies and programs to improve land and water use as well as agricultural services, specifically plant and animal health, research and extension, rural finance, and marketing. In addition, it will need to continue with its investments in rural infrastructure, especially roads. Also, to exploit Peru's vast forestry resources, the government should take measures to promote the sustainable development of forestry. Recommended policies and programs in each of these areas are outlined below. 3. Although the Costa has the most potential for growth, recent success stories in the Sierra with processed potatoes, insect dyes (cochinilla), and alpaca wool suggest that market-oriented policies could also permit rapid agricultural growth in the Sierra. Similarly, in addition to forestry, the Selva has potential to substantially expand its production of rice, coffee, cocoa, and palm oil production. Improvements in rural infrastructure and agricultural services will help realize the Selva's potential. 4. Rapid growth of agricultural production and incomes should help reduce poverty. However, there are still likely to be many rural residents, especially at high elevations (altiplano), for whom the reforms will have only minimal impact. In addition, there are many living in extreme poverty who would require many years of strong income growth before they could climb out of poverty. This study therefore also proposes a more targeted strategy for poverty alleviation. - ii - Land Productivity and Investment 5. Laws allowing sales of agricultural lands and removing limits on the size of land holdings has helped increase investment in the sector and improve land productivity. However, there is still valuable land, especially along the coast, that is being used in economically inefficient ways, even as there are investors who would like to put the land to better uses. Two factors prevent the market from bringing this about: (i) insecure land titles, which make investors reluctant to buy the land, and (ii) land held by peasant communities either as sugar cooperatives or as communal property, which cannot be sold. To resolve these issues, the government first needs to increase the pace of land titling by addressing some of the factors that have slowed down the process. In particular, the government should consider removing UNDP from the process and transferring its administrative and accounting functions to UGECI, whose functions should be limited administration and monitoring. Second, the government needs to find a politically acceptable way to transfer ownership and control of sugar cooperatives to other private investors. Even if this involves a fiscal subsidy, it may be preferable to continued fiscal bleeding caused by government loan guarantees and bailouts. The sale of lands and assets to private investors would bring in much-needed investment and management skills into the sector. Thirdly, where communities request it, the government should respond to their demands to individually title communal land. Water Use and Irrigation Infrastructure 6. Despite increasing water scarcity along the coast and large investments in irrigation infrastructure, water is often wastefully used and Peru's irrigation infrastructure is poorly maintained. For political and social reasons, water tariffs of small farmers cannot be raised to levels that would induce more rational use of the water and generate sufficient funds for system maintenance. Awarding water rights to existing users without charge and allowing them to be traded offers the best incentive for conserving water and allocating it to more productive purposes. Following several years of discussions, a draft law has emerged which enjoys widespread support. The Government should consider passing the law. If, however, the government prefers to establish tradable water rights on a trial basis, an appropriate way to do so would be to use the irrigation schemes being rehabilitated under the irrigation sub-sector project as pilots. 7. Under any system of water rights, providing technical assistance to water user associations to allow them to better operate and maintain systems deserves top priority. An important change, especially for irrigation schemes in the Sierra, is to revamp its system of water distribution from one based on water on demand to water distribution patterns that are contiguous in order to reduce water losses through evaporation and canal seepage. A third recommendation in the water area is to consider a strategy for the privatization of the large irrigation projects executed by INADE in order to get the schemes completed faster and cheaper. Under such a strategy, the control and management of the projects would be handed over to existing users or private investors. The government's role would be limited to that of financing components of the project - iii - that have significant positive externalities or are of a public goods nature and to providing any necessary guarantees. Forestry 8. Peru has vast forestry potential -- 75 million hectares are forested, the bulk in the Amazon basin. Even after excluding the 25 million hectares in protected reserves, 40 million hectares could potentially be commercially exploited -- an area the size of Switzerland. However, Peru produces only one million cubic meters of wood annually and exports only $25 million, about $0.3/hectare as opposed to ratios of $210 and $185/hectare in Malaysia and Chile. This is mainly because logging contracts are typically granted for small areas and for relatively short periods. Moreover, because they are scattered all over the jungle, enforcement and control of illegal logging is difficult. An estimated 260,000 hectares of forest are deforested every year. A new law providing long-term concessions for forestry is essential for promote forestry. However, this law needs to have sufficient protection built in to ensure sustainable forestry management and provide incentives for reforestation, especially in the Sierra. Present drafts of the law before Congress do not pay sufficient attention to this issue. The new law should also permit joint forestry ventures between peasant communities and private investors. Providing employment and income in the highlands should reduce the migration and illegal logging in the jungle. Animal and Plant Health Services 9. Animal and plant health issues are increasingly important to improving the productivity of the crop and livestock sector. They are also crucial to enabling fruit and vegetable exports and to protecting the public from foreign pests and diseases. Thus recent initiatives to strengthen SENASA and fund biological pest control, eradication of fruit flies, and animal health are very important. The government now needs to make sure that the program is implemented expeditiously. In addition, to be credible and effective, SENASA's recommendations need to be based purely on technical grounds and not used as a means of agricultural protection. In cases where there is legitimate threat of pests or diseases being imported into Peru, it should work closely with INDECOPI to find the most cost-effective way of prevention. To help SENASA become more independent, it should be overseen by an autonomous and independent Board, elected by both the government and the private sector. Finally, to clarify fragmented health laws and regulations going back decades, issuance of a new law which consolidates, updates and clarifies laws will greatly facilitate proper enforcement of health standards while ensuring that such laws are not used as an unjustified non-tariff trade barrier. Rural Finance 10. The liquidation of the Banco Agrario del Peru (BAP) and the collapse of other institutions such as credit unions left a large gap in Peru's rural financial market. In response, the the government has funded several ad hoc agricultural credit programs such as FONDEAGROs and Fondos Rotatorios. These schemes have discouraged financial - iv - discipline and continued farmer efforts for additional public subsidies. The government has also encouraged the formation of private rural banks (Cajas Rurales) and provided banks access to COFIDE funds for agricultural lending. During the past 5 years substantial progress has been made in rebuilding parts of the rural financial system: 16 fledgling Cajas Rurales have been formed, some Cajas Municipales now provide loans to farmers, and commercial banks have steadily increased their lending for agricultural purposes. However, major gaps persist in rural financial markets: most of the formal agricultural credit is concentrated in the coastal region, there is little long-term lending, most agricultural loans are tied to real estate collateral, small farmers have limited access to formal loans, and many rural people still do not have access to deposit services. 11. To deal with the situation, one option is to play a passive role in rural financial market development and to wait patiently for markets to work better. This approach risks social unrest and continued ad hoc programs which hamper development of efficient rural financial markets. A preferred option is to play a more aggressive role in rebuilding these markets. Under this option, the government would form an inter- agency committee to foster rural financial development. The committee would be responsible for formulating and carrying out a development strategy for rural financial markets, including how to lower transaction costs and how to reduce lending risks. It would also monitor and deflect government and regulator actions that boost transaction costs in rural markets and coordinate donor efforts to ensure that only practices and policies that enhanced the overall performance of rural financial markets would be allowed. This might include insisting on cost-recovery rates being charged on loans. Fostering training and research on the development of efficient of rural financial markets could be another important function of the committee. Development partners can best contribute to Peru's rural financial development by supporting the implementation of policy reforms and institutional and technical assistance rather than via lines of credit. Research and Extension 12. A good system for the delivery of agricultural research and extension services can help increase productivity and safeguard the natural resource base. Peru has many extension programs - an estimated 260,000 farmers receive some form of technical assistance. Among the diversity of programs, the most successful appear to be those that are demand-oriented in nature, providing farmers and farmer organizations technical assistance in non-traditional areas such as marketing, credit, and water and soil management, often phasing out their involvement over a 3-4 year period. These schemes are worth expanding further, with public resouces concentrating on those small, but commercially viable farmers. However, the top-down extension services of INIA, which focus on expanding yields of traditional crops, need to be revamped. 13. In contrast, resources spent on agricultural research are limited. Most of the country's research is carried out by INIA under a $5 million research budget. There is an urgent need to expand agricultural research activities. The expansion needs to be more demand-based, focusing on exploring market opportunities and the introduction of new -v- crops rather than focusing exclusively on traditional crops for domestic markets, as is the case with INIA presently. To help bring this about, INIA should develop a research strategy in collaboration with domestic and international research providers, NGOs and universities. The strategy should focus public research funds on small commercially- viable farmers but consider co-financing mechanisms and fiscal incentives to support other kinds of agricultural research. The research should also respond to post-harvest and marketing needs. INIA needs to explore the extent to which CGIAR may assist in establishing a new research program. Marketing 14. Although the costly system of guaranteed producer prices and government marketing of crops and inputs has been dismantled, several interventions in the output and input markets remain. The most significant of these are those by PRONAA, ENCI and Fondos Rotatorios. While the size and fiscal costs of intervention are small, they distort the market and have an adverse impact on product quality and market development. They should be phased out in favor of increased programs providing technical assistance and rural infrastructure to the same groups. 15. At the same time the state could usefully facilitate marketing efforts in in three ways. First, the public information system of providing prices on cultivated areas needs to be improved to allow farmers to make planting decisions in time. At present this information comes with a two to three month delay, too late for most planting decisions. Second, the government should support the Lima municipality in its efforts to resurrect the project establishing a new wholesale market in Lima. The present municipal market, which handles 40 percent of perishable crops in the country is inefficient, chaotic, and congested. It lacks adequate refrigeration facilities and results in output losses and oligopsonistic pricing behavior. A new wholesale market for perishables would help reduce losses and raise farngate prices. Third, the government should support the recent opening of the privately-established commodity exchange (bolsa de productos) for non- and semi-perishables. In particular, it should help CONASEV develop a system of warehouse certification, especially for semi-perishables such as rice and sugar. This could be a useful form of collateral for farmers. Rural Poverty Programs 16. About 65 percent of Peru's rural population is classified as poor, with 38 percent classified as extremely poor. The bulk of the poor are in the Sierra, followed by the Selva. Among the rural poor, those engaged in agricultural activities tend to be the poorest, with the landless being the poorest. The poor tend to have surprising high levels of education, but the education has low returns, suggesting that the quality of education is poor. Family size and the dependency ratio has a strong correlation with poverty and a large quantitative impact. In contrast, access to land has a small impact, suggesting that a land redistribution policy is likely to be costly and ineffective. - vi - 17. Over the years, the Government, donors, and NGOs have implemented many programs to alleviate poverty. Traditional programs, which focussed on increasing agricultural production through input and marketing subsidies, have proven to be ineffective and costly, while benefiting the non-poor disproportionately. In contrast, recent poverty alleviation programs, such as those of FONCODES, PRONAMACHCS, PAR and FEAS, have had better success. Their novel features are their reliance on communities and beneficiaries for the design and implementation of the programs. The programs correctly target poor households in the Sierra and Selva, require communities to contribute to the costs of the program, and foster community participation, often helping strengthen the community. 18. Although Peru gets high overall marks for its poverty alleviation programs, five principles could improve their cost-effectiveness and widen its scope: (i) follow the demand-driven principle: While most already are, there are still some such as PRONAA and Fondos Rotatorios which do not meet this criterion. In addition, these programs hamper the development of agricultural markets. (ii) improve donor coordination: Because may programs operate in the same area, coordination, which could be done by regular meetings between the various instititions, is desirable. (iii) improve rural education: The education components of programs are typically oriented to building classrooms. More emphasis should be given to incentives for improving teacher quality, encouraging parent participation in school activities, and improving technical skills and on-the-job training. (iv) expand family planning programs and introduce social security: Given the strong impact of family size on poverty, more resources could be devoted to contraceptive programs in areas where there are problems of the supply. Over the longer-term, consideration should be given to implementing a social security system for the elderly, who are often overlooked in current programs. Such a system could also contribute to lowered birth rates. (v) establish an evaluation system: To ensure that the programs are functioning well, funds to carry out independent evaluations should be budgeted, preferably not in the agencies responsible for the projects. 1. THE STRUCTURE AND EVOLUTION OF PERUVIAN AGRICULTURE Agricultural Resource Base and Ownership 1.1 Although Peru's land area of 129 million hectares is greater than that of France and Italy combined, Peru is not intrinsically an agricultural economy. Only 6 percent of its land is cultivable. With its population of 25 million, Peru's per capita availability of cultivable land is only 0.30 hectares, much lower than the average for South America (0.44 hectares). Furthermore, while Peru's water resources are abundant, they are poorly distributed. Ninety-eight percent of the country's fresh water is located in the Amazon basin, which has limited agricultural potential, while the fertile, low-rainfall (below 50 mm/year) coastal region is entirely dependent on uncertain and highly variable seasonal water from 53 rivers originating in the Andes. 1.2 Peru has tremendous forestry potential. Of its 84 million hectares of natural forests, 90% are tropical heterogeneous forests located in the amazon basin. Even after excluding 25 million hectares that have been reserved as protected areas and an estimated 10 million hectares of land that have been deforested, 49 million hectares could potentially be exploited for commercial purposes. 1.3 Farm sizes tend to be small, especially in the coastal region (Costa) and in the highlands (Sierra). According to the 1994 Agricultural Census, Peru has 1.7 million farming units cultivating 5.9 million hectares of agricultural land. Over half of the 249,000 coastal farms are less than 3 hectares. In the Sierra, 63 percent of the 1.2 million existing farms are below 3 hectares. In contrast, three-quarters of the farms in the jungle area (Selva) are greater than 3 hectares, and two-fifths greater than 10 hectares. Following the 1969 Agrarian Reform and subsequent parcelization of land, land distribution is better than that in other Latin American countries (Figure 1.1). Farm units are heavily fragmented. Farmers own an average of 3.3 plots, over half of which are below 3 hectares. In the Sierra, a quarter of the farmers own 5 or more plots, some no more than a few square meters. Recent Policies and Performance 1.4 Since taking office in 1990, the Fujimori Administration instituted a program of profound economic reforms aimed at restoring macroeconomic equilibria and introducing structural reforms to promote a more competitive market economy. These reforms swept away much of what had been a highly interventionist regime. Following a year or two of slow recovery, GDP has grown at an average annual rate of 7.3 percent since 1992 and annualized inflation during the last 12 months has dropped to 7.1 percent. In agriculture, early reforms included: (a) the substantial liberalization of the agricultural trade regime; (b) the elimination of price controls over agricultural products; (c) the elimination of most agricultural input subsidies; and (d) severe downsizing or liquidation of most public 2 FIGURE 1.1 LAND DISTRIBUTION GINIS FOR SELECTED LATIN AMERICAN COUNTRIES 100 92% 90% 80% 101 70% 60%, 54%' 20%6 10 Panzam Costa Rica Peru Cuba Brasil Nicaragua Ecuador OMile Colombia Paraguay uatemala Uriguay Source: Land Tenure Center, Univ. of Wisconsin. Peru estimate by GRADE based on the 1994 Agricuttural Census. agricultural institutions including the Ministry of Agriculture, marketing agencies (ENCI and ECASA), the Agrarian Bank (BAP), and the agricultural technology institution (INIA). 1.5 While these agricultural reforms were necessary for macroeconomic stability, they were detrimental to short-term agricultural growth because they removed agricultural subsidies and dismantled public institutions. Recognizing the importance of creating new laws and institutions to support private sector-led agricultural growth, the Government began a series of additional reforms. These reforms aimed to: (a) encourage private sector provision of agricultural research activities through innovative contractual relationships with the private sector and NGOs for the use of INIA's agricultural research field stations and facilities; (b) strengthen property rights for land through the passage of a new land law and through the initiation of a Land Titling and Registration Project financed by the IDB; (c) rationalize the use of water through the drafting and consideration of a proposed new water law, and also through technical assistance and loans to user groups to rehabilitate deteriorated irrigation systems; (d) make credit available to agricultural producers through programs run or promoted by the Ministry of Agriculture; (e) establish a modem system of animal and plant health services through a large technical assistance project to strengthen SENASA, the regulatory and implementing agency for animal and plant health services; and (f) improve small farmer income and production through programs providing selected production inputs (such as fertilizer, credit, seed, and equipment) on a grant basis to small farmers in economically backward areas. 3 1.6 Peru's agriculture sector has performed well since 1992, when there was a decline in agricultural output caused by a combination of poor weather and reduced agricultural subsidies. The sector has since grown at an annual rate of 9.4 percent, reaching $3.6 billion in 1996 At current prices this represents six percent of GDP. However, because of problems associated with price indices during periods of hyperinflation, a more reliable picture of the importance of agriculture is obtained from calculating its share of GDP at constant prices, which is 12.6 percent. Over this period, non-traditional exports grew at an annual average rate of 24 percent, exceeding the value of traditional agricultural exports 1.7 Despite the spectacular export performance, agricultural exports only account for 15 percent of agricultural GDP, about half the South American average. The bulk of the recent agricultural growth came from increases in the production of potato, cotton, rice, coffee and poultry. Impressively, this growth occurred even as the sol appreciated by 20 percent between 1990 and 1992 and an additional 2 percent during the period 1992-1996 (Figure 1.2). Real farmgate prices for most crops also declined substantially during the nineties (Figure 1.3). FIGURE 1.2 REAL EXCHANGE RATE AND AGRICULTURAL PRODUCTION 1988-1996 252,0 10S 100 202,0 \ 95 90 152,0 85 80 102,0 Real Exch mge Rate 52,0- 70 Agncu 65 2,0 60 1988 1989 1990 1991 1992 1993 1994 1995 1996 So 1 co ilateral Re d Exchan e Rte Agricultur ad C tral Production Valu(1 Sources: Compiled by GRADE based on Ministry of Agriculture and Central Reseve Bank 4 Figure 1.3 REAL FARMGATE PRICES FOR SELECTED CROPS 1988-1996 9,0- 8,0 7,0 6,0 5,0 4,0- 3,0, 2,0- 2,o ..-.- . ................. . . . 1,0 0,0 . I I I 1988 1989 1990 1991 1992 1993 1994 1995 1996 RICE. BEANS - YELLOW MAIZE -COTTON COFFEE POTAT Source: Compiled by GRADE from Minisby of Agriculture data 1.8 The Living Standards Measurement Surveys (LSMS) indicate that, despite the contrary claims of skeptics, Peru's level of rural poverty decreased substantially since 1991 (Table 1.1). However, rural poverty levels are still worse than in the mid-eighties and the improved performance appears to be due to the decline in terrorist activity in rural areas and to the stable macroeconomic climate, rather than to the impact of agricultural reforms. For instance, the increased production has come largely from increased area under cultivation rather than more increased yields (Annex 1, Table 4.3), possibly reflecting a return to farms abandoned because of terrorism. Somewhat puzzlingly, although rural poverty has declined somewhat since 1994, overall poverty has increased slightly, despite strong GDP growth. TABLE 1.1 POVERTY IN PERU, 1985-1996 (Percentage of poor households) Poverty Extreme Poverty Region 1985-86 1991 1994 1996 1985-86 1991 1994 1996 Urban 35.7 51.0 38.1 41.9 12.9 18.8 6.3 7.3 Rural 54.8 72.7 66.1 65.3 35.0 54.5 40.4 37.9 Total 41.6 55.3 46.5 49.0 18.4 24.2 16.6 16.6 Source: Moncada (1996); for 1996, GRADE estimates based on LSMS surveys. 5 1.9 The benefits from the second wave of agricultural reforms have yet to be realized. Several reform initiatives have either faltered (such as the water law) or been incompletely implemented (land registration, rural credit, research and extension, and plant and animal health). The cropping pattern still does not seem to reflect crop profitability The rank correlation between land distribution among the 29 most important crops (controlled by region) and crop value per hectare is negative for each of the last 20 years. That is, agricultural production fails to move towards those crops that are potentially most profitable. This asymmetry between the overall structure of incentives and sector performance suggests that distortionary policies and programs or other constraints prevent resources from moving toward more profitable activities Constraints to Agricultural Growth 1.10 Based on past studies, field interviews and an analysis of existing data, the team preparing this report finds the insecurity of property rights to land, forestry and water to be important obstacles to increased investments in agriculture. Laws allowing sales of agricultural lands and removing limits on the size of land holdings has resulted in an active market in agricultural land sales in recent years -- farmers have sold an estimated 37,000 hectares of agricultural coastal land in the last one and a half years. However, many irrigated coastal lands suitable for growing high value crops such as asparagus are still used for growing low-value crops such as corn. This is mainly because of a lack of registered titles. Not only does it make access to formal credit extremely difficult, it deters potential buyers with better farming and marketing skills and access to capital from buying this land. Only 35 percent of coastal land plots and 17 percent of plots nationwide are registered. The rural land titling and registration unit, PETT, has difficulty in registering more titles. Of the 251,000 titles granted by PETT between 1993 and May of 1997, less than 36,000 have been registered. Because of the lack of registered titles, multiple titles for the same property are common, which makes investors reluctant to buy agricultural lands. 1.11 Even in areas of water scarcity along Peru's coast, water is often used to irrigate low-value crops using inefficient irrigation technology, while vast tracts of fertile coastal land along major roads remain undeveloped for lack of water. This is because there are few incentive for farmers to conserve water or reallocate it to higher-value uses. Water charges are low and water rights are insecure and cannot be traded separately from land. The benefit from water saved by more efficient irrigation technology or growing less water- intensive crops accrues to other water users rather to the farmer saving water. This situation reduces the incentive to use water efficiently. Similarly, the lack of secure long-term concessions for forestry encourages illegal logging using unsustainable practices, even as Peru's tremendous forestry potential remains largely untapped. Current forest concessions are for short periods - no more than 20 years -- in small dispersed areas, so that enforcement and control of logging is extremely difficult. As a result, Peru exports only $25 million worth of wood annually, about $0.3/hectare, while countries such as Malaysia and Chile export $210/hectare and $185/hectare respectively. 1.12 Agricultural services are another source of obstacles to sustained high growth of the sector. Based on interviews of many farmers in the all regions of the country as well as an analysis of Peru's agricultural successes (Annex 1), animal and plant health services are 6 mentioned as a key constraint to increasing agricultural productivity and exports. As a result of limited public resources devoted to this activity since the eighties (less than US$ 2 million per year), major diseases and pests have attacked Peruvian agriculture, lowering production and preventing exports. Similarly, agricultural research has all but disappeared in Peru. And following the closure of Peru's Agrarian Bank (BAP) in 1992, rural financial services, especially credit, are often cited as a constraint to growth. Formal finance is limited to those with registered land as collateral and there is little long-term finance in agriculture. 1.13 Peru's vast area and terrain makes transport infrastructure a natural constraint to more competitive agriculture. Recent government policies and expenditures to rehabilitate road infrastructure and improve ports and airports have done much to alleviate these constraints. However, Peru's geography imposes important limits to agricultural growth, mainly in the Sierra and Selva, and makes marketing problematic. An efficient marketing system would do much to overcome these constraints. However, marketing of inputs and outputs being carried out by the government agencies, ENCI and PRONAA, while small suffer from many of the inefficiency problems plaguing most agricultural marketing parastatals around the world. In addition to the administrative inefficiencies, these parastatals introduce distortions and uncertainties in these market, hampering its efficient and sustainable development. At the same time, marketing of perishables through the Lima's municipal wholesale market, which handles 40 percent of products nationwide, is inefficient, chaotic and congested. This market lacks refrigeration facilities and results in output losses and oligopsonistic pricing behavior. Government Expenditures Oriented to Agriculture 1.14 The government runs a variety of programs to stimulate agricultural production and provide assistance to poor rural households. Presently four government ministries manage these programs: Ministry of Agriculture, Ministry of the Presidency (MIPRE), Ministry of Women and Human Development, and Ministry of Transportation, Communications, Housing and Construction. During 1995, public expenditure for just the agricultural components of these programs was about $650 million, roughly 18 percent of agricultural GDP and 1.3 percent of total GDP. (In the case of INADE, all expenditures have been attributed to agriculture because it was difficult to distangle the expenditures and because these projects have yielded virtually no non-agricultural benefits to date). In 1996, this figure dropped to only $400 million, as a result of a drastic reduction in INADE's investments (Table 1.2). These figures exclude expenditures by the Ministry of the Women and Human Development for the National Program for Food Assistance, PRONAA, and the Program for Rehabilitation Assistance (PAR). Although these programs total to over $100 million a year, their agricultural support component is small. 7 TABLE 1.2 GOVERNMENT EXPENDITURES IN AGRICULTURE BY INSTITUTION (Thousands Of Dollars) 1995 1996 Ministry of Agriculture 133,598 129,889 PETT 488 6,561 INIA 8,491 9,024 SENASA 4,106 6,064 PRONAMACHCS 32,355 50,158 Other (INRENA, FEAS, Central Administration) 88,159 58,081 Ministry of the Presidency (MIPRE) 507,315 246,988 INADE 426,228 168,865 FONCODES 27,019 34,583 Regions 54,068 43,540 Ministry of Transp. and Communications 7,506 24,301 Total 648,419 401,177 Share of public exp. in agric. GDP (with INADE) 17.9% 10.3% Share of public exp. in agric. GDP (without INADE) 6.1% 6.0% Sources: Ministry of Economy and Finance (MEF) and Congressional Budget domnuttee 1.15 These fiscal expenditures do not include indirect support to the sector such as condoning the tax debt of sugar cooperatives and refinancing their debt at favorable interest rates. Nor do they include fiscal costs of exonerating the sector from the main labor tax, FONAVI, or from reducing income taxes from 30 percent to 15 percent, thereby exempting the sector from the minimum income tax affecting the rest of the economy. Similarly, they do include lost revenues from exonerating the sector from the value-added tax or from import duty exemptions on agricultural inputs or additional income tax exemptions for investments in non-irrigated land. Such tax and import tariff exemptions for agriculture distort incentives in favor of agriculture at the expense of other sectors, consumers, and taxpayers as well as encourage lobbying efforts for special treatment by farmers and others. 1.16 Although INADE's expenditures on large multipurpose projects with major irrigation components has decreased since 1995, they still represent 42 percent of the public expenditure in agriculture (Table 1.3). Because of changing budgetary priorities and large inefficiencies associated with their construction, the projects have been under execution for decades, with huge cost-overruns. Under the new market-oriented paradigm, there is little justification for the government to continue to incur huge expenditures for these projects. In contrast, the increasing trend on public expenditures for rural roads and marketing is welcome. Improved rural infrastructure will do much to integrate rural households in the country's development. Similarly, increased expenditures on land titling and registration and plant and animal health services are desirable because land insecurity and plant and animal diseases are major impediments to continued agricultural growth. Also given externalities arising from protecting natural resources, the increased expenditures by INRENA and PRONAMACHCS appear desirable. However, increased expenditures on production subsidies, which consist mainly of grants to small farmers in the form of agricultural machinery and equipment, fertilizer, and pesticides, are worrisome. Not only do such 8 distortionary grants hamper the development of local input market, they are likely to lower incomes of farmers that are excluded from these schemes. TABLE 1.3 GOVERNMENT EXPENDITURES IN AGRICULTURE BY TYPE OF SUPPORT PROGRAM (Thousands Of Dollars) 1995 1996 (1) Production subsidies 21,253 29,666 (2) Marketing and Transport Infrastructure 13,634 33,557 (3) Land Titling and Registration 651 6,561 (4) Research and Development 8,969 9,660 (5) Plant and Animal Health Services 877 6,922 (6) Irrigation Infrastructure (other than INADE) 24,880 37,533 (7) Technology Transfers 7,054 5,048 (8) Conservation of Natural Resources 17,439 29,759 (9) Other 73,366 30,065 Sub-Total 168,123 188,772 (10) Regions (MIPRE) 54,068 43,540 (11) INADE Investments 426,228 168,865 Total 648,419 401,177 Share of INADE expenditures 66% 42% Note: Row (9) includes administrative overheads Sources: MEF and Congressional Budget Committee 1.17 In conclusion, the recent picture of public expenditures for agricultural activities is largely positive. There is a desirable increasing trend towards expenditures on rural roads, land titling and registration, plant and animal health services, and small-scale investments in irrigation and natural resource conservation. In addition, the reduced expenditures on INADE's projects is probably desirable, though a longer-term strategy of the future of these schemes needs to be designed and implemented (see next chapter). In contrast, the increasing trend of production subsidies and transfers is disturbing as it hinders market development and hurts those that are unable to obtain the subsidies. It would be better if targeted poverty alleviation measures were based on the demand-driven, participatory models of programs such as FONCODES, PRONAMACHCS, and FEAS. These programs use community labor and the use of external contractors for implementation, thus minimizing distortions in the input or output markets. Toward a Strategy for Agricultural Development 1.18 The scenario above suggests that Peru's agricultural economy has tremendous potential and is poised for continued strong growth in the future if certain constraints could be removed. This is particularly true along the coast where fertile soils and easy access to ports could sustain a large expansion in high-value commodities, especially fruits and vegetables. However, as the experience with processed potatoes, insect dyes (cochinilla), and alpaca wool demonstrates, market-oriented policies can also lead to strong agricultural growth in the Sierra (Annex 2). Similarly, with improved rural infrastructure 9 and agricultural services, the Selva has potential to substantially expand its production of coffee, cocoa, and palm oil production. In addition, it has tremendous potential for forestry development. 1.19 The objective of this report is to suggest a strategy for agricultural development to more fully realize Peru's potential in agriculture. More specifically, the strategy aims to. * achieve dynamic and sustainable growth in agricultural production and agricultural productivity; * increase growth in commercial agriculture, particularly in export-oriented commodities; and * achieve these objectives through mechanisms consistent with macroeconomic stability, the promotion of an active private sector, integration in world markets, and the use of policies which do not distort relative prices The strategy will suggest specific policies, programs and institutional reforms to the extent possible. However, in many instances the implementation of the strategy will require additional detailed work and possible technical assistance. 1.20 The analysis above suggests that the key to strong and sustainable growth lies in: (i) strengthening property rights to land, water and forests and (ii) improving agricultural services, specifically plant and animal health, rural finance, marketing, and agricultural research and extension. These sub-sectoral issues are discussed in Chapter 2. Moreover, to ensure that alleviating these constraints will lead to economically efficient growth, the government will need to address some distortions in the fiscal and trade regime which divert resources in favor of agriculture at the expense of other sectors and taxpayers. The distortions arise from import trade protection through higher tariffs and variable import surcharges (sobretasas) as well as from fiscal exemptions for many agricultural activities. 1.21 While the resulting agricultural growth is likely to increase incomes and employment, thereby reducing poverty, there are still likely to be many rural residents, especially at high elevations (altiplano), for whom the reforms are likely to have little impact. Moreover, there are many living in extreme poverty, who would require many years of strong income growth before they could climb out of poverty. For these people, the government could design certain programs to alleviate poverty. Thus a fourth objective of this report is to recommend targeted programs aimed at alleviating the poverty of the rural poor, which is the subject of Chapter 3. There are three annexes: a statistical annex (Annex 1), an annex analyzing recent agricultural success stories (Annex 2), and an annex on rural finance (Annex 3). Background papers on rural poverty and agricultural research and extension are available upon request. 10 2. POLICIES TO STIMULATE AGRICULTURAL PRODUCTIVITY AND INVESTMENT Land Tenure and Registration 2.1 According to Peru's National Institute for Natural Resources, INRENA, six percent (7.6 million hectares) of Peru's land area of 129 million hectares is cultivable. Of the potentially cultivable area, the 1994 agricultural census reveals that only 5.9 million hectares are actually cultivated. The remaining 1.8 million hectares of cultivated land, mostly along the coast, is unirrigated land owned by either peasant communities (Comunidades Campesinas) or the government. This land is not exploited mainly because it lacks access to water. 2.2 Agricultural land in Peru is characterized by fragmented holdings. On average farm units own 3.3 plots, over half of which are below 3 hectares. The reasons behind the fragmented holdings, especially in the coast, can be traced to the 1969 Agrarian Reform, which imposed limits on rural land holdings (150 hectares for irrigated coastal land) and converted the large landholdings into cooperatives, most of which subsequently broke up into small plots (parcelas). For the Sierra, however, farmers maintain fragmented land holdings as a way of managing risks by cultivating lands in different ecological zones. This, together with a complicated mechanism of land division for bequests and land grouping through inter-community marriages, explain an important part of the fragmentation process. 2.3 Under the Agrarian Reform, 9.5 million hectares of land, previously held by 10,000 owners, were expropriated and transferred to over 400,000 families who worked on these farms. The cooperative structure did not work as envisaged and agricultural GDP stagnated, with the income of the landless and peasant farmers falling as the demand for their wage labor declined. By the mid-eighties, most of those Cooperatives had collapsed, leading to the division of the land in small plots among members.' However, the sale of these lands were prohibited and few land titles were registered.. 2.4 An important exception to the parcelization process was the land held by sugar cooperatives, which was maintained under cooperative ownership for reasons of economies of scale. However, mismanagement and reported corruption have led these cooperatives into bankruptcy and pressured the government into debt forgiveness and debt restructuring at favorable interest rates. Thus 60,000 hectares of prime agricultural coastal land with an ample supply of water is producing losses and bleeding the treasury. 2.5 The Fujimori Administration has done much to liberalize the land market. In 1991, the Law of Promotion of Investment in the Agricultural Sector (Legal Decree 653) explicitly allowed the sale of rural lands and raised limits on land holdings. However, the 1979 Constitution then in effect limited the Government's ability to allow sales of land belonging to peasant communities (comunidades campesinas and comunidades nativas). Under the umbrella of 1993 Constitution which removed such restrictions, the Government passed the Land Act (Law No 26505), which eliminated land size limitations and outlined IFor a fuller discussion of the land reform and subsequent economic policies, see World Bank 1992. 11 procedures to allow members of cooperatives and other communal land property schemes to individually title and register their land. 2.6 These reforms have spurred an active market of agricultural land in some coastal areas. An estimated minimum of 51,000 hectares of agricultural coastal land have been sold over the last 18 months, including 14,000 hectares of new agricultural land under the INADE-executed Chavimochic project. These sales should do much to increase private investment, land productivity and agricultural production and exports. 2.7 Even so, a lot of agricultural land in Peru is still be used for crops with low profitability using low levels of technology, even as adjacent land enjoys very high rates of return. The sale of this land to farmers who are better able to use it does not occur mainly because the bulk of agricultural land lacks secure and registered land titles According to the 1994 Agricultural Census, of the estimated 5.6 million plots, only 971,669 have registered titles. Even this is likely to be an overestimate as it is based on self-declaration. The Land Registry is unable to provide more reliable figures. New investors and farmers wishing to expand their profitable activities are reluctant to buy lands without registered titles because of frequent instance of multiple titles to the same land and because of the difficulty in getting bank financing without a registered title. 2.8 In addition, the lands held by sugar cooperatives are used very inefficiently. Reportedly, there is much corruption in the management of cooperatives. Some believe that managers are reluctant to sell the land because they would lose an important source of income. Moreover, as a politically strong body, cooperative members have been able to win concessions from the government to help keep the cooperative afloat. Partly as a result, cooperative members and managers are reluctant to sell their lands without being assured of an unrealistically high minimum price. 2.9 A third source of inefficiency stems from the lands owned by peasant communities. Although several peasant communities have shown interest in taking advantage of the 1995 Land Act and dividing their communal property into individual titles, this is not occurring for two reasons. First, the legal framework and procedures relating to the mechanism to issue individual titles has not been developed for the Peasant Communities in the Andes. In fact, government officials have turned down requests for registering titles coming from these communities. Second, the communities lack the technical and legal skills to transform themselves into individual, titled, property owners, which would allow them to obtain formal credit or to sell their lands to others with the capital and management capability to utilize the land productively. 2.10 For lands that remain to be titled, the government needs to accord this a priority and immediately resolve some problems which have slowed down the registration process. An IDB-financed project supports the process of land titling and registry. This project, conceived initially to title and register properties affected by the Land Reform, has extended its scope to all the coastal region. The project unit, called the Rural Program for Titling and Registration (PRTR), provides funds to both the rural titling and registration agency (PETT) in the Ministry of Agriculture and to the Public Registry (Registro Publico) under the Ministry of Justice. 12 2.11 Although PETT has unrealistic target of registering I million properties per year, it has titled and registered since 1993 only about 36,000 titles, mainly because of limited resources and organizational problems. While considerable funds have been budgeted to this activity, disbursement are proceeding very slowly because of procedural delays resulting from receiving approval from many institutions. Before PETT can purchase goods under the IDB-financed project, it must first receive the approval of the PRTR and then the approval of the Management Unit for International Cooperation (UGECI) located in the Ministry of Agriculture. The request is then cleared with the IDB and passed onto the United Nations Development Program (UNDP), which administers the fund, for final purchase. The process can take months. 2.12 To speed up the registration process, the government should consider removing UNDP from the process and transferring its administrative and accounting functions to UGECI. It would also help to clarify that UGECI's role is limited to administrative matters and monitoring activities rather than substantive matters, for which the Coordination Unit in the Ministry of Economy and Finance is better equipped. In addition, setting more realistic registration targets (say 150,000 properties/year), may help. Finally, a way to merge the Public Registry (Registro Publico) with the Land Registry (Registro Predial) needs to be found. The Registro Predial was set up when the Registro Publico lacked the capacity or willingness to accept faster titling and registration procedures. However, there is a danger that the same property can be titled in both registers under different names. 2.13 To realize the full potential of the land under sugar cooperatives, a politically acceptable way to transfer ownership of sugar cooperatives to other private investors needs to be found. To do so, the government should consider negotiating a minimum guaranteed land price with cooperative members to induce them to sell. The lands could then be auctioned to investors who would bring in new investment resources and management to turn the cooperatives into profitable companies. The difference between the auction price and actual sale price would be borne by the treasury. In addition, golden handshakes to current managers based on their years of service could be provided to obtain the support of these managers. The costs of this incentive has to be compared to the fiscal costs that the government could incur in terms of bailouts and loan guaranties and to the tax revenue that a profitable sugar company could bring in. 2.14 To deal with the issue of lands held by peasant communities, the government needs to specify more detailed and clear procedures for land titling and registration for both communal and individual land in the Sierra. Given the degree of property fragmentation, this process should rely significantly on peasant communities for the design and implementation. In addition, the government could create a new wing within PETT devoted to helping peasant communities obtain first communal and then individual land titles. Irrigation and Water Use 2.15 Although Peru's water resources are abundant, they are poorly distributed. Ninety-eight percent of the country's water is located in the Amazon basin, which has limited agricultural potential, while the fertile, low-rainfall coastal region is entirely dependent on water from over 53 rivers originating in the Andes. To exploit Peru's vast 13 water resources, successive governments have invested heavily in irrigation systems to store and divert water from Andean basins to coastal areas for use by agriculture, industry, and domestic consumption. The bulk (80 percent) of water is used for agriculture. Almost half of Peru's cultivated area, including all its coastal cultivation, is under irrigation. These areas account for the bulk of agricultural output. 2.16 Much of Peru's expenditures in irrigation are centered around large multi- purpose projects executed by the National Institute for Development (INADE). Nine such projects are still under execution along the coast. According to one study (Sucila and Gayoso 1993), these projects are estimated to cost $9.5 billion in 1993 dollars, of which $2.6 billion were spent through 1992. While some of the projects have significant hydropower potential, irrigation benefits have often been cited as the main reason for undertaking these projects. When completed, the nine projects are expected to bring 300,000 hectares under irrigation and to improve water delivery to an additional 285,000 hectares. Expenditures on these large schemes have historically dwarfed other public expenditures in agriculture. 2.17 Despite large and on-going investments in irrigation infrastructure, Peru's irrigation infrastructure is poorly maintained, with some public irrigation schemes in danger of collapse. Many schemes have irregular water delivery and suffer from water theft. Soil salinization affects an estimated 300,000 hectares of agricultural land. Moreover, even in areas where water is scarce, it is often used wastefully. Valuable water from costly irrigation schemes are used to grow low-value crops using inefficient irrigation technologies. Overwatering in the upper reaches of river is a major cause of salinization in downstream lands. 2.18 Although Peru's existing water law specified priorities for water use and mandated that tariffs cover operations and maintenance (O&M) and construction costs, the system of priorities has proven unworkable and water tariffs typically do not even cover O&M costs. Despite low water tariffs, many users are unwilling to pay them, partly because water delivery is irregular. The low tariffs and delinquencies, in turn, make it difficult to maintain the system adequately, which results in even more irregular water delivery, resulting in a vicious cycle. Thus raising water charges to levels that would reflect water scarcity and induce farmers to use water more efficiently is even less feasible. 2.19 Recognizing its inability or unwillingness to continue to subsidize O&M activities, the government passed laws in 1989, 1990, and 1991 turning over to water user associations (Juntas de Usuarios) the responsibility for operating and maintaining the irrigation infrastructure and for setting user charges. However, few water user associations had the technical capacity to carry out needed O&M activities. Moreover, the model of the water user association suffers from many shortcomings that make it difficult to operate large distribution systems effectively (Caruso 1997). Some of these are: (i) the inability of water users to access the highest level of management, (ii) the voting structure of one-person-one- vote which gives unbalanced voting power to small landowners, (iii) lack of participation of the majority of water users in the decision making process, (iv) duplication of administrative efforts by internal commissions (Comisiones de Regantes) and adjoining Juntas, and (v) lack of cohesion between adjacent Juntas on the same delivery system. 14 2.20 Providing technical assistance to water user associations to allow them to better operate and maintain systems and to organize themselves for this purpose should be a top government priority in this sub-sector. Many of the larger coastal schemes should consider using the Jequetepec water user association model, which could readily adapt to a more modern system of organization with the issuance of tradable water rights (see below) However, smaller systems, especially those in the Sierra, have no need for such a comprehensive system and they could continue to operate, as they have for centuries, through their peasant community structure. The most important operating change for water user associations, especially in the Sierra, is to revamp its system of water distribution from one based on water on demand to water distribution patterns that are contiguous, or concentrated in one area at a time. The present system causes huge system losses as a result of evaporation and seepage of water through unlined canals. 2.21 To address the issue of wasteful water use, awarding water rights to existing users without charge and allowing them to be traded offers the best incentive for conserving water and allocating it to more productive purposes. A draft law for this purpose was approved by Cabinet earlier this year but has been held up in Congress because of the perception that it may be politically divisive and because of some technical concerns, mainly relating to the way that rights would initially be assigned. 2.22 Whereas the Government is becoming wary of the proposed water law, farmers now appear to be support it. A survey is likely to reveal widespread farmer support for tradable water rights. In Chile a 1981 water law assigning tradable water rights has worked well for consumptive uses (agriculture, water supply and most industry) but because of defects in the way non-consumptive rights were defined and allocated, conflicts between consumptive and non-consumptive uses (for hydropower) have emerged, which has led some to shy away from a Chile-style law. By designing the law appropriately, Peru could enjoy the benefits of tradable water rights but avoid the mistakes present in the Chile law. A study suggesting such appropriate legislation was discussed with the government, water user groups and think tanks in 1995 (World Bank 1995). The government should thus consider passing the law. 2.23 As an interim measure, there is a proposal to introduce tradable water rights in one river basin (the Santa, which includes water for the Chavimochic and Chicama schemes) on a pilot basis. While the idea is plausible, the choice of the river basin and the draft regulations prepared by the Ministry of Agriculture make the proposal unsuitable as a pilot for water markets. The Santa river has a large water flow which presently goes unused into the ocean. There are unlikely to be water shortages (which are needed for water markets to function) until the Chavimochic is fully developed -- which could take years. Moreover, the strong public authority proposed in the draft regulations to allocate and distribute water goes against the grain of the draft water law. Also, as a pilot, this basin is not suitable because likely issues in the newly auctioned lands under Chavimochic will be quite different from those in established schemes. 2.24 If the government wishes to develop a pilot program, a better alternative would be to use some of the irrigation schemes being rehabilitated under the irrigation sub- sector project as pilots. In addition to technical assistance being provided to water user associations (juntas de usuarios), long-term funds are being made available to commercial 15 banks to lend to the juntas. However, banks are unwilling to assume the risks because of collateral issues (originally the project was designed to work in conjunction with tradable water rights as collateral) and because of the current weak financial cash flow position of most juntas. Given the state of deterioration of the irrigation systems and the high rates of return of many of these rehabilitation projects, the government may be forced to assume the non-payment risk. It may also find it necessary to pick up part of the additional expenses for those small farmers who will not be able to make the switch to higher value crops. Under this redesigned project, the government could pass a law and regulations allowing tradable water rights in a few (3 or 4) of these schemes. In this way, any difficulties in the design of the law or its regulations can be ironed out. By simultaneously also having similar schemes operating under the current law, a good controlled experiment of a typical real-world situation is possible. 2.25 As Table 1.2 shows, public expenditures on INADE's large irrigation projects dwarf other public expenditures in agriculture. The projects have often been under execution for decades with large cost overruns. At least some of these projects have doubtful economic benefit, even after large sunk costs. A study suggesting a strategy for the execution and financing of these projects deserves a close look (World Bank 1995). Under this strategy, the control and management of the projects, along with corresponding water rights and any government-held land, would be handed over to existing users or private investors. The government's role would be limited to that of financing components of the project that have significant positive externalities or are of a public goods nature and to providing any necessary guarantees. Such a scheme is preferable to the gradual sale of lands under the schemes. Forestry 2.26 Ninety percent of Peru's 75 million hectares of natural forests, covering 55 percent of its territory, are tropical heterogeneous forests located in the Amazon basin. This area is slowly but steadily getting deforested. Satellite observations between 1990 and 1995 indicate that 261,000 hectares of natural forest were lost every year during that period. Agricultural migration and illegal logging by poor peasants from the neighboring highlands, especially along new road developments, account for a large part of that loss. 2.27 Until very recently, reforestation efforts have averaged only about 10,000 hectares per year and are focused in the highlands, where many forested areas have almost disappeared due to intensive exploitation. One estimate puts the area in need of reforestation to be 10 millions hectares, 70 percent of which are in the highlands. Very recently, the National Program for the Management of Hydrological Basins and Conservation of Soils (PRONAMACHCS) has begun an aggressive reforestation program, announcing that 100 million trees will be planted in the Sierra in 1998. Because it is focused on poor communities in the Sierra and requires their active participation, it appears to be a well-designed and well-implemented program, though somewhat ambitious. 2.28 In conjunction with reforestation program in the Sierra the government should consider ways to control deforestation in the Selva. Enforcement of illegal deforestation is difficult because the deforestation areas are dispersed and because much of the deforestation involves burning of forests for agricultural purposes by poor farmers or 16 communities, mostly migrating from the poverty of the Sierra. Instead the government should control deforestation effects by promoting more research and technical assistance on the impact and potential of secondary forests as an economic resource. Secondary forests are those which reappear in deforested Amazon lands that are abandoned by peasants when their soil productivity drops to uneconomic levels. For example, in the Napo River Basin, peasants rotate their land to allow the soil to recuperate, thereby also limiting deforestation Management of these altered ecosystems (called "purmas") could encourage the rapid growth of secondary forests as well as provide peasants with a flow of economic resources, as has been done in Napo river basin. An adequate integrated management of cattle ranching, agriculture and forestry could be a sustainable way to limit migratory agriculture and their pernicious deforestation effects. 2.29 Peru has 25 million hectares reserved as protected areas. However, this still leaves a staggering 49 million hectares -- an area the size of Switzerland -- that could potentially be exploited for forestry. Yet Peru presently only produces only I million cubic meters of wood products annually and exports only $25 million worth of wood products (about $0.62 per hectare). In contrast Malaysia and Chile have ratios of $210/hectare and $185/hectare respectively. 2.30 The main reason for this is the 20 year limit placed on forestry exploitation contracts and permits under the 1975 Forestry and Fauna Law (DS 21147) still in effect. Exploitation contracts are granted for commercial logging and extraction permits to farmers and native communities (comunidades indigenas). Currently there are about 800 forest contracts covering 1.3 million hectares and about 220 permits covering 208,000 hectares. Only 42 contracts (5.5 percent) encompass areas greater than 1,000 hectares, the maximum land area which can be exploited without an approved forestry management plan. As these contracts are scattered all around the Amazon jungle, enforcement and control of illegal logging is extremely difficult. Granting small dispersed forestry permits without major requirements has in practice turned the forest into a free good. The level of deforestation per capita is very high in Peru -- equal to 0.010 hectares per capita per year as compared to 0.006 for Brazil, 0.004 for Latin America and the Caribbean, and 0.003 for the world. 2.31 As a first step to improve forest management in the country, the National Institute for Natural Resources, INRENA, has just published the 1995 Forest Map of Peru. Among other things, this will help identify the Permanent Forest Areas (ZFP) -- those that could be assigned to permanent forest exploitation under management plans that guarantee forest coverage. The first of such areas was established in the National Forest Biabo- Cordillera Azul (Supreme Decree 008-97-AG, May 21, 1997). Using public bidding, twenty lots covering 650,000 hectares will be auctioned to the private sector. ZFPs covering between 8 and 10 million hectares are expected to be established during 1998. Enforcement of the forestry plan will be facilitated by concentrating the ZFPs in smaller areas and by making the minimum lots to be auctioned large (over 3000 hectares). At the same time, current forestry concessions are also under review and their management plans are being updated. 2.32 To better develop forestry in Peru, there is an urgent need for a new forest law that gathers successful experiences in the promotion of the sustainable commercial forestry. An important component of such a law would be the possibility of long-term 17 concessions (for a minimum of 50 years, if not indefinite), subject to some management plan with low monitoring costs. Congress is now deliberating drafts of laws allowing long-term concessions. However, these laws do not pay enough attention to ensuring sustainable forestry at low monitoring cost. In addition, the laws are focused on forestry resources in the Selva. But there is potential for commercial forestry in the Sierra based on reforestation. Promoting reforestation in the highlands would be an effective way to reduce poverty in rural areas. As the Chilean experience has shown, long-term investors, such as private pension funds, may be interested in reforestation efforts if market conditions and incentives are appropriate. The new law should provide a framework that will encourage contracts between investors and peasant communities owing forests with commercial potential. Providing employment and income in the rural Sierra via these schemes will not only help alleviate poverty, it reduce migration to the Selva, thereby having a positive impact on deforestation and illegal logging in the Selva. 2.33 Even after the new law is passed, there will be many areas where reforestation may not be commercially viable but where, for social reasons, reforestation programs in poor communities is desirable. This suggests that PRONAMACHS could continue with its reforestation programs. However, it should ensure continuous coordination with the private sector in order to avoid duplication of efforts or crowd out the private sector. Grant funds from donors could support some of these social reforestation programs. Plant and Animal Health 2.34 Endemic pests and plant diseases in Peru substantially reduce the productivity of crops and livestock. The most important problems arise from (a) fruit flies, which are widespread along the coast, (b) citric fruit plagues such as the tristeza, (c) black sigatoka affecting bananas, (d) moliniasis affecting cocoa, (e) a variety of animal parasites, including ones which affect alpaca wool production, and (f) brucelosis affecting sheep. One-third of crop production is estimated to be lost to plant pests in the field; an additional 10 percent is lost to pests while in storage (such as the potato weevil). While farmers cope with the problem by using agrochemicals intensively, this raises the costs and over time pests develop tolerance to the chemicals. Also because of the perception, often correct, that Peru suffers from many plant and animal diseases, several of Peru's potential agricultural exports, especially fruits, are subject to trade barriers by importing countries. Finally, because Peru lacks adequate safeguards against pests in its imports, it is vulnerable to new pests and diseases. Addressing plant and animal health issues will substantially improve agricultural production and exports. 2.35 The Ministry of Agriculture has jurisdiction over the legal and regulatory framework for the use of agro-chemicals as well over the management of animal and plants health problems. Within this Ministry, this function is carried out by the National Service for Agricultural Health (SENASA). SENASA was created in November 1992 as a decentralized institution that replaced the Ministry's Directorate of Pest Control. Its main function is to improve animal and plant health in support of production, processing and domestic and international trade of agriculture. 18 2.36 Despite the importance of the issue, SENASA has received few resources from the treasury and relies on user fees for a substantial part of its operations For example, during the first semester of 1997 62 percent of SENASA's expenditures came from the fees it charged Given the lack of resources, SENASA's work has focused on vaccinations, dissemination and training, especially in poor communities in the highlands. In the future, SENASA plans to encourage greater private participation, especially in vaccination, pest control and laboratory analysis. 2.37 SENASA's financial situation and capacity for additional work would improve substantially with the implementation of a proposed loan from the IDB. The IDB- supported program has two key components: (i) institutional strengthening and (ii) pest and disease control, which is further divided into three sub-components: biological pest control, eradication of fruit flies, and animal health. With the recent signing of this project agreement, speedy implementation of the program is a possibility. 2.38 While strengthening SENASA both technically and financially is the top priority in this area, making sure that it remains independent of political pressures is also important. To be credible and effective, SENASA's recommendations need to be based purely on technical grounds. In particular, it needs to be more autonomous and resist efforts to use sanitary measures to restrict imports. A credible SENASA will acquire an international reputation, which will allow it to help domestic farmer associations negotiate with potential importing countries for removing unjustified sanitary restrictions on potential Peruvian agricultural exports. A measure that will help SENASA maintain its independence and be less susceptible to political pressure is to have SENASA overseen by an autonomous and independent Board, elected by both the government and the private sector. Currently, SENASA reports to the Minister of Agriculture and the tenure of its Executive Director is insecure. 2.39 A third area of reform is the legal and regulatory framework. Legislation on animal and plant health dates back to the turn of the century. Laws have been modified many times since then and the current legislation is complex, fragmented, and in some instances, contradictory. For example, to obtain a complete picture of health restrictions on imports of cotton, one has to review the health code published over 50 years ago as well as an additional six decrees since then. A new law which consolidates, updates and clarifies existing laws and regulations will greatly facilitate the proper enforcement of health standards while ensuring that such laws are not used as an unjustified non-tariff barrier. 2.40 Finally, in cases where there is a legitimate danger of pests or diseases being introduced into Peru through the import of certain goods, SENASA needs to work with INDECOPI to come to an agreement as to the most economical but effective way of doing so. SENASA is sometimes prone to reacting to any threat by using a blunt instrument as imposing a ban or effective ban on imports, which is often unlawful and unnecessarily costly to businesses and consumers. For example, to avoid the introduction of possible pests in cotton, SENASA initially required imported cotton to be fumigated in a vacuum chamber, only two of which exist in the world (in California and Egypt), instead of requiring another cheaper but equally effective method of fumigation. This particular measure was overturned by INDEPCOPI in 1994, as were two supreme decrees and three ministerial resolutions issued between 1991 and 1994. As INDECOPI's ability for overturning unjustified decrees, 19 regulations and resolutions has recently been serious curtailed, there is an even more pressing need for cooperation. Rural Finance Status of Peru 's Rural Financial System 2.41 Farmers cite access to credit on affordable terms an important constraint to increased agricultural production. Most finance for farmers is informal, typically done by middlemen (acopiadores). The term of the loan rarely exceeds a few months and effective annual interest rates are relatively high. Similarly, with very few exceptions, even formal finance is short-term. The lack of longer term loans for agriculture limits the ability of the sector to make the investments required to modernize farming and to compete in international markets. An efficient financial system could also make available attractive deposit services to rural residents. The resulting increased savings could provide some with a path out of poverty. An efficient financial market could also provide an increasing number of creditworthy people in rural areas -- both farmers and non-farmers -- assistance in expanding their businesses and incomes. 2.42 Annex 2 provides a recent history of rural finance in Peru, focusing on the Banco Agrario de Peru (BAP) and the reasons for its failure and subsequent closure in 1992. Recognizing that the private sector would only slowly fill the void left by BAP's closure, the government implemented several measures to help alleviate the credit needs of farmers. These measures: * liberalized restrictions on the sale of farm lands with the hope this would enhance their attractiveness as loan collateral; * provided credit for farm inputs, called FONDEAGROS, which were managed by the Ministry of Agriculture. Between 1992 and 1995 the government committed $230 million to these programs. As only 10 percent of these loans were repaid, the program was dissolved and remaining funds rolled into the Revolving Fund for Assistance to Agriculture (FRASA), which is administered by Peru's second-tier bank, COFIDE; * created a revolving fund call Fondos Rotatorios into which $92 million was placed between 1992 and 1995. This fund was also managed by the Ministry of Agriculture and provided loans in areas outside the coast at subsidized interest rates; * promoted through the Ministry of Agriculture the formation of private rural banks, called Cajas Rurales de Ahorro y Credito ( Cajas Rurales). 2.43 Five years after shutting down BAP, formal credit to the sector has resumed close to its levels during the eighties. The bulk of formal agricultural lending is done by commercial banks rather than the Cajas Rurales or via credit schemes of the Ministry of Agriculture. After being largely driven out of agricultural lending by BAP, combined with disruptions caused by land reform and rural violence, commercial banks have increased their lending to agriculture ten-fold from about $30 million in 1990 to almost $300 million in mid- 1997 as compared to an average of $379 million lent by BAP during the eighties. The 1997 figure includes loans to agro-industrial enterprises, some of which are on-lent to farmers. 20 2.44 Interviews with officials from five banks accounting for 85 percent of lending to agriculture suggest the following: * over one-third of their agricultural loans were made directly to farmers, almost all to coastal farmers owning more than 20 hectares of registered land; * most of their agricultural loans went to marketing intermediaries who were involved with handling agricultural commodities or inputs; * many of these borrowers, in turn, made informal loans to farmers; * many of the agricultural loans made by commercial banks were associated with exports and imports; * most of the loans were extended to firms that concentrated their activities in the coastal regions; and * most of the agricultural loans were for periods of less than one year. 2.45 While their expansion in agricultural lending is impressive, commercial banks are still far from providing the backbone of a rural financial system outside the tosta. Of the 884 bank offices in the country nearly 90 percent are located in the coastal region--60 percent in Lima alone--and only slightly more than 10 percent are located in the interior of the country. Commercial banks are remote from providing loans directly or indirectly to the bulk of the farmers in the country and they are doing little in the way of providing attractive deposit services outside cities. 2.46 The other main providers of formal agricultural credit are: Cajas Rurales, Cajas Municipales de Ahorro y Credito, and COFIDE. Unless the policy environment changes, none of these institutions are likely to form the basis for an efficient and sustainable rural financial system. Of the 18 Cajas Rurales with a total loan portfolio of $44 million in early 1997, two have since been liquidated. Most are undercapitalized, have difficulty mobilizing deposits, continue to depend heavily on "outside" funding, and are restricted in their scope and range of lending. About a quarter of them are still not profitable and some have serious loan recovery problems. In mid-1996 overdue loans comprised 16 percent of the total loan portfolio with three having rates in excess of 50 percent. The Cajas finance less than a quarter of their loans through deposited funds. 2.47 The past several years some Cajas Municipales have begun to expand into rural areas through opening new branches. In mid-1996 agricultural loans comprised 13 percent of all loans made by the twelve Cajas Municipales. It is encouraging that Cajas Municipales are seeing farm lending as a profitable business. These Cajas have developed techniques and practices that allow them to make a profit on relatively small loans. However, their deposit-to-loan ratios are only 0.6 to 0.7, well below expectations. This is likely to limit their continued expansion into agricultural lending. 2.48 COFIDE manages publicly-provided funds that often have loan targeting objectives. It is an improvement, nevertheless, over the old practice of placing targeted funds in the central bank and then passing them along to final lenders via concessionary lines of credit. This arrangement removes the Central Bank from promotional activities, lessens 21 the disincentives that outside money have on deposit mobilization, and limits the transaction costs associated with managing multiple lines of funding from outside the country. Even though COFIDE charges market rates of interest on its loans to other elements in the financial systems, such as the Cajas Rurales and the Cajas Municipales, its loans are usually cheaper for these organizations to use than it would be to mobilize similar amounts through voluntary small deposits. In part, the tepid deposit mobilization performance of both the Caja Rurales and the Cajas Municipales is due to their access to COFIDE funding. Since COFIDE does no retail lending, its credit activities have only an indirect effect on the supply of rural loans. COFIDE officials estimate that approximately US$20 million of its total of US$200 million in loans outstanding in mid-1997 went into agricultural loans. Policy Options 2.49 Despite progress in rebuilding the rural financial system in Peru since its low point in 1990, major gaps persist. These include a paucity of medium- and long-term loans for agriculture, limited credit for non-export enterprises, little formal lending done in the interior of the country, too few loans available for non-farm rural firms, extremely limited deposit services for rural people, and almost no lending that is done without real estate collateral. The government has two main options to deal with problems in rural financial markets. The first option is to pursue a policy of benign neglect based on the hope that unfettered private actions will eventually fill most of the gaps that currently exist in rural financial markets. The other option is for the government to take concerted actions to stimulate building of additional rural financial infrastructure, along with other activities that reduce the transaction costs and risks associated with rural finance. 2.50 Under the benign neglect option, commercial banks, Cajas Municipales, and possibly Cajas Rurales are likely to continue expanding their agricultural lending if other measures to improve the profitability of agriculture and availability of collateral are implemented (for example, land registration, plant and animal health, marketing). This expansion, in turn, will most certainly result in further expansion in the amount of informal lending done in rural areas, and at least some of this informal lending will be to the rural poor. This cascading of loans from formal to informal through marketing intermediaries, however, is unlikely to fill the major gaps in the rural financial market. Informal finance typically does not provide relatively large loans, medium- and long-term loans, or attractive deposit opportunities. It will similarly be impossible for semi-formal and formal forms of finance to provide loans directly to a much large number of farmers until more financial infrastructure is built in rural areas. Finally, an important concern with the benign neglect approach is that there is no formal mechanism to protect the rural financial system against recidivism to directed and subsidized credit such as via the Fondos Rotatorios or now defunct FONDEAGRO programs. 2.51 For these reasons it may be appropriate to explore ways to accelerate the building of an efficient rural financial system through the following four types of actions. First, an inter-agency committee for fostering the well-being and growth of these markets should be established. Peru's microcredit experience as well as experience elsewhere in the world suggest that a clearly-defined support group is able to foster financial markets and help maintain discipline. In addition to defending the interest of rural financial markets in political arena, the committee should also be responsible for formulating and carrying 22 out a development strategy for rural financial markets, including how to lower transaction costs and how to reduce lending risks. In addition to addressing transaction costs, the inter-agency committee could monitor and deflect government and regulator actions that boost transaction costs in rural markets. The committee would also be responsible for assuring that donor efforts were coordinated and that only practices and policies that enhanced the overall performance of rural financial markets would be allowed. This might include insisting on cost-recovery rates being charged on loans, reporting and evaluation requirements that do not substantially increase transaction costs in these markets, and pricing of funds such that deposit mobilization is encouraged. Fostering training and research on the development of efficient of rural financial markets could be another important function of the committee. 2.52 Second, the government might act to strengthen and possibly rebuild some financial institutions in rural areas. It is readily apparent that more bricks and mortar are needed to expand rural finance in Peru. Many of the villages and communities that once had a BAP office or a credit union remain without formal financial services, especially access to deposit services. Bilateral and multilateral assistance could help to build or rebuild some of this vital infrastructure, including Cajas Rurales and credit unions. 2.53 Third, the government could mount an organized effort to reduce transaction costs in rural finance. Rural finance is costly, partly because many of the transactions are small. All financial intermediaries are impeded from doing more rural finance by these costs. New practices and state-of-the-art electronics is needed to drive down these costs throughout the rural financial system. 2.54 Fourth, the government could support a parallel effort to reduce the risks of rural lending through legal and court reforms aimed at easing loan collateral problems. Understandably, lenders in Peru are highly risk averse. They prefer to lend short because of their experience with inflation, and lend against real estate collateral because of serious loan recovery problems during the 1980s and early 1990s. Resolving legal and court problems could help overcome some of their reluctance. Accelerated land and water titling and registration, changes in laws that would allow banks to accept a broader range of collateral--cattle, machinery, and inventory, for example--along with more efficient courts would be steps in easing collateral difficulties (World Bank 1997). Fostering leasing agencies may be another way of overcoming the lack of medium- and long-run lending for farmers in the near future. 2.55 In conclusion, government and development partner efforts to support rural finance under the new paradigm must be more subtle, innovative, and done with more discipline than under the old directed-credit paradigm. Ad hoc programs that insert additional outside money into the system will discourage deposit mobilization, increase transaction costs, and deflect rural financial markets from filling important gaps. Development partners must be particularly creative in repackaging their assistance to ensure that it does not weaken incentives that are critical in the development of rural financial markets. Development partners can best contribute to Peru's financial development by supporting the implementation of policy reforms and institutional and technical assistance rather than via lines of credit. 23 Research and Extension Current Situation 2.56 A good agricultural research and extension system should provide information on technology to permit farmers to increase agricultural productivity and production, thereby improving farmer well-being and reducing rural poverty. It should do so not only by focusing on increasing crop yields but by meeting farmer needs for postharvest management, marketing, processing and export. Moreover, it should help protect its natural resource base, particularly water and soils. These issues are particularly important for Peru as its recent high agricultural growth rates have been due almost entirely due to bringing increased land under cultivation -- its crop productivity is roughly the same as it was over a decade ago (Statistical Appendix Table 4.3 and Table 4.4). To sustain high agricultural growth rates, Peru will need to increase productivity, in part by adapting and disseminating importing technologies and products to local conditions, but also by generating original technical ideas, which Peruvian scientists can develop into novel production technologies to compete with other countries. 2.57 Peru's current research and extension system fails to meet the above criteria, especially as they relate to agricultural research. For one thing, little is spent on research. Most research is carried out by the National Institute for Agricultural Research, INIA, whose annual research expenditures are about US $5 million, covering the salaries and research material of about 100 agronomists in nine experiment stations. Its research program appears to be managed in the traditional paternalistic top-down manner, without sufficient regard to farmer needs. As a result it focuses almost entirely on raising crop yields for traditional crops and pays little attention to post-harvest needs. Limited agricultural research is also undertaken by universities, especially the agriculture university La Molina, as well as by some private firms. For example, innovations in the production, marketing and processing of crops such as asparagus and broccoli have been developed entirely by the private sector. Even so, total agriculture research expenditures in the country are likely to be only about $8 million per year (excluding the International Potato Center), about 0.16% of agriculture GDP, well below the levels of most countries in Latin America. 2.58 In 1994, following many years of neglect of the public research system, several of INIA's experiment stations were privatized. The government donated seven coastal stations to the Peru Foundation (Fundaci6n Perfi), a private non-profit organization comprising representatives from farmers' organizations, agro-industry, and other sectors. Fundaci6n Per6 has reestablished the productive capacity of its stations and does some demonstration type of extension on its farms. It also does very limited research on a contract basis using external expertise. However, it has not built a research capacity of its own; nor does it appear to want to do so in the near future. 2.59 The agricultural extension picture is better. Peru has a large number of extension programs being carried out by NGOs, private firms, and the government. Total expenditures on extension is estimated at $38 million per year, almost 0.8 percent of agriculture GDP, comparable to that of other countries in the region (Table 2.1). An estimated 250,000 farmers receive some form of technical assistance. However, this 24 includes 178,000 farmers receiving extension services of PRONAMACHCS that are largely restricted to soil conservation or tree planting as opposed to agriculture, per se. 2.60 Among the large number of agriculture extension programs the most innovative are those by the Fund for Agricultural Extension Services, FEAS, the Association of Exporters, ADEX, and PRONAMACHCS, all of which have demand-driven components in contrast to INIA's supply-driven programs. About half of FEAS' funds are deposited directly into bank accounts of peasant organizations, who use them for hiring technical assistance from private suppliers. The rest of its funds are used to supply materials and services to the communities involved, to administration, and to miscellaneous programs such as the operation of eight agricultural radio stations. Another desirable feature of the program is its requirement that each peasant organization begin to collect fees for extension services and deposit them in a savings account, which would fund technical assistance as FEAS withdraws over a four-year period. In place since 1993, these accounts now have a balance of $574,000. 2.61 The AID-funded ADEX program is also demand-driven in that it requires farmers to apply to them for technical assistance, which is subsequently provided through sub-contractors. Aimed at small farmers with an export potential, ADEX also provides market information and identifies promising new products on the basis of their market studies. In addition to providing technical assistance on production, it provides assistance on storage, enhancing quality, basic processing and marketing. It also helps the farmers obtain access to formal credit. 2.62 The novelty of the PRONAMACHCS program is that it links technical assistance to natural resource management, particularly soil conservation and reforestation. The program is restricted to the Sierra, where soil degradation and deforestation are serious problems. The program provides most materials required for soil conservation and tree planting as well as technical assistance. Unskilled labor is entirely contributed by the peasants without cash payment. Of its annual expenditures of about $50 million, approximately 70 percent are devoted to materials and 30 percent to technical assistance. Most resources are spent on soil conservation activities, but an increasing share is being used for tree planting -- about 65,000 hectares were planted with trees in 1997. 2.63 In contrast to the above demand-driven programs, INIA's conventional training and visit type of program is almost entirely supply-driven. The program focuses almost exclusively on increasing output of traditional crops, mainly potatoes and corn. Focusing on traditional crops directed to domestic or even local markets is risky. Most of these commodities face an inelastic demand. Expanding supply could drive down prices and farm incomes, especially for commodities subject to high transportation costs. 25 TABLE 2.1 EXTENSION PROGRAMS IN PERU: 1997 ESTIMATED EXPENDITURES AND BENEFICIARIES Expenditures (1997) Institution ($ million) No. of Beneficiaries INIA 5.0 10,000 FEAS 3.0 43,000 PRONAMACHCS 15.0 178,000 ADEX 8.0 8,000 Others 7.0 10,000 TOTAL 38.0 249,000 Total Excluding 23.0 71,000 PRONAMACHCS Source: Mission estimates based on publicly available data. * Includes extension programs of non-profit organizations such as Fundaci6n Per(6 and public agencies such of SENASA and INADE. Recommendations for Research 2.64 Research needs to respond to demands of farmers in each of Peru's three physical regions and to emphasize market potential, both international and domestic. The range of research providers (INIA, universities, NGOs, private firms, foundations, international organizations) can help do this. To make the most of this diversity, it is useful to develop a strategic plan and priorities that indicate the comparative advantages of each. In particular, INIA needs to formulate, in consultation with farmer groups, a plan which identifies priority area by region, research themes and the needs for strategic, applied, and adaptive research. Based on this the government can decide: (i) which kinds of research deserve direct government support, for example via co-funding mechanisms, direct contracting or execution, or via having research firms and universities compete for public funds on a competitive basis, (ii) which could be supported indirectly, for example via tax incentives; and (iii) which should be undertaken by the private sector without any support from the government. It can also identify areas where user fees could recover some of the costs associated with the research. 2.65 In general terms, the government should be involved primarily in funding research for farmers who are commercially viable but who are too small, poor, and dispersed to organize themselves to implement agriculture research. Additionally, public funding should be provided to finance research aimed at reducing externalities, particularly environmental degradation. In general, large farmers should be able to develop their own research capabilities and the state should mostly limit its involvement to providing an adequate legal framework that protects intellectual rights. Even here, public subsidies may be justified in areas where there are positive externalities associated with the private research effort. In this sense, the privatization of most experiment stations in the coastal area originally operated by INIA is consistent with the above criteria. INIA retained experiment stations mostly located in the Sierra and Selva, which is where farmers are generally small and relatively dispersed with little capacity to implement their own research. 26 2.66 However, INIA needs to be more responsive to farmers demands and should consider the market potential for the products rather than focusing just on yields. To do so, it will need to have greater consultation with farmers and listen better to their needs. Often this will involve addressing the entire agricultural chain from production to post-harvest storage and processing, to marketing, including to overseas markets. INIA also needs to exploit its comparative advantage stemming from being the point of contact with a large number of national and international researchers. It needs to increase its collaboration with such national and international firms, including for example, by adapting new knowledge generated in the Consultative Group for International Agricultural Research (CGIAR). In this way, INIA could become an intermediary between CGIAR and Peruvian farmers, especially small-sized farmers. 2.67 Implementing the above options will require considerably more public funds than is currently the case. In overall fiscal terms, however, the needed expenditures will be small. Even a doubling of INIA's research budget implies only an additional expenditure of $5 million. And the proposed system of competitive grants for universities and others could start with as little as $1 or $2 million a year. To carry out this process on a larger scale, the government should consider utilizing technical assistance or a research and extension project financed by a bilateral or multilateral development partner. Agricultural Extension: Recommendations 2.68 As with agricultural research, publicly-funded extension services should be oriented to farmers who are commercially viable but who do not yet have sufficient financial resources to pay for extension services. The extension program should be regarded as an economic investment, not as an element of the antipoverty strategy. Farmers who are not commercially viable should be targeted for social assistance but not for extension programs. This strategy would lead the government to concentrate on farmers that operate farms with an area of between 2 hectares and 10 hectares. Most farmers operating more than 10 hectares are sufficiently well-off to privately hire technical assistance by themselves. According to the 1994 agriculture census, there are 344,000 farm households that operate in the 2 to 10 hectare range. With a slight push, most of them should be able to increase their productivity and marketing skills. According to the 1994 census, the bulk of such farmers know how to read, are already using some modern purchased inputs, and belong to peasant organizations. Utilizing peasant communities to disseminate the technical assistance when possible, as done by FEAS, is desirable. 2.69 To ensure that the public funds are targeted and spent effectively, the technical assistance should be phased out over a 3-4 year period, as in the ADEX program. This will allow a broader distribution of public resources while ensuring that farmers who fail to generate sufficient income to pay for technical assistance even after receiving free or subsidized technical assistance for several years be dropped in favor of others with greater potential. In addition, the extension programs should be oriented to the needs of farmers. Service providers should listen to farmer needs and inform them of their options. In particular, they should inform them not only of ways to improve production, but also management assistance and marketing. Public extension agencies should not promote any particular products, but should inform farmers about the potential of different products. Also, implementation of a rigorous system of independent evaluation of the various 27 extension programs would be useful in determining the successes of the various programs and future budgetary allocations. Finally, the government needs to expend resources in better training of extension agents. Marketing 2.70 Soon after coming to office, the Fujimori Administration dismantled Peru's costly system of price supports, regulated consumer food prices, and government marketing monopolies, thereby improving the fiscal situation and promoting the development of private marketing channels. However, several government interventions in the output and input markets remain. The most significant are the interventions by PRONAA and ENC. While the size and fiscal costs of these interventions are small, they distort the market and have an adverse impact on product quality and market development. For example, PRONAA's purchases of rice, potato, and kiwicha are done at prices much higher than those paid by local intermediaries. Instead PRONAA should consider channeling its purchases through a more transparent mechanism like the recently created commodity exchange. The resulting savings could be used for providing technical assistance and small- scale infrastructure in poor communities. 2.71 Similarly, ENCI's intervention in the input market continues to be sizable. For example, during 1994-95, over one-third of urea imports were done by ENCI. Most of this fertilizer was donated to small poor farmers through several governmental programs. Although ENCI's imports have begun to decline, given the likely greater cost-efficiency of the private sector, there is little justification for the intervention of ENCI. The idea of liquidating ENCI should again be reviewed and the savings used to expand targeted poverty alleviation programs (Chapter 3) 2.72 Several studies done by GRADE have shown that the Peruvian agricultural marketing system operates quite inefficiently (for example, Cannock and Geng 1994, Escobal and Agreda 1997). Oligopsonies are not uncommon; farmers make planting decisions with limited information as to market demand and supply; and marketing losses as a result of inadequate storage facilities, especially for perishables, are high. The key explanatory factor for the inefficiencies is the restricted access to publicly provided goods such as road infrastructure and information, especially in Sierra and Selva. Another reason is the lack of formal finance. For most farmers, informal finance provided by middlemen (acopiadores) is the only source of credit. This credit often comes with conditions on the sale of output. 2.73 While the issue of rural finance was covered earlier, there are three areas where marketing could be improved. First, the government needs to improve its agricultural information system, which has already improved substantially since 1990. This is important in order to facilitate timely planting decisions by farmers. At present, such information is provided with a two to three month delay, too late for most planting decisions. What is needed instead is a monthly monitoring system which also provides a projection of the harvest. The implementation of an Agricultural Census in 1994, 22 years after the previous one, now makes it possible to speed up the process by allowing extrapolations from carefully selected samples. 28 2.74 Second, the Government needs to address inefficiencies in Lima's wholesale market. Today almost 40 percent of the value of all marketed agricultural products that are perishable pass through Lima's old wholesale marketing system. Established in 1945, this market traded all perishables until 1971, when a second nearby premise was made available for trading fruits. During the last 26 years, these markets have not grown in area although trading volumes have grown by over 50 percent. As a result, the "load" managed by the market is equivalent to more than 80 MT per square meter per year, much higher than international recommendations of 30 to 35 MT per square meter per year. This affects the overall efficiency of the market, making it very chaotic and congested. In addition, the markets lack adequate warehousing, refrigerating or packing facilities, which results in output losses and leads to oligopsonistic pricing behavior. 2.75 To address this situation, there is an urgent need to establish a new wholesale market in Lima. Such a project had been approved for World Bank financing in the mid- eighties, but was scrapped in 1987, when the Bank stopped disbursing funds to Peru because of payments arrears. The Municipality of Lima has recently called for international bids for the construction and private operation of a new wholesale market. Given the importance of this measure, the government needs to give its backing to the project and facilitate the issuing of any regulations. 2.76 The third development that should improve marketing is the establishment of a commodity exchange in Lima. In 1995 and 1996, the Government put in place the necessary legislation to establish and regulate private commodity exchanges (Bolsa de Productos). In 1997, the National Supervisory Commission for Companies and Securities, CONASEV, authorized the creation of the Lima Commodity Exchange (Bolsa de Productos de Lima) which included as members many large private firms involved in warehousing, insurance and marketing. In December 1997, the exchange is expected to trade various varieties of: cotton, rice, sugar, coffee, corn and fertilizer. In addition, a number of non- standardized crops will be traded on a pilot basis, including potatoes, carrots, garlic, onions, asparagus, tomatoes, beans, lentils, quinoa, marigold flour, cochinilla and amaranth. To improve the operation of this exchange, the Government needs to help CONASEV to develop a system warehouse certification, especially for semi-perishables such as rice and sugar. This could be a useful form of formal credit for some farmers. 29 3. COMBATING RURAL POVERTY: EVIDENCE AND POLICY Rural Poverty Profile 3.77 While Peru's poverty levels have fallen substantially since 1991, they are still very high. Moreover, the share of rural households living in poverty, at 65 percent, is higher than it was in the mid-1980s and much higher than the share of urban households (Table 1.1). In fact, the share of rural households living in extreme poverty is more than five times as high. Most rural poverty is concentrated in the Sierra, where 513,000 households live in poverty (Table 3.1). The Selva has the next highest number of rural households -- about 150,000 households -- while the Costa has only 110,000 households living in poverty. This suggests that any targeted poverty allevation strategy should focus on the Sierra and the Selva regions. Table 3.1: RURAL POVERTY BY PHYSICAL REGION (1994) Number of Rural Households (thousands) Poor Non-poor Total Sierra 513 348 861 Costa 110 90 200 Selva 149 89 238 Total 772 527 1299 Source: 1994 LSMS. 3.78 An analysis of the 1994 LSMS reveals interesting insights into the poverty profile. If rural households are divided into farmers, agricultural workers, and non- agricultural workers, depending on their primary source of income, agricultural workers tend to be the poorest, followed by farmers (Lopez and Maggiora 1997). Rural households who receive their primary source of income from activities other than agriculture have the lowest levels of poverty. Although the rural population derives most of its income from agriculture, non-agricultural activities contribute more than one third of the rural income. However, the poorer a household, the greater is its dependence on agricultural sources of income. 3.79 The level of schooling of Peru's rural population is high compared to other Latin American countries with similar incomes. The average years of formal schooling of the adult population is over six years and, at 93%, the rate of school enrolment among the 6 to 15 year-old population is extremely high. The returns to education are much higher in non-agricultural activities than in agriculture. In fact, formal education has no effect on the per capita income of agricultural workers and only a small effect on farmers, about $34 for every year of schooling, half that for non-agricultural workers. Although it is hard to fully 30 explain this phenomenon without more detailed studies, this could be because of the poor quality of rural education, particularly in the more isolated areas where agriculture is almost the sole economic activity, and because farm-related activities are typically not very skill demanding in Peru. 3.80 The dependency ratio (household size/working members) of rural households, at 2.3, is high compared to Latin American countries with similar income levels. In El Salvador, Colombia and Paraguay it is 1.8, 1.1, and 1.9 respectively (Lopez and Valdes, 1997). Moreover, there is a clear positive correlation between family size and dependency ratio, on the one hand, and poverty of the household on the other. In addition, the quantitative impact is large, suggesting that where supply factors limit the availability of contraceptives, reducing family size and the dependency ratio could dramatically decrease poverty in rural areas. 3.81 Though access to land is a significant income-enhancing factor for farm households, the quantitative impact of land on per capita income is rather modest (in contrast to its large effect on farm production). This implies that programs to redistribute land are likely to have limited effectiveness. To surpass the poverty line many poor households would need very large amounts of additional land. Using land redistribution as a means to alleviate poverty would require massive land transfers, making it a very costly poverty alleviation program. 3.82 Non-spanish-speaking households are much worse-off than spanish-speaking households, but this is largely explained by their location. Controlling for household characteristics, non-spanish-speaking households located in villages where both types of households coexist do just as well as spanish-speaking households. Moreover, the effectiveness of education in promoting income growth among non-spanish-speaking households is not less than among spanish-speaking households. This suggests that location-specific factors (weather, soil conditions, access to water and markets) rather than market discrimination is the reason for the lower incomes of non-spanish-speaking households. Poverty Alleviation Measures Traditional Programs 3.83 Until this decade, Peru's poverty alleviation efforts in rural areas focused on ways to get small farmers to increase agricultural production. As such, the programs ignored the landless poor as well as rural residents engaged in non-farm activities. The emphasis was on providing subsidized inputs, cheap directed credit, free technical assistance, and price supports to enhance farm production and incomes of poor farmers. These programs tended to be run by central line ministries, were essentially supply-driven, with little input from farmers on the design or implementation of the assistance, and were directed at individual farmers rather than rural communities. The 1969 Agrarian Reform went further by expropriating land and redistributing it to workers who lived on the expropriated farms. 3.84 Traditional programs have proven to be ineffective, costly, and unsustainable. The subsidized inputs and marketing subsidies tended to benefit the non-poor 31 disproportionately. The centrally-run government programs were costly to administer and often poorly designed; for example, they did not address the post-harvest and marketing concerns of farmers, let alone the landless and other non-agricultural workers. As a result the plight of the rural landless worsened, especially following the Agrarian Reform, while the rest of the economy paid a high price for the costly subsidies, dillminating in the hyper- inflation of the late eighties. Recent Programs 3.85 Recognizing the failure of the traditional programs, the Fujimori administration has introduced some innovative programs aimed at alleviating rural poverty. Their novel features are their reliance on communities and beneficiaries for the design and implementation of the programs. The most important of these programs is the National Fund for Compensation and Social Development (FONCODES), which provides communities with funds to implement economic and social infrastructure projects such as potable water supply systems, small irrigation systems, terracing, classrooms, health clinics, and street paving. In addition, some social assistance programs such as a school breakfast program, a family planning program and a literacy program are funded. The projects are proposed, designed and implemented by the communities themselves, often with technical assistance provided by the government or NGOs. Communities are expected to contribute to construction costs by providing unskilled labor without compensation. During 1995 and 1996, FONCODES disbursed $138 million and $113 million respectively, mostly in rural areas, and reached over 100,000 rural households annually. 3.86 PRONAMACHCS is another innovative program which seeks to work with poor rural communities to improve their standard of living. Its projects are restricted to natural resource management and it operates only in the Sierra, which is where most of the rural poor live and which is the region most affected by natural resource degradation. The program provides materials and technical assistance for soil conservation, for reforestation, and for the implementation of small rural infrastructure, including improvement of degraded pasturelands. Communities provide their labor and land. During 1995 and 1996 it spent $32 million and $50 million respectively. Of its budget of $50 million for 1997, 70 percent is expected to be used for materials provided to the communities free of charge and 30 percent for technical assistance and administration. For 1998, its budget has been increased to $65 million. The program reaches about 3,800 peasant organizations, or 25% of all communities in the Sierra. PRONAMACHCS staff estimate that their program has reached 178,000 households over the last 7 years. 3.87 By working through communal organizations, PAR provides assistance for the resettle ment of part of the estimated 600,000 people displaced by Peru's civil war against terrorism. With the subjugation of terrorism, many displaced people are returning to their homes. The program provides emergency assistance for the returning people, including food and basic services as well as assistance for repairing houses and other properties. It has components for building schools and health clinics, for providing technical assistance, and for providing materials for small productive improvements. Although the program was originally concentrated in the four departments most affected by the war (Ayacucho, Apurimac, Huancalvelica and Junin), it is beginning to expand into other areas. PAR's 32 budget for 1996 and 1997 has averaged just over $20 million annually and the program is estimated to have benefited over 200,000 people. 3.88 FEAS is a fourth participatory poverty-alleviation program focusing on rural areas in the Sierra. In addition to some institutional strengthening of farmer associations, the program allocates cash grants to poor farmer organizations on a competitive basis. The communities use these grants to buy technical assistance from private providers. Since its inception in 1993, FEAS has distributed $9.2 million to 545 farmer organizations and is estimated to have benefited 43,000 households. FEAS' assistance to each community is designed to be phased out over 4 years. 3.89 In addition to these participatory demand-driven programs, the government still operates some of its more traditional programs. The largest such program is PRONAA, which distributes food to poor families in urban and rural areas. In addition to distributing food to poor communities, it buys about half its food from poor farmers at pre-determined prices, which are often substantially higher than market prices. The program reaches about 9 million people, of which less than 1 million are rural. The PRONAA budget in 1996 was $93 million and purchases from small size farmers are estimated to have been about $40 million. In addition, it has smaller programs run by the MAG, such as Fondos Rotatorios, which provide inputs in kind tb farmers. 3.90 In summary, Peru gets high marks for the orientation of the bulk of its rural poverty alleviation programs. The programs correctly target poor households in the Sierra and Selva, require or foment a significant degree of community participation, are oriented to the needs of poor households, and often work through established rural communities and help strengthen them. With the exception of programs such as PRONAA and Fondos Rotatorios, they create few market distortions and rely on the private sector to a great extent (paras. 1.16 and 2.7). Moreover they have broad coverage and relatively small leakages into the non-poor population. An estimated 60 percent of the 772,000 poor rural households are likely to be reached by at least one of the above programs and 80 percent of the beneficiaries are, in fact, likely to be poor. Recommendations 3.91 Although the rural poverty alleviation programs are quite good and a vast improvement over earlier traditional programs, there are a five measures which could improve their effectiveness and widen its scope: (a) Follow a demand-driven principle: To help ensure that the assistance reflects the priority needs of communities and is implemented in a cost- effective manner, the active participation of communities in the design and implementation of the programs should be sought and communities should be expected to contribute a significant share of costs, particularly their own labor. Often technical assistance to strengthen communities or farmer associations will be necessary to allow this to occur. Most projects implemented by FONCODES, FEAS, and PRONAMACHCS meet this criterion. However, there are components of some of these programs, including some new productive programs of FONCODES, which could be 33 further oriented in this direction. In addition, the top-down programs of the Ministry of Agriculture need to be phased out or revamped to meet this criterion. In addition to concerns about their relevance and cost- effectiveness, these programs hamper the continued development of agricultural markets. (b) Improve inter-institutional coordination: With several programs operating in the same areas, coordination between the programs becomes an issue. For example, until recently, FONCODES and PRONAMACHCS sometimes had similar projects in the same area but whereas one required communities to provide their labor without compensation the other did not. This has now changed with both requiring labor contributions from the community. To identify and resolve such problems quickly, regular meetings between the various institutions are desirable. (c) Improve rural education: The counter-intuitive result on the low returns to education suggests that the quality of rural education is low. While several of the poverty alleviation programs emphasize investments in education, they are typically oriented to building or rehabilitating classrooms. More emphasis is needed on improving the quality of teachers (better training, monetary incentives to teachers to relocate in poor areas, etc.) and in encouraging parent participation in the school activities. In addition, increased attention should be given to improving technical skills and on-the-job training. (d) Expandfamilyplanningprograms and introduce social security: Given the strong impact of family size and the dependency ratio on poverty, more resources should be devoted to family planning. While the short-term emphasis has to be on contraceptives and education, in the longer-term, an old age security system is likely to help a lot. Because an important motivation for having children is old age security, a social security system can make an important contribution to lowering birth rates. In addition, a social security system is an important mechanism for reducing poverty of the elderly. By and large, existing programs are aimed at children or working adults and only indirectly affect the lives of the elderly. (e) Establish an evaluation system:. Guidelines for the periodic evaluation of the effectiveness of the many programs that deal with poverty should be established and funds to carry out this activity should be budgeted. The evaluation should be conducted by independent and reputed consulting firms and to the extent possible, funds for the evaluation should not be provided by the agencies responsible for the projects. References Cannock, G. and R Geng (1994). "Diagnostico sobre la Eficiencia de los Sistemas de Comercializaci6n" In: Escobal, J. (Ed.): Comercializaci6n Agricola en el Peru GRADE 33Op. Caruso, Charles (1997). Agricultural Wateruser Organizations in Per&, USDA, Mimeo Escobal, J. and V. Agreda (1997). Anilisis de la Competitividad en la Formaci6n de los Precios en el Mercado Mayorista No I "La Parada" Mimeo. GRADE. Lima L6pez Ram6n and della Maggiora, Carla ( 1997). Rural Poverty in Peru Stylized Facts and Analytics for Polic (Second Draft). University of Maryland at College Park, Maryland The World Bank (1992). Peru- Agricultural Policies for Economic Efficiency. Report No. 10605-PE, Washington DC The World Bank (1995). Per6: A User-Based Approach to Water Management and Irrigation Development, Report No. 13642-PE, Washington, DC The World Bank, (1997). Rural Poverty in Latin America. Report No. 16792-LAC, Washington, DC The World Bank, (1997). Peril: How Problems in the Framework for Secured Transactions Limit Access to Credit. Report No. 15696-PE, Washington, DC ANNEX I STATISTICAL APPENDIX Statistical Appendix ANNEX I: STATISTICAL APPENDIX 1 POPULATION 1.1 and Area by Natural and Political Regions, 1996 1.2 by Department, Selected Years 2. NATIONAL INCOME AND EMPLOYMENT 2.1 GDP by Expenditures (in 1986 prices) 2.1A Growth Rates 2.2 GDP by Sectors (in 1986 prices) 2.2A Growth Rates 2.2B Sectorial Shares 2.3 GDP by Sector (in current prices) 2.3A Sectorial Shares 2.4 Labor Force Structure by Sector, Selected Years 3. EXTERNAL SECTOR 3.1 Balance of Payments Summary 3.2 Value of Principal Agricultural Exports 3.3 Value of Principal Agricultural Imports 4. AGRICULTURAL OUTPUT 4.1 Value of Production of Major Agricultural Commodities 4. 1A Production Shares 4.2 Quantity of Production of Major Agricultural Commodities 4.2A Growth Rates 4.3 Yields of Major Crops 4.3A Growth Rates 4.4 Area Harvested by Crop 4.4A Growth Rates 5. LAND USE AND TENURE 5.1 Farm Size and Land Use (1994) 5.2 Number and Distribution of Plots by Region and Farm Size 5.3 Farm and Farm Size Distribution by Region According to Number of Plots Owned 11 Table 1.1 Population and Area, By Natural Region and Department, 1996 Natural Regions Political Costa Sierra Selva Total Regions Population Area Population Area Population Area Population Area (Thousands) (Km) (Thousands) (Kmr) (Thousands) (Km) (Thousands) (Km) Total 12,4039 137,134 8,5208 404,990 3,022.2 743,095 23,9469 1,285,219 Amazonas - 1050 7,262 271.3 31,987 3763 39,249 Ancash 419.8 9,402 6047 31,225 - - 1,0245 40,627 Apunmac - 4095 15,757 - 409 5 15,757 Arequipa 153.5 22,465 845.5 40,880 - - 999 0 63,345 Ayacucho - 479.5 48,501 38.3 453 517.8 48,954 Cajamarca 25.0 2,087 1,021.9 21,662 296.7 9,499 1,3436 33,248 Cusco - - 982.2 42,756 1214 29,136 1,103 6 71,892 Huancavelica - - 413.8 18,567 - - 4138 18,567 Huanuco - - 5739 17,270 143.8 14,867 717.7 32,137 Ica 601.6 18,946 6.1 2,382 - - 6077 21,328 Junin - 947.8 24,356 1854 23,589 1,133.2 47,945 La Libertad 956.9 10,329 408.8 15,241 - - 1,3657 25,570 Lambayeque 982.0 13,503 26.5 729 - - 1,008 5 14,232 Loreto - - - - 798.6 368,852 798.6 368,852 Madre de Dios - - - 74.1 85,183 741 85,183 Moquegua 100.9 5,392 36.9 10,342 - - 137.8 15,734 Pasco - - 177.4 7,011 66.2 18,309 243.6 25,320 Piura 1,290.7 29,411 176.8 6.481 - - 1,467.5 35,892 Puno - - 1.127.1 64,456 16.3 7,556 1,143 4 72,012 San Martin - - - 643.2 51,253 643.2 51,253 Tacna 217.4 7,768 28.7 8,295 - - 246 1 16,063 Tumbes 173.6 4,669 - - - 173.6 4,669 Ucayali - - - - 366.9 102,411 366.9 102,411 Source Instituto Nacional de Estadistica e Informatica - Dreccion Nacional de Censos y Encuestas Compiled by GRADE. 1.2 Table 1.2 Evolution of Population by Department, 1940-96 Census 1996 Region 1940 1961 1972 1981 to June 30 (June 9) (July 2) (June 4) (July 12) TOTAL 7,032 10,420 14,122 17,762 23,947 Amazonas 90 129 213 268 376 Ancash 465 606 755 854 1,025 Apurirnac 280 304 321 343 410 Arequipa 271 407 561 738 999 Ayacucho 414 430 479 524 518 Caiamarca 568 787 957 1,083 1,344 Callao 84 219 332 454 700 Cusco 565 648 7S2 875 1,104 Huancavelica 266 316 347 362 414 Huanuco 277 355 432 506 718 Ica 145 261 373 447 608 Junin 500 547 721 897 1,133 La Libertad 404 609 808 992 1,366 Lambayeque 200 354 S33 709 1,009 Lima 849 2,094 3,595 4,993 6,932 Loreto 321 411 541 475 799 Madre de Dios 25 25 2S 36 74 Moquegua 36 53 78 103 138 Pasco 151 185 231 244 Piura 432 692 888 1,155 1,468 Puno 645 727 813 910 1,143 San Martin 121 170 234 332 643 Tacna 38 68 100 148 246 Tumbes 27 57 79 108 174 Ucayali 219 367 Source: Instituto Nacional de Estadistica e Informatica - Direccion Nacional de Censos y Encuestas. Compiled by GRADE. Table 2.1 GDP BY EXPENDITURE, 1950-96 (Nuevos soles In constant 1986 prices) 1950 1960 1970 1975 1980 1985 1990 1991 1992 1993 1994 1995 1996 GDP 82,890 142,518 242,840 319,503 346,871 337,420 325,323 319,405 314,874 334,993 378,739 406,171 416,758 Private Consumption 59,801 92,870 175,596 218,324 221.087 219,092 240,501 243,248 239,525 249,739 267,326 284,721 300.228 Public Consumption 6,010 13,415 25,278 35,964 41,698 41,910 25,903 17,667 22,923 25,308 31,547 37,588 37,128 Gross Fixed Investments 15,932 28,726 42,955 87,373 89,738 67,928 72,626 57,535 57,319 64,475 83,565 96,324 89,431 Change In Stocks 23 2,749 271 312 8,125 2,407 (a) 6,858 5,931 7,194 7,643 8,334 8,477 Exports 11,132 24,613 36,583 36,910 48,978 50,567 39,726 31,046 32,647 37,244 43,889 46,750 50,039 Imports 10,008 19,855 37,879 59,380 62,755 44,475 53,433 36,950 43,471 48,968 55,231 67,546 68,543 GDP per-capita 10,859 14,351 18,404 21,074 20,056 17,378 15,096 14.521 14,086 14,731 16,374 17,260 17,403 Source: INEI Table 2.1a GROWTH RATES OF GDP BY EXPENDITURE, 1950-86 (Percent) 1950- 1960- 1970- 1980- 1990- 1960* 1970* 1980* 1990* 1996, 1985 1990 1991 1992 1993 1994 1995 1996 GDP 56 5.5 3.6 -0.6 42 17 -2.4 -1.8 -1.4 6.4 131 7.2 26 Private Consumption 4.5 66 23 0.8 3.8 0.3 4.4 1.1 -1.5 43 70 65 5.4 Public Consumption 84 6.5 51 -4.6 6.2 3.5 -18.8 -31.8 298 104 247 191 -1.2 Gross Fixed Investments 6.1 4.1 76 -2.1 3.5 -7.9 8.0 -20.8 -04 125 296 153 -72 Change in Stocks . . - - 898 - -13.5 213 62 90 1.7 Exports 8.3 40 30 -21 39 4.0 -157 -21.8 5.2 141 178 65 70 Imports 7.1 67 52 -1.6 4.2 -8.1 230 -308 176 126 128 223 15 Source: INEI Table 2.2 GDP BY SECTOR, 1950-1996 (Nuevos Soles In constant 1986 prices) 1950 1960 1970 1975 1980 1990 1991 1992 1993 1994 1995 1996 GDP 82,890 142,518 242,804 319,503 346,871 325,323 319,405 314,874 334,993 378,739 406,170 416,758 Agriculture 18,421 23,048 32,093 33,253 32,499 38,393 40,073 36,987 40,416 45,985 49,664 52,420 Fishing 174 1,186 3,918 1,290 1,566 3,461 3,107 3,508 4,245 5,474 4,430 4,655 Mining 1,810 4,290 6,346 6,401 12,941 9,496 8,746 8,519 9,212 9,620 9,833 10,089 Manufacturing 16,570 35,787 63,041 80,991 87,183 73,758 76,617 74,802 78,398 90,746 94,859 97,207 Construction 7,226 11,610 18,879 30,712 28,706 27,305 27,752 28,968 33,122 43,769 51,297 48,937 Services 38,689 66,597 118,527 166,856 183,976 172,910 163,109 162,091 169,600 183,146 196,086 203,449 Source INEI Table 2.2a GROWTH RATES OF GOP BY SECTOR, 1950-96 (Percent) 1950- 1960- 1970- 1980- 1990- 1960* 1970* 19800 1990* 1996* 1990 1991 1992 1993 1994 1995 1996 GDP 5.6 5.5 3.6 -0.6 4.2 -2.4 -1.8 -1.4 64 13.1 72 26 Agriculture 2.3 3.4 0.1 1.7 5.3 -9.8 4.4 -77 9.3 13.8 80 55 Fishing 21.2 12.7 -8.8 8.3 5.1 -3.8 -10.2 12.9 21.0 29.0 -191 51 Mining 9.0 4.0 7.4 -3.0 1.0 -2.7 -7.9 -26 8.1 44 22 26 Manufacturing 80 5.8 3.3 -1.7 4.7 -2.1 3.9 -2.4 4.8 15.8 45 25 Construction 4.9 50 4.3 -0.5 10.2 3.3 16 4.4 14.3 32.1 172 -46 Services 56 5.9 4.5 -0.6 2.7 -1.5 -5.7 -0.6 4.6 80 7.1 38 Source: INEI Table 2.2b SECTORAL SHARES OF GOP IN CONSTANT PRICES, 1950-1990 (Nuevos Soles In constant 1986 prices) 1950 1960 1970 1975 198b 1990 1991 1992 1993 1994 1995 1996 GDP 100 100 100 100 100 100 100 100 100 100 100 100 Agriculture 22.2 16.2 13.2 10.4 9.4 11.8 12.5 11.7 121 12.1 122 126 Fishing 0.2 0.8 1.6 0.4 0.5 1.1 1.0 1.1 1.3 1.4 1 1 1.1 Mining 2.2 3.0 2.6 2.0 3.7 29 2.7 2.7 27 25 2.4 2.4 Manufacturing 20.0 25.1 26.0 25.3 25.1 22.7 24.0 238 234 240 234 233 Construction 8.7 8.1 7.8 9.6 8.3 8.4 8.7 92 99 116 126 11.7 Services 46.7 46.7 48.8 52.2 53.0 532 51.1 51.5 50.6 48 4 48 3 48.8 Source: INEI Table 2.3 GDP BY SECTOR AT CURRENT PRICES, 1950-96 1950 1960 1970 1975 1980 1985 1990 1991 1992 1993 1994 1995 1998 (Thousands of Intis at Current Prices) (Nueves Sols at Current Prices) (Thousands of Nuevos Soles at Current Prices) GDP 18,278 67,519 279,969 665,389 5,967 197,903 8,564,352 32,937,328 52,170,385 80,010,143 109,315,764 132,598,960 149,780,380 Agriculture 5,210 14,258 46,692 103,685 581 17,434 377,763 2,365,316 3,374,543 5,255,895 7,090,078 8,138,921 9,543,241 Fishing 35 577 5,584 5,256 29 1,199 16,271 220,872 445,884 739,841 1,130,998 1,036,443 1,221,301 Mining 728 3,425 13,716 22,255 904 19,513 143,185 597,586 970,668 1,418,994 1,905,757 2,289,958 2,514,943 Manufacturing 2,842 12,632 55,359 133,252 1,206 48,066 1,424,548 7,880,048 12,006,017 18,229,848 25,151,937 30,149,481 33,189,197 Construction 864 3,252 17,345 50,798 342 14,087 627,234 2,601,845 4,507,420 7,667,290 12,280,835 16,249,837 16,605,017 Services 8,599 33,375 141,273 350,143 2,905 97,604 5,975,350 19,271,662 30,865,855 46,698,275 61,756,160 74,734,322 86,706,680 Source: INEI Table 2.3a SECTORAL SHARES OF GDP IN CURRENT PRICES, 1950-96 1950 1960 1970 1975 1980 1985 1990 1991 1992 1993 1994 1995 1996 GDP 100.0 1000 100.0 100.0 100.0 100.0 1000 100.0 100.0 100.0 1000 1000 1000 Agriculture 28.5 21.1 16.7 15.6 9.7 8.8 4.4 7.2 6.5 6.6 65 6.1 64 Fishing 0.2 0.9 20 0.8 0.5 0.6 0.2 07 0.9 09 10 08 08 Mining 4.0 5.1 4.9 33 15.1 99 17 1.8 1.9 18 17 17 1.7 Manufacturing 15.5 18.7 19.8 20.0 20.2 24.3 166 239 230 228 230 227 222 Construction 4.7 4.8 6.2 76 57 7.1 7.3 7.9 86 96 112 12.3 11.1 Services 47.0 49.4 505 52.6 48.7 49.3 698 585 592 584 565 564 579 Source: INEI Table 2.4 Labor Force Structure by Sector, 1970, 1980 and 1990 Economic Sector 1970 1980 1990 1996 (1,000) Percent (1,000) Percent (1,000) Percent (1,000) Percent Total 4191 100.0 5587 100.0 6748 100.0 7490 100 Agriculture 2016 48.1 2223 39.8 2665 39.5 2333 31.1 Mining 59 1.4 112 2.0 135 2.0 89 1.2 Industry 528 12.6 648 11.6 756 11.2 962 12.8 Electricity 8 0.2 17 0.3 20 0.3 23 0.3 Construction 176 4.2 218 3.9 263 3.9 313 4.2 Commerce 478 11.4 732 13.1 897 13.3 1432 19.1 Transport 172 4.1 246 4.4 290 4.3 426 5.7 Financial Institution 59 1.4 140 2.5 162 2.4 69 0.9 Services 695 16.6 1251 22.4 1559 23.1 1843 24.6 Source: INEI 31 TABLE 3.1 BALANCE OF PAYMENTS SUMMARY (Mililons of U.S. dollars) 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 I. CURRENT ACCOUNT BALANCE -2065 -1819 -570 -1384 -1510 -2101 -2302 -2662 -4298 -3605 1. Trade balance -500 -134 1246 399 -189 -341 -607 -997 -2170 -2000 a Exports f.o.b. 2715 2731 3533 3321 3406 3661 3516 4598 5591 5897 b. Imports f.o b. -3215 -2865 -2287 -2922 -3595 -4002 -4123 -5596 -7761 -7897 2. Services -371 -333 -307 -365 -413 -575 -575 -502 -773 -680 a. Exports 793 831 836 799 826 836 837 1064 1131 1371 b Imports -1164 -1164 -1143 -1164 -1239 -1411 -1412 -1565 -1904 -2050 3. Income -1312 -1515 -1685 -1733 -1367 -1632 -1615 -1800 -1997 -1573 a. Private -145 -117 -76 -35 -51 -226 -239 -353 -1402 -896 b Public -1168 -1398 -1609 -1698 -1316 -1406 -1376 -1447 -595 -677 4. CurrentTransfers 118 163 176 316 459 446 495 637 642 647 II. FINANCIAL ACCOUNT -1292 -1746 -1249 -699 -61 912 1865 3908 3135 3689 1. Private Sector 3 4 49 41 108 206 1204 3863 2643 4130 2. Public Sector -886 -1121 -807 -1022 -126 -381 497 -337 -141 -414 3. Short-term capital -409 -629 -491 282 -43 1087 164 382 633 -27 III. EXCEPTIONAL FINANCING 2181 2529 2476 2491 1390 1490 629 1611 1504 922 1. Reprogamming 19 28 736 25 5529 691 1313 705 676 586 2DebtForgiveness 0 0 0 0 12 53 7 138 20 60 3Netarrearsflow 2162 2501 1740 2466 -4151 746 -691 768 808 276 IV. BCRP NET INTERNATIONAL -950 -1877 RESERVE FLOW 1148 590 -540 -176 -837 -695 -724 -3032 (Negative sign indicates increment) 1. Change of NIR stock 915 395 -709 -174 -773 -697 -741 -2976 -923 -1899 2.Gold valuation and monetization effec -233 -195 -169 2 64 -2 -17 56 27 -22 IV. NET ERRORS AND OMISSIONS 28 446 -116 -231 1018 394 532 175 610 872 Note Direct investment from privatization 0 0 0 0 0 6 168 2241 547 1688 Source Banco Central de Reserva del Per6 3.2 Table 3.2 Value of Principal Agricultural Exports, 1970-1996 (Millions of US$ FOB) Product 1975 1980 1985 1990 1991 1992 1993 1994 1995 1996 Traditional Agricultural Exports 385.8 227.0 225.0 182.7 194.8 114.4 80.6 223.9 338.1 290.5 Cotton 59.8 71.8 51.0 48.2 57.4 23.1 4.6 4.9 24.3 30.2 Coffee 51.9 142.2 151.0 98.3 104.3 68.9 56.2 186.3 283.8 223.1 Sugar 269.1 13.1 23.0 36.2 33.1 22.3 19.9 32.7 30.0 37.2 Non-Traditional 10.8 40.9 63.7 93.4 112.0 140.6 157.3 182.4 227.3 331.5 Cocoa 0.1 14.8 27.0 13.0 14.0 10.5 11.5 14.1 21.6 19.9 Asparagus 0.6 3.3 5.9 30.9 44.3 62.2 74.0 79.3 99.7 130.4 Brazil Nuts 0.9 2.7 3.3 4.1 3.1 2.1 2.3 3.5 4.3 5.9 Marigold a/ a/ 5.0 6.6 7.7 16.2 19.2 19.3 13.8 14.8 Cochinilla 2.9 3.5 7.1 3.5 5.8 2.6 3.5 2.3 2.7 12.9 Mangoes a/ a/ 1.7 2.3 1.7 6.4 4.9 6.6 6.3 11.3 Others 6.3 16.6 13.6 33.0 35.5 40.5 41.8 57.4 79.0 136.2 Total Agricultural Exports 396.6 267.9 288.7 276.0 306.8 255.0 237.9 406.3 565.4 622.0 Total Exports 1,330.0 3,916.0 2,978.0 3,321.0 3,406.0 3,661.0 3,516.0 4,598.0 5,591.0 5,897.0 Traditional as Share of Total Agricultural Exports b/ 97.3 84.7 77.9 66.2 63.5 44.9 33.9 55.1 59.8 46.7 Agricultural Shares of Total Exports 29.8 6.8 9.7 8.3 9.0 7.0 6.8 8.8 10.1 10.5 Table 3.3 VALUE OF PRINCIPAL AGRICULTURAL IMPORTS, 1970-1996, Selected Years (Millions of US$ FOB) 1970 1975 1980 1985 1990 1992 1994 1996 RICE 1.8 33.4 85.4 0 93.8 115.9 642 1423 WHEAT 32 137.2 141.1 99.4 881 118.3 1459 2382 SUGAR 0 0 31.8 0 54.0 63.5 944 104.7 DAIRY PRODUCTS Powdered Whole Milk 0 0 8 5643 219 336 551 752 Powdered Skim Milk 2.9 25 233 8.8 18.5 142 170 197 Milk Fat 1.7 144 16.2 68 7.4 49 43 50 SOYBEANS 0.6 7.4 0 25 00 0.1 05 25 SOYBEAN OIL 4.9 39.5 23.5 192 23.4 32 7 72 7 525 BARLEY 0.8 6 7.1 89 96 90 71 120 MALT 0.4 12.3 11.1 9 121 127 126 16.0 LAMB 2.7 1.2 1.7 2.9 34 0.5 04 04 BEEF 22 5 5.6 6.4 7.3 4.5 116 115 TRIPE 0.7 1.2 2.2 1.8 4.1 3.6 4 7 4.8 CORN 0.2 52.4 67.3 28.1 60.5 82.9 81 6 123.0 SORGHUM 0 4.8 0.3 0.1 0.2 2.2 2.9 TORTA DE SOYA 0 0 12.2 108 28.8 298 343 546 POTATOES 0 0 0.3 0 0.0 2.0 0.3 00 CHEMICAL INPUTS 7.4 27.9 27 9 14 5 33 3 37.4 526 83.1 UREA 4.1 58 0.1 0 216 24.7 250 458 AMNONIA NITRATE 0.1 4.9 35 0 10 0.3 00 06 AMMONIA SULFATE 0.7 2.2 2 1.5 0.0 1.5 07 33 CALCIUM PHOSPHATE 0 3.7 28 19 1.7 1.7 51 23 POTASSIUM CHLORIDE 0.2 1.1 2.2 12 02 1.0 44 19 POTASSIUM SULFATE 0.2 0 09 0.6 00 0.2 09 27 MAGNESIUM SULFATE 0 0 0.1 01 00 00 01 06 HERBICIDES 04 6.3 2.9 1.1 18 1.9 42 49 NATURAL RUBBER 1.7 3.9 88 4.9 42 3.1 71 100 AMMONIA PHOSPHATE 0 0 4.6 32 28 30 52 112 TOTAL AGRICULTURAL IMPORTS 71 340 437 210 433 530 607 865 TOTAL IMPORTS 700 2427 3090 1806 2922 4002 5596 7897 AGRICULTURAL SHARE OF TOTAL IMPORTS (perce 101 14.0 14.1 11.6 148 133 108 110 Source, Ministerio de Agricultura Banco Central de Reserva Table 4.1 VALUE OF PRODUCTION OF MAJOR AGRICULTURAL COMMODITIES (Nuevos Soles at 1979 prices) 1970 1975 1980 1985 1990 1991 1992 1993 1994 1995 1996 RICE 294 26.8 22.0 44.2 47.9 409 41.4 48.3 699 569 600 WHITE CORN 12.3 11.3 8.8 10.5 7.3 111 6.3 91 92 11.4 123 BEANS 4.5 4.2 4.0 4.0 38 4.0 3.6 42 46 5.0 5.9 POTATOES 50.4 43.0 39.6 40.9 30.3 38.5 26.3 391 48.3 594 60.5 WHEAT 5.8 6.0 3.5 4.2 4.6 62 3.5 5.1 6.0 59 6.9 COTTON 320 29.6 34.8 37.0 30.7 23.0 14.1 12.8 21 9 28.3 35.1 QUINOA 0.3 0.4 0.8 0.5 0.2 08 02 0.7 0.9 0.7 08 BARLEY 6.3 5.6 4.5 3.9 2.7 44 2.6 4.2 4.8 4 9 5.7 YELLOW CORN 13.6 14.8 11.2 17.3 17.1 16.3 13.8 20.6 18.9 172 19.7 SUGAR CANE 241 28.4 17.8 23.3 18.8 18.5 150 138 17.2 200 194 SOYBEANS 0.0 0.1 0.8 02 0.2 01 00 0.1 0.1 0.1 0.2 SORGHUM 0.4 09 1.1 0.7 04 0.9 0.1 0.5 0.1 01 01 COFFEE 31.6 30.3 42.5 44.8 41.5 38.3 40.1 397 42.4 449 494 CHICKEN 12.5 282 31.1 43.6 53.2 60.8 69.5 65.7 76.8 89.1 891 LAMB 4.0 3.6 3.5 2.9 4.0 3.2 3.3 3.1 31 32 35 PORK 8.8 10.4 10.5 10.2 12.6 130 13.8 144 148 15.2 15.8 BEEF 13.2 134 13.0 15.8 18.2 17.0 172 166 15.8 167 17.1 EGGS 6.8 123 14.7 19.1 24.2 28.7 26.2 262 28.4 353 31.9 MILK 36.3 35.8 34.3 35.6 34.1 34.3 338 35.3 36.5 37.7 39.8 TOTAL 292.3 304.9 298A 358.5 352.1 360.2 330.7 359.6 419.8 452.2 473.2 Source: Ministerio de Agricultura Table 4.ia PRODUCTION SHARE OF MAJOR AGRICULTURAL COMMODITIES (percent) 1970 1975 1980 1985 1990 1991 1992 1993 1994 1995 1996 RICE 10.1 8.8 7.4 12.3 13.6 11.3 125 134 16.7 126 12.7 WHITE CORN 4.2 3.7 3.0 2.9 2.1 3.1 1.9 25 22 25 26 BEANS 1.5 1.4 1.3 1.1 11 1.1 1.1 1.2 11 1.1 13 POTATOES 17.2 141 13.3 11.4 86 107 8.0 109 11.5 131 12.8 WHEAT 20 2.0 1.2 12 13 1.7 1.0 1.4 14 13 15 COTTON 10.9 9.7 11.7 103 8.7 64 4.3 3.6 52 6.3 7.4 QUINOA 01 0.1 0.3 0.1 01 02 0.1 02 02 02 02 BARLEY 22 1.8 1.5 11 08 12 0.8 12 1.2 11 12 YELLOWCORN 46 4.9 37 4.8 4.9 45 42 57 45 38 42 SUGARCANE 82 93 60 6.5 53 51 45 38 41 44 41 SOYBEANS 0.0 0.0 0.3 0.0 0.1 00 0.0 00 00 00 0.0 SORGHUM 0.1 0.3 0.4 0.2 0.1 0.3 0.0 0.1 0.0 0.0 00 COFFEE 10.8 9.9 14.2 12.5 11.8 10.6 121 11.0 101 99 10.4 CHICKEN 4.3 92 10.4 122 15.1 169 210 183 18.3 19.7 18.8 LAMB 14 12 12 08 1.1 09 1.0 09 0.7 07 0.7 PORK 3.0 3.4 3.5 2.8 3.6 36 4.2 4.0 3.5 34 33 BEEF 4.5 4.4 4.4 4.4 5.2 47 5.2 46 38 37 36 EGGS 2.3 4.0 49 53 6.9 8.0 7.9 7.3 68 7.8 67 MILK 12.4 11.7 11.5 9.9 9.7 9.5 10.2 9.8 87 83 84 Source: Ministerio de Agricultura Table 4.2 PRODUCTION OF MAJOR AGRICULTURAL COMMODITIES (Thousands of metric tones) 1970 1975 1980 1985 1990 1991 1992 1993 1994 1995 1996 RICE 590 537 441 886 961 819 829 968 1401 1140 1203 WHITE CORN 251 230 180 215 150 227 128 186 188 233 251 BEANS 53 49 46 46 45 47 41 49 54 59 69 POTATOES 1922 1640 1513 1559 1157 1469 1003 1493 1843 2265 2309 WHEAT 123 126 74 89 98 130 73 108 127 125 146 COTTON 245 227 266 284 235 176 108 98 168 217 269 OUINOA 7 8 15 10 4 15 4 14 17 14 16 BARLEY 170 149 121 103 73 117 69 113 130 131 153 YELLOW CORN 385 421 317 491 486 462 392 586 537 489 560 SUGAR CANE 7589 8958 5599 7328 5929 5842 4741 4343 5429 6306 6119 SOYBEANS 0.4 1.5 10.9 22 25 14 0.3 10 08 1.4 22 SORGHUM 12 29 36 23 13.0 31.8 36 16.6 41 43 29 COFFEE 68 65 92 96 90 83 86 86 92 97 107 CHICKEN 58 130 144 201 245 280 320 303 354 411 411 LAMB 24 21 21 17 24 19 19 19 18 19 20 PORK 47 55 55 54 67 69 73 76 78 80 83 BEEF 85 86 84 101 117 110 111 107 102 107 110 EGGS 28 50 60 78 99 117 107 107 116 144 130 MILK 825 813 780 809 776 780 768 803 830 858 905 Source: Ministerio de Agricultura Table 4.2a GROWTH RATES OF PRODUCTION OF MAJOR AGRICULTURAL COMMODITIES (Percent) 1970-80* 1980-90* 1990-96* 1985 1990 1991 1992 1993 1994 1995 1996 RICE -2.9 8.1 3.8 -21.6 -11.7 -14.8 12 167 448 -18.6 5.5 WHITE CORN -3.3 -18 9.0 0.2 -34.6 51.8 -437 455 12 238 75 BEANS -12 -0.4 7.6 -63 -212 44 -108 182 91 98 175 POTATOES -24 -2.6 12.2 7.2 -31.8 270 -317 488 235 229 19 WHEAT -49 2.9 6.9 4.7 -40.5 330 -439 480 175 -16 169 COTTON 08 -1.2 2.2 41.6 -27.5 -250 -388 -94 nd nd nd QUINOA 85 -134 28.8 -241 -732 3383 -74.3 255.9 17.9 -16.0 152 BARLEY 00 -4.9 13.1 -4 9 -38.7 60.4 -41.2 634 154 1.1 164 YELLOW CORN -19 4.4 2.4 -14.1 -37.1 -50 -152 496 -85 -89 145 SUGAR CANE -30 0.6 0.5 49 -64 -1.5 -18.8 -84 25.0 161 -30 SOYBEANS 391 -13.6 -2.2 227 -100 -46.4 -77.5 2166 -153 693 600 SORGHUM . 115 -98 -220 -47.1 -64.4 1452 -887 361.2 -75.2 33 -316 COFFEE 30 -0.2 2.9 22.2 -214 -7.6 46 -1.0 nd nd nd CHICKEN 95 55 9.0 107 193 144 14.2 -55 16.8 161 -01 LAMB -13 1.4 -2.6 -90 190 -20.6 26 -46 -27 50 74 PORK 1.7 1.9 38 -18 -94 30 6.3 40 26 3.1 36 BEEF -0.1 3.4 -1.0 -1.6 47 -6.5 11 -3.7 -46 53 28 EGGS 80 51 47 197 2.9 187 -88 03 83 244 -98 MILK -0.6 -01 2.6 36 -33 0.5 -15 45 34 3.3 55 Source Ministeno de Agrcutura Table 4.3 YIELDS OF MAJOR CROPS (Metric ToneslHectares) 1970 1975 1980 1985 1990 1991 1992 1993 1994 1995 1996 RICE 4.2 4.4 4.3 4.6 52 5.1 5.0 5.5 5.9 5.6 5.7 WHITE CORN 1.1 1.0 1.0 1.1 1.0 1.2 0.9 1.1 1.1 1.1 1.2 BEANS 0.8 0.8 0.8 0.8 0.8 0.9 0.9 0.9 0.9 0.9 0.9 POTATOES 6.1 6.5 7.2 8.4 7.9 7.9 7.4 8.4 9.8 9.4 10.1 WHEAT 0.9 0.9 0.9 1.1 1.2 1.2 1.0 1.3 1.2 1.3 1.2 COTTON 1.7 1.7 1.8 1.8 1.7 1.5 1.3 1.5 n.a. n.a. n.a. QUINOA 0.4 0.6 0.8 0.6 0.4 0.7 0.5 0.8 0.8 0.7 0.9 BARLEY 0.9 0.9 1.1 0.9 1.0 1.0 0.8 1.1 1.2 1.1 1.2 YELLOW CORN 2.5 2.9 2.4 2.8 2.8 2.9 2.9 3.1 3.1 3.0 3.0 SOYBEANS 1.0 1.4 1.7 1.8 2.1 1.7 1.3 1.6 1.3 1.4 1.6 SORGHUM 3.4 3.0 3.0 3.2 2.7 3.1 2.6 4.6 4.2 4.3 5.2 SUGAR CANE 156.8 162.0 113.8 138.0 122.5 109.2 99.3 91.2 106.4 105.8 112.5 COFFEE 0.6 0.5 0.6 0.6 0.5 0.5 0.5 0.5 n.a. n.a. n.a. Source: Ministero de Agricultura Table 4.3a GROWTH RATES OF YIELDS OF MAJOR CROPS (Percent) 1970-80 1980-90 1990-96 1985 1990 1991 1992 1993 1994 1995 1996 RICE 0.7 2.0 1.6 -5.9 2.2 -3.2 -3.1 9.4 7.4 -3.9 1.8 WHITE CORN -0.5 0.3 2.5 0.2 -6.6 16.4 -20.2 18.3 -1.6 5.6 2.4 BEANS 0.4 0.1 2.2 -5.8 5.1 9.2 -3.7 2.6 -0.3 5.7 -4.1 POTATOES 1.6 0.9 4.1 1.2 -10.1 1.0 -6.5 13.0 16.5 -4.1 7.3 WHEAT 0.6 2.6 0.7 10.8 -10.1 0.9 -16.9 23.1 -2.6 1.8 -1.1 COTTON 0.4 -0.3 n.a. -12.3 -8.2 -8.6 -16.5 20.8 n.a. n.a. n.a. QUINOA 5.8 -6.0 12.1 -18.2 -49.9 70.1 -31.6 57.1 1.6 -7.2 15.3 BARLEY 1.9 -1.4 3.7 -4.8 6.0 8.2 -18.4 32.1 3.8 -0.9 4.0 YELLOW CORN -1.0 1.4 1.3 -2.3 -2.7 3.4 -2.3 9.9 -0.6 -3.4 0.2 SOYBEANS 3.9 2.2 -4.8 8.8 4.2 -10.0 -21.3 22.9 -22.4 11.4 11.0 SORGHUM -0.3 -0.9 11.5 -0.3 -23.5 10.4 -16.4 79.5 -8.2 1.3 21.7 SUGAR CANE -4.1 0.7 -1.4 4.9 -9.5 -9.9 -9.1 -8.2 16.7 -05 6.3 COFFEE -0.2 -2.2 n.a. 6.7 -18.6 12.6 4.2 1.3 n.a. n.a. n.a. Source: Ministero de Agricultura Table 4.4 AREA HARVESTED BY CROP (Thousands of Hectares) 1970 1980 1981 1986 1991 1992 1993 1994 1995 1996 RICE 140 103 161 161 159 166 178 239 203 210 WHITE CORN 228 180 191 204 196 138 170 175 205 216 BEANS 66 58 56 68 51 47 54 60 62 76 POTATOES 315 210 224 192 185 135 178 189 242 229 WHEAT 136 82 114 98 105 70 85 102 99 117 COTTON 144 148 157 166 118 86 65 nd nd nd QUINOA 16 19 18 20 21 8 18 21 19 19 BARLEY 186 110 118 123 113 82 101 113 115 129 YELLOW CORN 154 132 151 226 158 137 187 172 162 185 SOYBEANS 04 6.4 8.4 2.1 1 0 1 1 1 1 SORGHUM 4 12 14 10 10 1 4 1 1 1 SUGAR CANE 48 49 39 50 54 48 48 51 60 54 COFFEE 113 153 153 163 164 165 161 nd nd nd Source: Ministerio de Agricultura Table 4.4a GROWTH RATE OF AREA HARVESTED BY CROP (Percent) 1970-80 1980-90 1990-96 1985 1990 1991 1992 1993 1994 1995 1996 RICE -3.1 6.1 2.2 -181 -134 -13.8 4.5 6.6 349 -153 37 WHITE CORN -2.4 -1.8 63 0.2 -28 31 -29.4 230 29 173 50 BEANS -1.2 -0.4 5.3 5.5 -21.2 -8.4 -73 152 95 39 225 POTATOES -4 -3.5 7.8 5.9 -24 26 3 -27.0 31 7 6 0 282 -50 WHEAT -4.9 -0.1 6.2 -4 8 -30.6 28.1 -325 20.3 206 -33 182 COTTON 0.3 -0.7 - 65.2 -19 -14.9 -26 7 -25 0 nd nd nd QUINOA 1.3 -0.1 14.8 -7 1 -465 157.6 -625 1266 160 -95 -0 1 BARLEY -5.1 0.0 9.0 56 -421 483 -279 237 112 20 12.0 YELLOW CORN -1.5 28 1.1 -11 -37.1 -9 -13.2 360 -7.9 -57 143 SOYBEANS 32 -15.4 2.7 91 -14.3 -333 -71 4 157.6 92 520 44 1 SORGHUM 129 -8.8 -30.1 -47.1 -53.8 116.7 -865 1570 -729 20 -438 SUGAR CANE 02 -0.2 20 0 3.4 10.5 -107 -03 71 168 -87 COFFEE 3 1.6- 18 -57 -8.4 04 -23 nd nd nd Source: Ministerio de Agricultura Table 5.1 Farm Size and Land Use by Region, 1994 Farm Size No. of Individual Total Land (ha.) Landholders (thousands) Percent (thousands) Percent Total 1,745.8 100.0 5,477.0 100.0 Less than 0.5 has. 270.5 15.5 43.0 0.8 0.5 to 4.9 990.1 56.7 1,592.8 29.1 5.0 to 9.9 240.2 13.8 1,067.2 19.5 10.0 to 19.9 130.7 7.5 910.2 16.6 20.0 to 49.9 78.2 4.5 864.1 15.8 50.0 or more 36.1 2.1 999.6 18.3 Irrigated Farms 792.4 100.0 1,729.1 100.0 Less than 0.5 has. 115.8 14.6 22.6 1.3 0.5 to 4.9 511.7 64.6 649.7 37.6 5.0 to 9.9 102.5 12.9 390.7 22.6 10.0 to 19.9 39.5 5.0 241.0 13.9 20.0 to 49.9 15.5 2.0 152.4 8.8 50.0 or more 7.3 0.9 272.6 15.8 Rain-fed Farms 1,173.1 100.0 3,747.9 100.0 Less than 0.5 has. 103.0 8.8 20.4 0.5 0.5 to 4.9 676.2 57.6 943.1 25.2 5.0 to 9.9 180.0 15.3 676.5 18.1 10.0 to 19.9 110.4 9.4 669.2 17.9 20.0 to 49.9 71.1 6.1 711.8 19.0 50.0 or more 32.4 2.8 727.0 19.4 Source: III Censo Nacional Agropecuario 1994 Elaborated by GRADE. Table 5.2 NUMBER AND DISTRIBUTION OF PLOTS BY REGION AND FARM SIZE Costa Sierra Selva Total Plots Plots per farm Plots Plots per farm Plots Plots per farm Plots Plots per farm Total 393054 1.6 4935108 4.1 392951 1.3 5721113 3.3 Less than 3 has 50.7% 1.4 56.7% 3.7 22.4% 1.2 53.9% 3.2 3 to 9.9 39.5% 1.7 33.0% 5.2 36.1% 1.4 33.7% 3.8 10 to 49.9 8.9% 2.0 8.8% 4.2 34.9% 1.4 10.6% 2.8 SO or more 9.0% 2.1 1.5% 2.4 6.6% 1.4 1.8% 2.0 Source: Il Censo Nacional Agropecuario, 1994 Table 5.3 Farm and Farm Size Distribution by Region According to Number of Plots Owned Number of Total Costa Sierra Selva Plots Farms Hectares Farms Hectares Farms Hetares Farms Hectares 1 45.3 65.6 69.1 80.4 32.7 54.3 76.9 87.3 2 19.5 12.7 18.1 8.5 20.6 15.1 16.1 8.3 3 10.8 6.3 6.6 4.5 13.3 8.3 4.5 2.3 4 7.3 4.7 3.1 2.1 9.5 6.5 1.6 1.3 5 4.6 2.8 1.4 1.2 6.2 4.1 0.5 0.3 6 or more 12.5 7.9 1.7 3.3 17.7 11.7 0.4 0.5 Total 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 Num. 1745773 35381808.8 248809 2604361.8 1204243 22694097.8 292721 10083349.2 Source: III Censo Nacional Agropecuario, 1994 ANNEX II CASES STUDIES OF RECENT SUCCESSFUL SMALL AGRICULTURAL PRODUCERS IN PERU by Victor Agreda and Javier Escobal Orizinal Text, Still to be Edited ANNEX 2: Cases Studies of Recent Successful Small Agricultural Producers in Peru by Victor Agreda and Javier Escobal I. Introduction I I This annex presents seven relatively successful cases in Peruvian agriculture. The objective is to identify the principle factors explaining the success of each case, as well as point out some factors which could work to impede expansion of the sector. The selected producers are involved in cultivating the following products: mangoes, in Irrigaci6n San Lorenzo, Piura; asparagus and cotton, in Chincha Valley; potatoes for industrial processing, in Mantaro Valley; coffee beans, in Ja6n-Moyobamba; insect dyes, in Ayacucho; and alpaca wool, in Puno. 1.3 The specific factors identified as fundamental determinants of success in most cases are: a) stable or growing market conditions; b) adequate seasonal patterns (counter- cyclical harvesting) and/or favorable climatic conditions; c) adequate provision of public goods and services (e.g., road infrastructure and public and animal health services); d) attention to the problem of scarce managerial skill among agricultural producers. 1.4 It is important to note that farm size is not a key issue for success in Peruvian agriculture. Some cases show that small farming can be successful if managerial bottlenecks are appropriately dealt with, through either contract farming or producers' organization. I. Case Studies I.1 Mango Producers in Irrigaci6n San Lorenzo 1.1.1 In the market for mangoes (both fresh and processed), total area of cropland, volume of production, and level of exportation have all increased substantially since 1991. As a result, while in 1986 export volumes equaled 1,958 tons, principally destined for European and Canadian markets, this number has now risen to 5,600 tons, of which 70% of revenues corresponds to the exportation of fresh mangoes. 1.1.2 One of the most important factors explaining this growth is the solution of sanitation problems in the use of pesticides to treat and control fruit flies, which had previously resulted in a U.S. ban on importation of fresh mangoes from Peru, in effect PERU: An Agricuftul Developn=t Sltraegy Annex 2 since September 1987. Since 1991, hydrothermal treatment' for fruit flies, which was approved by the U.S. government's Environmental Protection Agency (EPA), has been applied to mango cultivation methods, allowing fresh mango exports to reach the substantial U.S. market, whose annual demand is close to 45,000 tons per year, in comparison with a European market demand of approximately 15,000 tons. Government intervention, through the Institute of Foreign Trade and the INIAA, played a critical role in solving this pesticide problem. 11.1.3 Other factors explaining the observed growth is adequate turnover, both natural and market. The most important production zone of fresh mangoes for export is Irrigaci6n San Lorenzo. This zone, due to exceptionally favorable climatic conditions, is able to produce high quality fruit precisely in those months (December to February) when other large exporters of mangoes (Mexico, India) are out of season. In other words, their disposition to favorable seasonal production with respect to the worldwide market enables these exporters to reach U.S. and European markets. Furthermore, the quality and variety of the mangoes exported from this region (including Hayden, Kent, and Tom Atkins) face a high demand in the foreign market, providing a certain level of price stability. II. 1.4 A third explanation for the rise in mango exports is the recent opening of the Port of Paita, which has allowed exporters to make use of marine export services from Chile, thereby lowering significantly the costs of naval export. 11.1.5 The above-mentioned factors would also explain the recent formation of a mango export industry in Piura and San Lorenzo, which has also contributed to the growth in the amount of domestic farmland dedicated to the cultivation of mangoes for export. These firms have established production contracts with middle-sized domestic producers (those with more than 20 hectares- of crops), who benefit from additional credit and technical assistance offered by exporting firms. All of these farmers enjoy registered titles. 11.1.6 Nonetheless, in order to ensure access to raw materials, some exporters of fresh mangoes have opted for vertical integration, buying out middle-sized producers instead of contracting their services. Fundamental in explaining this behavior is the fact that contracts with local growers have become relatively less convenient for agro-industry on account of the increasing liberalization of the market for land. However, it is important to keep in mind that the local mango growers that have been contracted by agro-industry have levels of capital and managerial skill well above the average local farmer. In this sense, it is also probable that a separate factor explaining vertical integration in this industry is the insufficient number of producers with desirable characteristics to enable ' Hydrothermal treatment is a means of quarantine control which proportionally attacks larvae of the varieties Medfly, East Indian fruit fly, South American fruit fly, etc. The process consists in submerging the fruit in water at a temperature of 46 degrees Celsius. PERU: An Agicuftual Development Stralegy Annex 2 expansion of agro-industrial operations. Given this scarcity, the only option in terms of production expansion is the transfer of land directly into the hands of agro-industry2 H.2 Asparagus Producers in Chincha Valley 11.2.1 In Chincha Valley, a significant substitution in land uses has occurred in recent years. Up until a few years ago, cotton was one of the most important crops for medium- sized and large producers. Nonetheless, many agricultural producers have recently substituted this crop primarily for asparagus cultivation, thereby expanding the asparagus industry that has been in this valley since the end of the last decade. 11.2.2 The country's southern coast provides an exceptional climate for the cultivation of asparagus, offering producers two harvests per year. The yields obtained (12,000 stalks/hectare) are also far superior to those registered in Spain (7,000 stalks/hectare), one of the principle worldwide producers and consumers of asparagus. Additional factors which contribute to explaining why the asparagus industry has settled and grown in the valley include: the proximity and accessibility of the Lima market, connected by a central highway; the atmosphere of security which presently characterizes the zone; and the presence of transport services and other commercial inputs. 11.2.3 However, one characteristic is particularly pertinent to the analyses presented in this document: As opposed to other valleys - for instance, that of Ica --, the significant presence of modern farming equipment has guaranteed industrial firms of the ability to implement desired productive technologies while insuring the availability of high quality raw materials. Among the factors contributing to the success of links between agro- industry and local asparagus growers include: the sufficient agricultural knowledge and management capacity of the majority of land-owners; the availability of modern technical equipment; the availability of arable land (on average, more than 50 hectares per holding), most of which are equipped with irrigation systems. In the case of these producers, again all of them show registered titles. II.2.4 These agreements are established through contracts in which the local producers obtain exceptional market conditions (including rent contracts, credit, technical assistance, etc.), resulting in reports of higher earnings with respect to those received in cotton production. It is clear that Chincha Valley is one of the only valleys in the country in which agricultural contracting (vertical coordination) has been a viable option. Nonetheless, in this type of arrangement between agro-industry and local growers, the small land-owner does not receive the direct benefits that are generated. In the end, these agents benefit solely as providers of labor during harvest season. 2 One finding which supports this hypothesis is the fact that land purchases by firms in the industry for mango concentrate have been substantially less frequent. This is easily explained by the fact that this particular industry demands a lower level of technical production methods from growers, which signifies lower fewer incentives for vertical integration. PERU. An Aicultural Development Strategy Annex A H.2.5 Just as in other zones of the country, as in the case of mango producers in San Lorenzo, the asparagus industry situated in Chincha Valley continues to buy up local lands, also opting for vertical integration. Though this process has not yet ended, it is possible to affirm that agro-industrialists have opted for ensuring access to quality raw materials for two reasons: a) many asparagus plantations have already entered a phase of diminishing yields and their infrastructure needs renovation - which also presents the opportunity to renew contracts with land-owners under a new type of arrangement. Until recently, outside businesses could not buy out local land-owners, yet under new legislation it is now possible. Secondly, there is a need for additional areas for cultivation beyond that which is in the hands of middle-sized land-owners. Given this shortage, sufficient contracted lands are no longer guaranteed to agro-businesses, and therefore they are now interested in buying up lands. It is clear that new legislation has offered large firms greater bargaining power in their negotiations with middle-sized land-owners, while the situation has worsened from the viewpoint of local land-owners. In order to assure a continuation of the beneficial contracts they receive from agro-businesses, middle-sized asparagus producers must now be more efficient in their operations due to the fact that firms in the asparagus industry now have a reasonable indicator of the cost of producing a kilo of asparagus in their own fields of production with which to compare with the prices of local producers. II.3 Cotton Producers in Chincha Valley 11.3.1 In Chincha, the abandonment of cotton production on middle-sized and large land holdings in favor of asparagus crops has consequently meant that small land-owners are the principle cotton producers in the valley. 1.3.2 A recent phenomenon in this valley has been the recent emergence of contractual arrangements between a small private consulting firm and an important group of producers, with the objective of augmenting the participatory role of local growers in the commercial aspects of cotton production. 11.3.3 The most important characteristics of this experience are the following: a) the consulting firm is composed of a group of agrarian businessmen with a long history in the valley; b) they receive no type of outside support (neither from the government, nor from International Technical Cooperation); c) in order for the producers to receive support from the consultants, each group must organize themselves into one company in the category of "Anonymous Society". To date, six firms have been formed which comprise a total of close to 260 small farmers and 1,200 hectares of land. The level of capital required of each association is relatively low, and is shared among all participating individuals. All of the afrmers are requested to have registered titles. The consulting firm occupies the role of managers contracted by each company of farmers; d) in a meeting with the director of each of the six companies of farmers, an agreement with each producer is approved, in which the consulting firm is obligated to complete the goals stipulated by each of the divisions. This offers the small farmer a role in supervising the company in meeting its PERU: An Aicultural Development Strategy Annex 2 objectives and e) as payment for their consulting work, the contracted firm charges each group of producers 25% of the direct costs (which represent an average US$239 per hectare). 11.3.4 The principle economic benefits which the small farmers receive from hiring the services of the management consultants are the following: a) a reduction in the price of inputs (an average discount of 24% with respect to the prices a small farmer would receive at a warehouse in Chincha), which represents a total saving for the producer of US$116 per hectare; b) a reduction in the interest rate charged by private banks with respect to that charged by the cotton gin companies when offering direct credit to the farmers, which results in an average savings of US$90 per hectare; c) a higher market price for the cotton they produce. Whereas cotton produced by small farmer has traditionally reached the market via the cotton gin companies which serve as middlemen, by forming their own associations producers can instead contract the services of capital owners during the period of production, and later sell their product directly to the textile companies. In terms of the price of raw cotton, this arrangement has allowed the producer to move from a price of S/. 85 (Peruvian soles) per 50 kilos (the fixed price offered by the middlemen capital-owners), to a price of S/. 101 (when sold directly to the textile companies). For the producer, this has signified an additional earnings of US$150 per hectare; d) as part of a company, the land-owners are able to switch from the government's simplified single income tax scheme to the "general" scheme, thereby allowing them to deduce from taxable income that which is spent on factor payments. In this manner, producers are able to obtain an additional savings of US$180 per hectare. In sum, by participating in an association of producers, the small farmer will increase his income by approximately US$536 per hectare. 11.3.5 With this method of operations, the management company has identified a role which is profitable for them, while also generating a "circle of benefits" which allows local economic agents to interact with greater ease and thereby increase overall utility. With this arrangement, private banks are more willing to offer loans to producers on account of the reputation, experience and knowledge of the owners of the consulting agency which supports them (given, of course, that each small farmer complies with the bank's minimum requirements, such as property titles, collateral, down payments, etc.). By connecting themselves with the matrix of conunercial activity of input producers located in Lima, they can acquire significant price discounts for large volume purchases. By gaining direct access to the textile companies, they receive higher prices with respect to those previously received from intermediaries. In other words, this arrangement generates economies of scale in production that each individual farmer is unable to achieve, thereby providing economic gains for the local growers as well as the private management consultants. 11.4 Growers of Potatoes for Industrial Processing, in the Montaro Valley II.4.1 The processing of potato-based snackfoods is one of the most dynamic and important areas of activity in the market for processed potatoes. In this market, important PERU: An Agricultural Development Sta Annex G changes are occurring as a result of private producers' efforts to improve their market position and increase sales. II.4.2 The most important firms involved in this line of products are multinational affiliates, in which 85% of sales of these products are concentrated. Until three years ago, the system of provision of potatoes employed by these firms occurred through intermediaries (which included wholesale buyers from La Parada). The system is now based in a direct relationship between mutlinationals and diverse groups of producers which vary according to seasons of production. From August to November, the firms receive potatoes from medium-sized producers located in the coastal valleys near Lima. These growers provide potatoes of the variety Tomasa Condemayla, that which is most readily employed in the production of chips (low in sugar, low in water, sufficiently large, and uniform in shape). The varieties of potatoes which are grown in the mountains do not posses such characteristics, making it necessary for the chip industry to warehouse potatoes from the coast in the last months of harvest and freeze them in rented storage space. Nonetheless, this method of production has not been able to meet the industry's factor demand for potatoes, given the existence of three other competing companies also demanding this input. Therefore, the industry has depended on another transnational firms' imports from Columbia of a different variety of potato, Diacol Capiro, in order to meet the demand. II.4.3 However, in the last two years, the firm has established an arrangement with a group of producers situated in the Mantaro Valley, with which they hope to ensure the provision of potatoes during all critical months of production. The valley's proximity to Lima, the location's present environment of security, and the transport facility provided by some of the country's best highways make this valley a clearly desirable location for acquiring agricultural inputs. Of equal importance is the presence of leading producers with significant managerial capacity. The firm has selected a group of such producers which collectively provide close to 1,000 hectares of land for cultivation, with average yields above 20 tons per hectare. The majority of corresponding seed-growers are professional technicians with years of experience in seed production, depending on areas of cultivation located in agro-ecological zones appropriate for the seed development. Similarly, the selected potato growers are producers specialized in this crop, whose production has been traditionally destined for the wholesale market of La Parada. II.4.4 The firm has provided these seed-growers with free test tube materials with the objective of developing greenhouse methods of producing young tubers of the Dialcol Capiro variety. The company has also guaranteed these seed-growers a regular market for this eventual product in accordance with a determined price and quality. The market for these tuber seedlings is naturally destined for the potato producers selected by the same firm, whom they have also guaranteed purchase of their product at a given price and level of quality. In exchange, the producers must agree to provide this product exclusively to the transnational firm, and also allow the firm's technicians to oversee their production advances and manage the delivery of the product. PERU: An Agricultural Developmnt Strategy Annex 2 11.4.5 The viability of extending this mechanism of pro.duction to other producers depends on the level of organization of local growers in order to conform to the requirements of the transnational's technicians. The interviews now being conducted by personnel from the firm indicate an interest on the part of the firm of establishing similar agreements and signing long-term contracts with other local producers. 11.5. Producers/Collectors of Insect Dyes in Ayacucho II 5.1 The demand for natural colorants from the food, pharmaceutical, and cosmetic industries has increased significantly in the past few years, due to growing restrictions on the use of synthetic dyes. This is true in the case of erithrosine, an artificial substitute of caramel acid found in parasitic insects. The evolution of the market for natural colorants has benefited countries which produce natural dyes, such as Peru, the principle producer of insect dye (commanding 80% of the worldwide production of this product). The exceptional climate, in terms of temperature and humidity, of such mountainous regions as Ayacucho, enables producers to depend on extensive wild cactus groves which are inhabited by the plant's natural parasite, Cochinilla Dactylopius coccus, a parasitic insect which feeds off of the cactus nectar. The caramel and caramel acid it collects in this manner has a grade of purity between 18-20%, much higher than that obtained form other countries, which is only 14% pure. 11.5.2 Beyond the favorable external market conditions, the atmosphere of security and the repair of the central highway, Los Libertadores, are other factors contributing to the growth in the area of land dedicated to the cultivation of insect dyes. However, the majority of this production is obtained from wild cactus groves located on communal peasant lands, making use of traditional methods for harvesting the parasites (without irrigation, periodic hedging, or adequate fertilizer, nor adequate post-harvest procedures). As a result, caramel acid yields are relatively low in this zone. On this type of croplands, a grade of 18% purity is obtained for caramel acid, whereas managed groves obtain a grade of 22%. 11.5.3 Nonetheless, the production of insect dyes obtained from these wild groves constitutes the principal provider of this particular input for export production. They have installed a system of harvest which makes use of the equipment of intermediary buyers, who join local harvesters to buy fresh and dried cochinillas. This system ends up insufficient to ensure export businesses a sufficient quantity of high quality raw materials, despite the fact that demand for this product has grown at the same rate as its price (in 1990 the price of collecting a kilo of dried insects was US$23; it has since risen to USS44). 11.5.5 With the goal of ensuring adequate supply, three of the six most important export companies, in which 80% of export volumes are concentrated, have installed in Ayacucho close to 80 hectares employing exploit-intensive harvest methods (technical irrigation systems, higher crop density, insecticide, adequate harvesters and dryers, etc.). PERU: An Agricultural Developmnent Souca Annex 8 Nonetheless, the lack of available lands offered on the market is an important factor leading other firms to install insect farms in other mountainous regions of the country, as well as in coastal zones close to Lima, such as in the case of Barranca. 11.6 Producers of Alpaca Fibers in Puno 11.6.1 The commercial production of alpaca fibers is the most important economic activity of alpaca farmers, and their principal source of money income. It is also the activity through which organizations of producers have attempted to increase their participation in the commercial realm of production, with differing results for each case. 11.6.2 In the market for alpaca fibers, there exist two clearly distinct flows of commercial production. The traditional commercial circuit consists in the alpaca farmer selling his fibers to the local buyer and/or product claimant who works for the collection companies of the sector's most important industrial groups. The alternative commercial circuit is direct sale of the raw materials to producer guilds. The guilds differ in the mode of price classification of raw materials for sale, though the majority choose to sell the fibers in bulk (without classifying them), and only a small number of firms make the effort to classify fibers. The sale of alpaca fibers in bulk to these organizations has little effect in raising the producer's price; on average, the price for guilds is no more than 5% above the price received by intermediate buyers. This price differential is achieved, in most cases, through indirect subsidies and with little long-term perspective, such that organizations of producers have little incentive to opt for alternative strategies to the traditional commercial circuit. 1.6.3 The commercial sale of distinct classes of alpaca wool is an alternative which the Center of Alpaca Farning Cooperatives of Puno (CECOALP) has promoted in more than five separate sales campaigns, in the southern mountain region. CECOALP has been successful in implementing a system of wool classification that is relatively simple and extremely effective. Producers receive a price in accordance with the quality of the delivered wool, a classification which takes place in the purchasing centers by trained female employees. The system of collection and classification of the wool at CECOALP is dependent on the availability of finances for the purchase of raw materials, now that the producers demand up-front payment at the moment of receipt of the materials. This has caused CECOALP to amplify its sources of credit beyond the cooperative sources. This method of selection and classification has meant a significantly higher effective market price. When the producers in the 1994-1995 sales campaign sold wool to the guilds for commercial sale of the product in bulk, they received a price equivalent to one kilo of top? (US$ 1.470), barely 3% higher than the prices paid by traditional intermediate buyers (US$1.375). When the product was sold to CECOALP for commercial classification they obtained an average price (US$1.747) 27% higher than that of the intermediate buyer, and 23% higher than sales in bulk. 3 "Tops" refers to the form of the product inunediately after the first stage of industrial processing. PERU: An Agrcultural Development Strategy Annex 2 II.6.4 Through the national public foods program, PRONAA, the government has decided to intervene in commercial wool production for the past two sales campaigns, buying bulk fibers directly from producers, offering prices slightly above the market price, in an attempt to remove the negative welfare role of the iutermediaries. Unfortunately, this intervention has the opposite effect: their role has increased speculative behavior as other economic agents now tend to hold off all sales operations until discovering the price offered by PRONAA. Those producers with problems of liquidity were in this context forced to sell their wool at prices far below those of the last sales campaign given that few buyers were willing to raise prices. The guild organizations, because they did not represent direct producers, were not considered eligible by PRONAA for the purchase of raw materials. The government had in the past already tried similar forms of market intervention, with similar results -- that is, without having achieved their objective of regulating prices. Apparently, they had not learned their lesson from these experiences. 1.6.5 Currently, it is common for producers' guilds to form direct links with industry in order to commercialize bulk fibers. In terms of classified wool, producers' insufficient managerial capacity and lack of infrastructure (centers of collection and access to credit), have prevented them form successfully expanding this method of commercial sale'. 11.7 Coffee Producers in Jain and Moyobamba H.7.1 Since October of 1994, the program ADEX-AID-MSP has executed a group of activities targeting small coffee producers in various zones of the country, which incorporate a total of close to 11,000 hectares of production. Two of these zones are Jain and Moyobamba, which comprise 525 producers possessing a total of 1,218 hectares. The objective of the program is to increase volumes of production and levels of productivity and income of coffee producers through: direct technical advising; installation of treatment plants for hydrated coffee in the same locales; and credit provision for the harvest and commercialization of coffee beans through local committees and the National Coffee Association. 11.7.2 The organization of producers into local committees has facilitated the promotional efforts of the national program, particularly the installation of treatment plants, which have raised levels of productivity and increased the market price producers receive from industrial processors. In terms of the prices, the have obtained between S/.25 and S/.30 more per 50 kilos for industrially treated coffee. However, a major problem continues to be the financial demands of small producers each five months before harvesting. The lack of access to formal credit has also prevented producers from utilizing adequate fertilizer and pesticides. 11.7.3 The program has also coordinated the participation of the most successful industrial coffee companies by informing them of the high quality of coffee obtained from In the beginning, the success of this system was negatively affect by the industry's lack of support for this method of commercial sale. However, this restriction is no longer a factor of influence. PERU: An Apicultural Developnmt StCV Annedg the treatment plants. In a similar fashion, they have shared with them information on the stable climate of security which now characterizes the zone of production, allowing producers to conduct commercial operations in normal conditions. Yet the impact of this program on productivity and yields still has not been felt by the majority of coffee growers, due to the several restrictions to expansion of the program that have not been solved. 11.7.4 Among the principle restrictions to expanding the program, one of the most important is the absence of property titles that limit access to credit from private banks. As a transition move, the program has managed to generate resources from the International Cooperative, but not sufficient to meet the demand for credit in this sector. In the same manner, the majority of roads which constitute the routes of access to the fields of production have been reported to be in bad condition, which has been a major obstacle to a greater number of producers benefiting from the services of treatment plants. III. Conclusions Ill.1 In all of the cases analyzed above, the producers depend on a favorable market environment (i.e. in which they have identified a product with a stable or growing market, for which the specific region has comparative advantage in production due to advantageous climatic conditions and/or appropriate seasonal patterns). Beyond these external factors, in some of the cases, government intervention by way of the provision of public infrastructure - roads, in particular - has been a determining factor (for cochinilla, in Ayacucho). In other cases, encountering methods of improving sanitation problems through the joint efforts of the State and the private sector represents the fundamental variable (mango in San Lorenzo). In this case, the participation of SENASA appears to be critical for technical support, and also for negotiating with the importing country in an effort to eliminate the restrictive sanitary requirements. Finally, in other cases, the deciding factor is attention to the scarcity of managerial capacity. M.2 The solutions encountered in the cases of scarce managerial capacity deserve a separate chapter. The cases discussed above show at least three distinct forms of approaching the same problem. In the case of asparagus production in Chincha Valley and of exporter producers of fresh mangoes, for example, the industrialists have had to limit their contracts to medium-sized producers, which are characterized by a higher than average level of managerial capacity. In the case of alpaca farmers in Puno, the key has been organization on the part of producers, supported by international technical cooperation and the presence of adequate leadership. Finally, perhaps the most novel and interesting case is that of the small cotton producers in Chincha Valley. In this case, the key factor has been the presence of a private firm offering managerial services to producers. Due to the prestige of the private consultants, they have been able to convince several small producers to hire their consulting services in order to receive help in commercializing their product and obtaining credit. PERU: An Agricultural Developmt U Anned2 111.3 One aspect important to emphasize are the recent land purchases on the part of many of the exporters of mangoes and asparagus. Such behavior could indicate limitations to the method of augmenting their scale of operations. Given that the rest of the producers from both zones are small land-owners with limited financial and managerial capacity, it is necessary for industrial exporters to manage their own lands in order to expand their scales of operation. 111.4 One form of encouraging the influence of those factors indicated as important to the success of small producers, is by analyzing those cases in which the results are less positive due to the absence of one of these factors. On the one hand, we have the cases in which government intervention has taken place through subsidies or direct purchases. State intervention in the market for alpaca wool, by way of PRONAA, is one example in which this type of intervention has negative effects for the producers organizations that had achieved higher market prices through the classification of wool. Other cases with similar types of results include the creation of processing and commercialization companies for loan recuperation in the cultivation of rice in Ferrenafe and of yellow corn in Picota (Tarapota); the purchase of quinoa, dried fish, and alpaca wool in Puno; the purchase of rice in Santa Rita de Castilla (Samiria), etc. M.5 With respect to problems of poor sanitation, there is a long list of cases: banana production in Tumbes, plagued by the pest sigatoka negra; orange cultivation in the coastal and central jungle zones, with the pest tristeza; cocoa, produced permanently in the high jungle areas such as Ja6n and Moyobamba, and plagued by moliniasis de cacao; chirimoya production in warm Andean valleys, such as the case of Andamarca Valley in Concepci6n, plagued by the fruit fly. 1.6 The case of insect dyes from the jungle illustrates the importance of communication infrastructure. Currently, close to 300 hectares of cultivation have been installed on both sides of the central highway connecting Igquitos to Nauta. Similarly, a processing plant has been installed near the entrance to the city of Iquitos. Furthermore, the government is currently finishing paving the highway, which will greatly facilitate transport of this highly perishable good, as well as for other of the region's exports. Nonetheless, repair of the region's secondary roads, which connect production centers with the central highway, has not been initiated. Thus, in the rainy season, producers remain isolated and the harvests of insect dyes is either lost, or must be transported by mule on roads of distances greater than 3 kilometers. Due to the perishable nature of the good, such transport methods significantly lower the quality of the product. ANNEX III RURAL FINANCE IN PERU Dale W Adams and Juan Jose Marthans Original Text, Still to be Edited ANNEX 3: RURAL FINANCE IN PERU by Dale W Adams and Juan Jose Marthans Like many other low-income countries Peru suffered major disruptions in its rural financial markets the past several decades, including drastic changes in institutional makeup, in relative size, in geographic coverage, in functions, and in donor and government support. Large segments of these markets collapsed during the late 1980s and only a few portions have been partially rebuilt. The major gaps that persist in these markets limit the ability of Peruvian agriculture to modernized, to compete in international markets, and reduce rural poverty. Although the private sector has made some progress in expanding rural lending since 1991, experience in other countries suggests that rural finance-because of its costs and risks-nearly always requires some special attention from government and donors. In the following we focus on the problems and opportunities associated with filling gaps in the rural financial system in Peru. Heeding Santayana's warning that "Those who cannot remember the past are condemned to repeat it," we first present an historical review of the rise and collapse of the system. We then offer explanations for what went wrong. After this we detail the steps that have been taken since the early 1990s to rebuild the system and describe the status of these markets and the gaps in them in late 1997. We conclude with a discussion of options and steps that might be taken to more rapidly expand financial services in rural areas. RISE AND COLLAPSE Four components of rural financial markets in Peru merit attention in an historical review: informal finance, commercial banks, the government owned agricultural development bank, and other financial institutions such as credit unions. Informal finance In terms of number of rural people who participate, informal finance has likely always been the most important component of rural financial markets in Peru. Before formal rural financial markets began to expand after World War II, self finance, borrowing from friends and neighbors, loans from landowners, merchant lending, and finance substitutes such as share cropping and traditional work groups comprised most of rural finance. Unfortunately, there is a paucity of historical information that documents activities in informal finance and one can only speculate on how these activities have reacted to the rise, collapse, and rebuilding of the formal financial system in rural areas over the past 5 decades. Several recent studies by the Centro Peruano de Estudios Sociales (Alvarado 1997a and 1997b) suggest that substantial amounts of informal PERU- An Agrcultural Development Strategy 2 Annex 3 finance continue to be common in rural areas.1 Recent interviews with commercial bankers strongly suggested that much of their agricultural lending to agro-industries and others substantially reinforces informal lending to farmers. In some respects, rural financial markets in Peru have gone full circle: from dominance by informal finance soon after World War II to similar dominance in the 1990s. The expansion of informal finance in rural areas the past few years has allowed farmers to sustain agricultural production. Commercial banks Likewise, there is little historical information available on what commercial banks did in rural finance, aside from how much they lent for agricultural purposes, the relative importance of their lending, and where their lending was done. Prior to the 1960s commercial banks provided the majority of formal loans that were called agricultural in Peru. Most of these loans were concentrated in the coastal region, almost all involved real estate as collateral, and the bulk of the loans were made to farmers or agro-industries involved in agricultural exports. After the 1950s commercial banks were increasingly crowded out of rural lending by the heavily subsidized activities of the government-owned agricultural bank. Land reform turmoil, rural violence, and unfavorable terms-of-trade for agriculture further discouraged commercial banks from seeking agricultural clients It wasn't until the agricultural bank was liquidated in 1992, macroeconomic reforms were in place, and rural violence was dampened that private banks began to return to agricultural lending. The Agricultural Bank The agricultural bank in Peru experienced several name changes after being formed in 1931; it was called the Banco de Fomento Agropecuario (BFA) but this name was changed in 1975 to the Banco Agropecuario del Peru (BAP). Initially the bank focused most of its activities in the coastal region and lent mainly to producers of export crops. In the late 1960s and early 1970s BFA/BAP's activities were reoriented to support land reform and central planning throughout the country. In addition to providing subsidized credit, the bank softened its collateral requirements and attempted-- with little success--to make more term loans with donor assistance. This was accompanied by laws that required BAP to lend at least half of its loans to poor farmers. These shifts in mission were pronounced when the political party in power changed. The close ties between the bank and the Ministry of Agriculture often resulted in BAP being forced to carry out political mandates. Both the number of employees and number of BAP offices rapidly expanded during the 1970s and early 1980s to provide increasingly subsidized loans in remote 'A proposed research project on informal finance sponsored by the Ministry of Economics and Finance, COFIDE, and the Inter-American Development Bank should shed additional light on informal finance in Peru. This section draws heavily from Vogel, Adams and Marthans. 2 PERU: An Agricuftural Development Su*tU 3 Annex 3 areas. From early 1975 to late 1979, for example, the number of BAP employees increased by 46 percent, even though the real value of its loan portfolio declined by more than a quarter (World Bank, 1981, p. 2).' BAP also rapidly enhanced its communication network during the late 1980s: putting two-way radios in more than 200 offices, purchasing computers, and providing telephones for more than 100 offices.' The adverse effects of this expansion on administrative costs were paralleled by hefty rates of inflation that substantially exceeded the nominal interest rates charged on BAP loans. In the later 1980s BAP was required to make loans in disadvantaged areas at zero nominal rates of interest, with the promise that government would reimburse BAP for the interest rate subsidy, a promise that was not kept.' High administrative costs, sticky interest rate policies and accelerating inflation evaporated the purchasing power of BAP's loan portfolio and converted the bank into a dependent of the Central Bank by 1990. This was accompanied by wide swings in the purchasing power of BAP's loan portfolio. Between 1987 and 1988, for example, the purchasing power of its loan portfolio was about cut in half and the bank was forced to suspend term lending (Table 1). By the end of the 1980s BAP had approximately 5 thousand employees, had at least part-time offices in most rural communities, but was on a life-support system maintained by the Central Bank. The large monetary emissions needed to keep BAP afloat became an important factor fueling inflation. Dissatisfaction with its performance and attempts to limit the supply of money led to the liquidation of the bank in 1992. Notable trends: At least five notable trends stand out in BAP's history: First, as can be seen in Table 1, over the period 1970 to 1989 BAP steadily increased its loan coverage from less than 4 percent of all farmers to about a quarter of them in 1989, a substantial achievement. The more than doubling in percentage of farmers receiving BAP loans from 1985 to 1986 resulted from a reorientation of the bank to making many more and smaller loans outside the coastal region. In the late 1980s BAP extended loans to farmers who cultivated about half of the farm land in the country, although many borrowers likely received smaller loans than their opportunities justified because of severe credit rationing by BAP. 'The real rates of interest charged by BAP were negative throughout the 1970s and 1980s. In the late 1970s and early 1980s these negative rates ranged from 40 to nearly 60 percent and later in the 1980s became even more highly negative. 4At closing, BAP had the most sophisticated, country-wide communication system of any bank in Peru. sThe loans made during the late 1980s at zero interest rates comprised 10 to 15 percent of BAP's portfolio and at the time of closure the government's debt to BAP for the interest rate subsidy amounted to something between $150 million to $380 million, depending on what exchange rate (official or black market) is used to convert local currency to dollars. 3 PERU: An Agpcultural Development StUeg 4 Annex 3 Second, it can also be noted in Table 1 that the ratio of new-BAP-loans to the- gross-domestic-product-from-agriculture grew steadily from 1970 to 1980 and then fluctuated substantially over the 1980s until plummeting below early 1970s levels in 1987. In large part, the fluctuations in this ratio were due to ebbs and flows in the real value of BAP's new lending caused by inflation and ups and downs in the amounts of donor and government money available to the bank. A third trend was the crowding-out of commercial banks by BAP. In 1950 BAP provided a third of all formal agricultural credit in the country, with commercial banks supplying the rest. Over the next three decades BAP increasingly extended a larger percentage of total agricultural loans, until capturing more than 90 percent of the formal agricultural loan market in 1976 and 80 percent-plus thereafter (Palomino 1993b, p. 5). Commercial banks were crowded out of the rural financial market by BAP's subsidized loans and were further discouraged from making agricultural loans by land reform, food price controls, and war-like conditions in some rural areas. Fourth, despite being a development bank that supposedly emphasized medium- and long-term loans, BAP was unable to increase, or to even sustain, the portion of its portfolio used for long-term investments. Despite a number of large credit projects funded by the World Bank and the Inter-American Development Bank (IDB) that- were primarily aimed at lengthening the term structure of BAP's loan portfolio, the portion of total BAP lending for investment purposes was usually less than 10 percent over the 1970-1988 period, and in some years all new loans made were for periods of less than a year (Palomino 1991, p.8). Fifth, the mix of money used by BAP to fund loans changed dramatically during the latter part of the 1980s; the importance of additional deposits and loans repaid declined in real and relative terms. Despite being allowed to offer slightly higher interest rates on deposits than commercial banks, and receiving preferential treatment on reserve requirements, BAP was unable to attract many deposits from households. As can be noted in Table 1, the real value of deposits in BAP fluctuated widely from year-to- year due to inflation and changes in exchange rates, but fell sharply after 1986. Shortly before being liquidated in 1991-92 most of BAP's deposits were comprised of quasi- compensatory balances deposited by merchants who sold inputs to BAP clients and congealed deposits owned by farm organizations. From 1986 through 1989 the increase in nominal deposits provided less than 3 percent of the funds used in BAP lending (Palamino 1993b, p. 23). 4 PERU: An Agricultural Development Straete 5 Annex 3 Table 1: Agricultural Bank of Peru: loans, deposits, and inflation, 1970-1991 Year Loans New Loan*** Deposits** Annual $US million* Loans coverage $US inflation New loans Outstanding /GDP ag. % million* rate balance % 1970 84 124 7 4 62 5 1971 84 112 6 4 54 7 1972 93 129 7 4 64 7 1973 122 149 7 4 85 14 1974 186 230 9 4 104 19 1975 212 237 9 5 97 24 1976 242 269 15 6 57 45 1977 213 296 13 6 9 32 1978 206 206 16 6 12 74 1979 364 252 22 7 40 67 1980 491 411 25 7 41 61 1981 532 466 21 8 66 73 1982 402 452 16 7 81 73 1983 322 358 15 7 83 125 1984 336 490 14 9 84 112 1985 284 294 14 10 76 158 1986 690 552 25 23 121 63 1987 257 469 5 22 43 115 1988 52 92 3 20 10 1,722 1989 183 203 6 25 34 2,775 1990 205 272 5 - 27 7,650 1991 nil 82 - - 25 139 Source: Various Central Bank; Superintendent of Banks and Insurance, and Agricultural Bank of Peru publications. Also Palomino, 1993a. *Converted to $US using parallel market rates of exchange. **Includes substantial amounts of deposits made by agricultural associations and organizations. ***Percent of all farmers in Peru who received a loan from BAP. 5 PERU: An Agricultural Development Strategy 6 Amex 3 Donors' support for BFA/BAP: Compared to other agricultural development banks, BFA/BAP received extensive donor support, possibly too much. The World Bank early played a lead role in this. Its first loan to BFA was for US$5 million in 1954 (105-PE).6 This was followed by 5 additional loans in 1957, 1960, 1965, 1973, and 1983 (162-PE, 257-PE, 415-PE, 933-PE, and 2302- PE) totaling US$175 million.7 In addition, the Agency for International Development provided about US$ 43 million to BFA during the 1960s to develop a costly supervised credit program (Carrenza; North Carolina State Mission). Starting in 1971 the IDB also provided BFA/BAP US$226 million through a series of seven loans, including US$11.5 supplied by the International Fund for Agricultural Development (IDB 1983; IDB 1995). The total value of donor support for BFA/BAP summed to well over a half billion 1990 US dollars. Although the objectives of the various donor projects with BFA/BAP varied, several goals were common in most of these efforts, especially those funded by the World Bank. Emphasis was placed on increasing the volume of medium- and long-term lending to stimulate agricultural investments, and also on increasing the total volume of formal agricultural lending. Until the last World Bank project (2302-PE), little donor emphasis was placed on strengthening BAP or on enhancing the efficiency of its operations. Likewise, aside from one IDB study in the early 1980s, none of the evaluations of donor projects with BFA/BAP attempted to track the overall financial performance of the bank (IDB 1983). Credit Unions In the 1960s and 1970s the credit union movement in Peru was often cited as being one of the strongest in Latin America. In 1982 the movement had 800 thousand members, 412 credit unions, and deposits/shares worth nearly US$60 million. A substantial number of these being in communities and towns. Many of its members outside Lima earned at least some income from agricultural activities and some loans were made to farmers. Unlike BFA/BAP, the credit unions largely relied on "inside money" rather than "outside funding" for lending. All of its members contributed to the share capital of the cooperative, and for a time many of them deposited additional amounts in savings account. Because credit unions provided opportunities for some depositors to obtain small loans, credit unions were a more attractive place for poor people to hold savings than were banks that were much less likely to offer small loans to depositors. The vitality of the credit union movement in Peru began to deteriorate during the 1970s. Accelerating inflation combined with sticky interest rate polices in most credit 6 This was the first-ever agricultural credit loan made by the World Bank. By the late 1980s the World Bank would commit worldwide more than $US 16 billion in subsequent agricultural credit projects, many of which were patterned after this first project in Peru (World Bank 1993). 7 In 1987 US$31.3 of the last loan (2302-PE) was canceled. 6 PERU: An Agricuftural Development Strtegy 7 Annex 3 unions steadily eroded the real value of share capital and deposits in these increasingly fragile cooperatives The traditional credit union model in Latin America stressed the importance of providing inexpensive loans and it was common practice to charge a nominal annual rate of interest of only 12 percent on loans. Shareholders and depositors, by necessity, received even lower rates of return on their savings. Since annual rates of inflation from 1973 onward were three to six times the rates of interest charged on loans, borrowers from credit unions captured hefty income transfers in the form of negative real rates of interest on their loans. At the same time, shareholders and depositors were effectively "taxed" by even higher negative rates of return paid on their savings to provide credit subsidies. Managers and members of boards of directors often resisted increasing rates of interest because of tradition, and because many of them were beneficiaries of the credit subsidies (Gadway). Because "insiders" soon colonized the subsidized credit, other members of the cooperative were increasingly denied access to loans and protested by withdrawing their share capital or deposits and becoming inactive members; others effectively did the same thing by refusing to repay loans. The hefty negative rates of interest paid on deposits hastened this flight. Subsequently, many of the credit unions, especially in rural areas, imploded because they were converted into borrower-dominated, rent-dispensing organizations. The Agency for International Development (AID) attempted to arrest the disintegration of the credit union system through a small, two-year, technical assistance project that focused on strengthening the Banco Nacional para Las Cooperativas (BANCOOP), along with helping to reform several credit unions (Vogel 1984). The project started in mid-1979 and is noteworthy because it broke ranks with the traditional donor approach of providing funding for on-lending, and, instead, focused on deposit mobilization (Burkett).' This included encouraging BANCOOP and participating credit unions to offer much higher interest rates on deposits, improving the quality of deposit services, and opening new branches. Despite operating in a stagnant economic environment, combined with accelerating inflation, the BANCOOP project was highly successful in boosting deposit mobilization in participating organizations. The amounts mobilized substantially exceeded the targets set for the project, the quality of deposit services were enhanced, and all participating organizations were strengthened financially (Larson and Adams). Several important lessons can be drawn from the BANCOOP project, some positive and some negative. First, relatively lethargic financial institutions can effectively mobilize deposits if they offer attractive incentives to savers and provide high quality services. Second, poor people in stagnant economic environments will deposit substantial amounts under conditions they find attractive. One can only speculate on how much better the performance of the BANCOOP project would have been if it had operated in the more ' The funding for this AID project totaled about one-half million US$, only about one-tenth of one percent of the total volume of donor funds provided to BFA/BAP for credit purposes. 7 PERU: M Agricultural Development Strategy 8 Annex 3 hospitable economic environment of the 1990s in Peru, rather than the hostile environment of the late 1970s-early-1980s. Third, mobilizing deposits is one thing and protecting them after they are collected is another. Being a weak cooperative organization, BANCOOP was not a full-fledged bank. It was not as carefully regulated and supervised as other deposit-taking banks in Peru. This lack of prudential regulation contributed to irregularities in BANCOOP lending that resulted in the liquidation of the bank in the mid-1980s. With the benefit of hindsight, it was not prudent to foster more deposit mobilization by BANCOOP before first subjecting the bank to better prudential regulation.9 Even with adequate supervision and regulation BANCOOP and the reformed credit unions would likely have been drowned in the tidal wave of inflation that washed over Peru in the late 1980s. In mid- 1997 there were still 219 registered credit unions in Peru with nearly 243 thousand members. Only about half of these credit unions were active, however. About 40 percent of the credit unions are concentrated in Lima with the remainder in other towns and cities. The total volume of outstanding loans made by the 115 credit unions that reported to the National Federation of Credit Unions was about US$116 million, and the amount of deposits was US$112 million. All of the lending by the credit unions depended on deposits, share capital, and retained profits.10 Despite the substantial contraction of the Credit Union system in Peru, in 1997 these cooperatives lent more than the Cajas Municipales and mobilized more deposits than the Cajas Municipales and Cajas Rurales combined. POSTMORTEM OF BAP" Because BFA/BAP played such a large role in government and donor efforts to treat rural problems in Peru, it is useful to sift through the ashes of that experience for lessons. Clearly, some of BAP's difficulties were due to forces beyond the bank's control while other wounds were self inflicted. The outside forces included an unstable and often hostile macroeconomic environment, drastic changes in the role assigned to BAP, and macrofinance policies that made it impossible for BAP to operate profitably. The internal problems included a non-bank culture, dependency on outside funding, lack of deposit mobilization, and excessive administrative costs. Economic Environment Providing formal financial services in rural areas is costly--especially in Peru. The lack of investments in public services in rural areas, policies that turned the terms- of-trade against agriculture, and natural disasters often dampened the economic 9 The appropriateness of mobilizing deposits in the 1990s has been reinforced by the substantially strenghtening of the Superintendencia de Bancos y Seguros. 10 Unpublished information provided by the National Federation of Credit Unions. The appropriateness of mobilizing deposits in the 1990s has been reinforced by the substantial strenghtening of the Superintendencia de Bancos y Seguros. "This section draws on Vogel, Adams and Marthans 8 PERU: An Agricultural Development Strateg 9 Annex 3 environment in agriculture. The relatively small size of transactions, combined with the geographic spread of clients, elevated the average cost of financial intermediation and further discouraged bankers from,penetrating rural areas. Price- or exchange-rate controls, the rugged Andes that divide the country, and limited amounts of fertile land often provided an unfavorable economic environment for BAP borrowers and depositors. 12 Inflation and the heavy hand of donors and politicians were the most powerful outside force that adversely affected BAP, especially in the two decades before its liquidation. Inflation eroded the real value of its capital base, increased the difficulty of mobilizing deposits, and, combined with concessionary interest rate policies, melted the real value of its loan portfolio. Terrorism during the 1980s and 1990s likewise dampened the economic environment in rural areas and added substantially to BAP's costs. During the worst periods of terrorism the BAP was spending an average of $2,000 a month per office for added security. Role Changes BFA/BAP's problems were compounded by abrupt changes in the roles assigned to it. These ranged from largely financing commercial agriculture in the coastal region, through supporting land reform throughout the country, to offsetting with subsidized credit the distortions caused in agriculture by macroeconomic policies, to channeling fiscal transfers to borrowers. The BAP passed through several cycles where it was first asked to assist with boosting production, and then told to be primarily a social or civil defense agency. It was extremely difficult to maintain discipline within the bank and between the bank and its clients with these types of dramatic changes in roles. Macrofinance Policies During most of its life BFA/BAP was forced to operate under interest rate ceilings. This limited its ability to price financial products according to their costs and risks. In some cases the bank was required to charge the least on loans that were the most costly (and risky) per unit of money to administer. Combined with periodic surges in inflation, concessionary interest rate policies repressed the term structure of BAP's loan portfolio and eventually decapitalized the bank. Battling to survive, BAP was often forced to shorten the term structure of its loans, thus avoiding lending that supported long-term investments--a negation of the primary justification used in forming agricultural banks. Lack of a Banking Culture Close ties with the Ministry of Agriculture limited the ability of BFA/BAP to create a banking culture and provided few fire walls between the bank and politics. It 12 The operations of financial markets were also adversely affected by laws and courts that limited the range and attractiveness of various types of loan collateral (World Bank 1997). 9 PERU: An Agrialitura Development Strate 10 Annex 3 was essentially a division of the Ministry of Agriculture and many of its employees came from that Ministry--especially its leadership. It tended to promote from within and employed individuals with knowledge of agriculture, rather than those with banking talents. Most of its employees viewed credit as an input in achieving agricultural or social objectives, rather than viewing loans as being half of financial intermediation, and seeing deposits as a ladder out of poverty. The measures that BAP used to gauge success were such things as loans made, crops planted, livestock financed, investments funded, geographic locations of loans, and size of the organization." Little attention was paid to bank profits, to administrative costs, to transaction costs, to quality of services, or to financial innovations. Because of its links with the political system, loan-making and loan-repayment decisions were sporadically taken out of the bank's hands. Dependency Because BAP mobilized few depositors and did not generated significant real profits, it depended on "outside funds. " This had unintended consequences that adversely affected the vitality and sustainability of BAP; it increased the vulnerability of BAP to political intrusions and to donor fads. In addition, managing a relatively large number of idiosyncratic lines of credit provided by the World Bank, IDB, other donors, and the Government of Peru substantially increased the administrative costs of BAP, as well as the transaction costs of borrowers. An unknown number of BAP staff undoubtedly spent millions of hours processing data on dozens of lines of credit administered by the bank, preparing periodic reports on same, and assisting with evaluations.14 In 1988, for example, BAP handled 22 lines of credit from different sources. Much of the data processing in the bank tracked these lines of directed credit and this displaced critical management information such as loan recovery status. Over much of its life it was cheaper for BAP to draw on Central Bank money, donor funds, or concessionary loans from commercial banks than it was to mobilize small deposits from rural households." The Bank had no incentive to promote deposit mobilization, to provide innovative deposit options, to lower savers' costs of making deposits, or to broaden its client base by aggressively seeking more depositors. "The skimpy information about BAP's financial performance in its annual reports is an indication of the non-bank role assigned to the bank (see, for example, Palomino 1993b). " In the early 1980s BAP had more than two dozen employees in its Lima office working full time on collecting and processing data on IDB programs with the bank. "In some years the government imposed agricultural loan quotas on commercial banks. These banks could meet the terms of these quotas by lending money to BAP at subsidized rates. 10 PERU An Agricultural Development Sratey 1 Amex 3 Flawed paradigm With the benefit of hindsight, BAP's demise was built into its birth and nurtured throughout its life. It was designed and operated under the old directed credit paradigm that was only compatible with subsidies, suppression of market forces, and central planning (Vogel and Adams). Given the inflexibility of BAP and its habits, it was seen by some as an anachronism when a new financial market paradigm came into vogue in the late 1980s and early 1990s that emphasized market forces, elimination of subsidies, and efficient and sustainable financial markets. It is ironic that the organizations which had earlier used and abused the BAP (donors and various Peruvian governments) were the ones who ultimately turned on the bank and forced its liquidation. The explicit reasons given for closing the bank included the perception that BAP was insolvent, that it had far too many employees, and that it had become a black hole into which government funds went, never to be seen again. In addition, a few decision makers concluded that BAP no longer performed a useful function. Some key policy makers, including the Minister of Economics and Finance at the time, felt that the relationship between formal credit and agricultural production was weak (Palomino 1993a, p. 30)." A strong push to disentangle the Central Bank from providing development funds reinforced the decision to liquidate BAP. REBUILDING THE RURAL FINANCIAL SYSTEM In the early 1990s it was widely assumed that private-sector lending would soon fill most of the void left by BAP's closure." As a bridge between the liquidation of BAP, and the hoped-for expansion of private sector lending to farmers, the government implemented several measures before or soon after BAP was closed in May, 1992 (Palomino 1993a). The first measure continued a regulation that imposed an agricultural loan quota of 10 percent on the loan portfolios of commercial banks. A second measure liberalized restrictions on the sale of farm lands with the hope this would enhance their attractiveness as loan collateral. In 1992 the government initiated a third measure that provided temporary credit for funding farm inputs, called FONDEAGROS, that was managed by the Ministry of Agriculture in the form of loans in kind." From 1992 to 1995 the government committed about $US230 million to these programs (Falconi). A companion program called Fondos Rotatorios was the fourth measure that involved 16This minister was earlier head of BAP and later Minister of Agriculture during the period when BAP experienced rapid growth in number of employees, number of offices, and in use of donor funds. "Most of the rural credit unions and mutuales collapsed during the 1980s. 1Only a minor portion of these loans were repaid, approximately 10 percent. The funds recovered rolled into a Fondo Revolvente de Apoyo al Sector Agriario (FRASA) that is administed by COFIDE and largely on-lent to Cajas Rurales. 11 PERU: An A8ncalural Developient Strategy 12 Annex 3 government expenditures of about $US 92 million from 1992 to 1995 (Falconi). It was also managed by the Ministry of Agriculture and provided loans in areas outside the coast at subsidized interest rates. The fifth measure was to promote through the Ministry of Agriculture the formation of private Cajas Rurales de Ahorro y Credito (CRAC). By 1997 these efforts had resulted in only a partially rebuilt rural financial system with major gaps in it. Compared to 1980, formal rural financial in Peru in 1997 was much more diffused and involved a larger cast of institutions. The gap in lending to farmers left by the liquidation of BAP has been only partially filled by the 18 new fledgling Cajas Rurales, by a small amount of lending by a few non-governmental organizations, by some tentative steps by Cajas Municipales into agricultural lending, by ad hoc government lending programs administered by the Ministry of Agriculture, by commercial bank lending, and by funding for agricultural purposes provided by the second-story bank COFIDE. Caias Rurales With the liquidation of BAP, many policy makers felt that new private banks could be formed in rural areas relatively quickly to provide financial services to poor people. With the Ministry of Agriculture taking the lead, the first privately owned Caja Rural de Ahorro y Credito (CRAC) was opened in late 1993 and an additional 8 Cajas started business the next year (Alvarado and others 1996). To help fund these new banks, the Ministry of Agriculture offered them compensation for collecting defaulted loans made earlier under FONDEAGROS. In early 1997 there were 18 Cajas Rurales with a total loan portfolio of about $US 44 million, primarily funded by COFIDE (Falconi). Two of these Cajas have since been liquidated. As might be expected, CRACs are encountering substantial problems. They are undercapitalized, have difficulty mobilizing deposits, continue to depend heavily on "outside" funding, some have serious loan recovery problems, their scope and range of lending is sharply restricted, and about a quarter of them are not yet profitable. In mid- 1996 overdue loans comprised 16 percent of the total loan portfolio of the Cajas with three of them having rates in excess of 50 percent (Alvarado and others 1996, pp. 30- 31). It is particularly disturbing that the Cajas finance less than one-quarter of their loans through deposited funds. Few of the Cajas are seen as being creditworthy by other segments of the financial system, or even by potential depositors. The problems in the CRAC are, in part, a hangover from the BAP: many of the employees working for the Cajas formerly worked for BAP; a number of the Cajas are housed in BAP buildings, many of the owners of the Cajas earlier defaulted on BAP loans, most of the clients of the Cajas were former clients of BAP, Caja employees use many of the lending procedures they learned in BAP, and the Ministry of Agriculture tends to view the Cajas as an extension of Ministry activities, much like the view they had early of BAP. In large measure, the CRAC have been build out of rubble left by the collapse of BAP and it is understandable why the Cajas have some of the same flaws that led to the demise of the BAP. They are high cost operations that are heavily dependent 12 PERU: An Agrcultural Development StrUaey 13 Annex 3 on COFIDE funding (Daly). The legal limits on the types of loans they can make and the restricted geographical areas within which they must work also severely limit the ability of these fledgling banks to diversify their lending risks. Until these flaws are corrected, it is unlikely that the Cajas Rurales will become the backbone of a new rural financial market in Peru.19 Cajas Municipales The past several years some Cajas Municipales de Ahorro y Credito (CMAC) have begun to expand into rural areas through opening new branches and by placing some emphasis on attracting new clients who are farmers (Alvarado and others 1996, pp. 50-61). In mid-1996 agricultural loans comprised about 13 percent of all loans made by the 12 CMAC and about half of them did at least some agricultural lending. The CMAC in Piura, Ica, and Sullana are the ones most heavily involved in loans to farmers. The CMAC in Ica operates in an areas with about thirteen thousand farmers, the majority of whom own less than 3 hectares. The Caja began to make agricultural loans in 1994 and by mid-1996 had 330 farmer borrowers with outstanding loans amounting to about US$400 thousand for an average loan size of approximately US$1,300. These loans made up one-fifth of the Caja's total loan portfolio and managers of the office were planning to further expand lending to farmers until about one-third of the Caja's loans were to farmers, possibly reaching as many as a thousand farmer borrowers. Most loans were secured by real estate collateral and most loans were for less than a year. The CMAC in Sullana started making agricultural loans in 1993. In mid-1996 the Caja had about 2,300 farmer borrowers with a total loan value of about US$1.7 million. Agricultural loans comprised about one-quarter of the CMAC's total loan portfolio and the average size of these loans was US$1,200. Most of these loans involve collateral in the form of real estate, and loans were seldom made for more than 12 months. Several aspects of the CMAC's activities are noteworthy. It is encouraging that the CMAC are increasingly seeing farm lending as profitable business. Through their earlier experience with making small loans in urban areas, these Cajas have developed techniques and practices that allow them to made a profit on relatively small loans. Some of the strongest and most aggressive CMACs will likely continue to open new branches and attract more farmer clients. At the same time, it is disappointing that the CMACs are not mobilizing more deposits in towns and in rural areas. Their deposit-to- loan ratio is only .6-.7, well below what might be expected from financial institutions such as the CMAC. In part, the availability of outside funding may have deflected these Cajas from placing more emphasis on attracting deposits. Their negative experience during the 1980s with deposit mobilization may have caused leaders to be too skeptical ' For additional background on the Cajas Rurales see Alvarado 1993; Alvarado and others 1996; Aznaran 1991 and 1994; Daly; Huaman; Lazarte; Valdivia. 13 PERU: An Agiculural Developmnt Strat 14 Annex 3 about the possibilities of attracting deposits in the more favorable economic environment of the 1990s.20 Commercial Banks After being largely driven out of agricultural lending by BAP, combined with disruptions caused by land reform and rural violence, commercial banks have begun again to lend for agricultural purposes. As can be noted in Table 2, the total volume of this lending increased ten-fold from only about US$ 30 million in 1990 to almost US$ 300 million in mid-1997.21 Five of the 23 commercial banks in Peru provided more than 85 percent of this agricultural lending in 1997: Banco de Credito, Banco Wiese, Banco Sudamericano, Banco Continental, and Interbank. TABLE 2: AGRICULTURAL LENDING BY COMMERCIAL BANKS IN PERU, 1990 to mid-1997 (in current Million US$) Bank 1990 1992 1994 1996 mid-1997 Banco de Credito 19 34 54 121 134 Banco Wiese 15 33 38 53 Banco Sudamericano 2 18 27 Banco Continental 2 20 29 20 21 Interbank 4 3 17 20 Other banks 5 17 21 41 43 Totals 30 86 142 255 298 Source: Superintendencia de Bancos y Seguros, Informacion Financiera Mensual, Junio, 1997. Interviews with officials in these five banks in November, 1997 revealed the following: The Banco de Credito had an outstanding balance in agricultural loans in October, 1997 of US$206 million, all secured with real estate collateral, and most of this lent in the coastal region. Of this, about US$20 million was in loans to millers, who probably provided informal loans to farmers. The bank had other loans totaling US$10 million to intermediarios (people who knew agriculture, NGOs, and religious groups) who then lent to farmers or to groups of farmers informally with the bank securing the loan through land mortgages. All of these intermediary loans were made in the coastal region and the bank was experiencing loan recovery problems on more than one-third of these loans. Most of these loan recovery problems, however, were due to non-payment by intermediaries, rather than loan default by farmers. It was common practice for the bank to require a compensatory balance equal to 15 percent of the value of these loans. 20 For more background on the Cajas Municiipales see: Burnett; Instituto Peruano de Economia; IPC 1994 and 1996; Lepp; Tello; Zeitinger. In late October, 1997, Cajas Municipales were allowed to convert to sociedades anonimas by and SBS Circular if they choose to do so. 21 This includes loans made to agro-industries, but does not include loans made to coffee processors (acopiadores). These latter loans are classified in Peru as loans made to the commercial sector. 14 PERU An Agricultura Development Strategy 15 Annex 3 Another US$80 million was lent to agro-industries, and many of them were likely involved in buying or selling agricultural goods and inputs on informal credit. In addition, the bank had US$96 million outstanding in loans to farmers, all of whom had more than 20 hectares of land. Most of the bank's agricultural loans were for periods of less than a year. As an aside, the bank also lends to some Cajas Municipales, who in turn, may make a few loans to farmers. Banco Wiese is the second-largest commercial lender for agricultural purposes with an outstanding loan portfolio of US$53 million in mid-1997, most secured by real estate collateral These loans were of three types: loans to assist land sales by sugar producers, credit for agro-industrial exporters, and loans for small farmers. Only 15 percent of the bank's "agricultural loans" went to small enterprises and a substantial part of this likely went into non-farm microenterprises through loans to NGOs. The bank was experiencing loan recovery problems on about 9 percent of the loans made to microenterprises and small farmers, compared to loan recovery problems of about 5 percent in the rest of its loan portfolio. Almost all of the bank's agricultural lending was in the coastal region and the bank had few employees who were experienced in agricultural lending. Bank officials had little interest in offering more deposit services in rural areas and preferred to work through NGOs in its microcredit activities. Most of the bank's agricultural loans were for periods of less than a year. The third largest commercial lender to agriculture is Banco Sudamericano with outstanding agricultural loans totaling US$27 million in mid-1997. It is a small bank that has offices only in Lima. It's lending focuses on the top end of the market, and the bank makes no loans directly to farmers. The bulk of its agricultural lending is for specialized agricultural exports such as asparagus and its loans are made to the buyers (acopiadores) who manage the production, packing, and export of the product. It is common practice for the buyers to provide informal loans to participating farmers, and in some cases, to also provide technical assistance. The bank requires real estate collateral on its loans, most of its loans are for less than a years, and all of the loans are made in Lima. Its agricultural loans are all denominated in US dollars. Banco Continental had agricultural loans outstanding amounting to US$21 million in mid-1997. About 60 percent of this was in loans to relatively large farmers who secured their loans with real estate collateral. Another 10 percent was in loans to associations of farmers or to intermediaries (colectores) who, in turn, made informal loans to small farmers. The remaining 30 percent of the loans was made to marketing intermediaries (acopiadores) who may extend informal loans to farmers. Three-quarters of all of the bank's loans were in the coastal region, most loans were for less than 9 months, and only 5 percent of these agricultural loans had recovery problems. The bank was also lending to several Caja Municipales that may have been making some agricultural loans. In mid-1997 Interbank had outstanding agricultural loans amounting to about US$20 million. Approximately 40 percent of this was in loans to farmers with a minimum of 20 hectares of land and the other 60 percent was in loans to agro-industrial 15 PERU: An Agricuhural Development Stmtegy 16 Annex 3 firms such as Nestle which the bank called loans by triangulo. Most of these agro- industrial firms provided informal credit, often in kind, to farmers to completed the triangle. Most of its agricultural loans are for less than a year, are made to firms operating largely in the coastal region, and all of the loans involve collateral in the form of real estate. Although the bank has offices throughout the country, it does not stress deposit mobilization outside Lima. Rural deposits are seen as being too costly to mobilize, especially while many of the banks in the country have excess liquidity. Interbank is thinking about closing some of its rural offices because they do not generate sufficient revenue to cover their costs. Overall, the interviews with officials in the five banks led us to the following conclusions: * something over one-third of their agricultural loans were made directly to farmers, * most of their agricultural loans went to marketing intermediaries who were involved with handling agricultural commodities or inputs, * many of these borrowers, in turn, made informal loans to farmers, * many of the agricultural loans made by commercial banks were associated with exports and imports, * most of the loans were extended to firms that concentrated their activities in the coastal regions, * most of the agricultural loans were for periods of less than one year, * and managers of these banks did not look to rural areas as being attractive markets for deposit services. As is the case with the Cajas Rurales, the commercial banks are still far from providing the backbone for a rural financial system in Peru. Of the 884 bank offices in the country nearly 90 percent are located in the coastal region--60 percent in Lima alone- -and only slightly more than 10 percent are located in the interior of the country. Commercial banks are remote from providing loans directly or indirectly to the bulk of the farmers in the country and they are doing little in the way of providing attractive deposit services outside cities. At the same time, commercial banks are experimenting with ways to expand their lending to a heterogeneous agricultural sector. COFIDE COFIDE is partially an artifact of the old directed credit paradigm. It manages lumps of money provided by donors or government that often have loan targeting objectives. It is an improvement, nevertheless, over the old practice of placing targeted funds in the central bank and then passing them along to final lenders via concessionary lines of credit. The COFIDE arrangement removes the Central Bank from promotional activities, lessens the disincentives that outside money have on deposit mobilization, and limits the transaction costs associated with managing multiple lines of funding from outside the country. Even though COFIDE charges market rates of interest on its loans 16 PERU. An Agicultural Development Strategy 17 Annex 3 to other elements in the financial systems, such as the Cajas Rurales and the Cajas Municipales, its loans are usually cheaper for these organizations to use than it would be to mobilize similar amounts through voluntary deposits of relatively small amounts. In part, the tepid deposit mobilization performance of both the Caja Rurales and the Cajas Municipales is due to their access to COFIDE funding. Inadvertently, COFIDE is reinforcing the historical tendency in Peru for rural financial markets to stress lending and to ignore mobilizing deposits. Since COFIDE does no retail lending, its credit activities have only an indirect effect on the supply of rural loans. About half of the loans made by the Cajas Rurales are funded by COFIDE and some unknown portion of that ends up in loans to farmers. Some of COFIDE's loans to other segments of the financial system likely seep into farm loans. COFIDE officials roughly estimate that approximately US$20 million of its total of US$200 million in loans outstanding in mid-1997 went into agricultural loans. Gaps in the system Despite some progress in rebuilding the rural financial system in Peru since its low point in 1990, major gaps persist in these markets. These include a paucity of medium- and long-term loans for agriculture, limited credit for non-export enterprises, little formal lending done in the interior of the country, too few loans available for non- farm rural firms, very limited deposit services for rural people, and almost no lending that is done without real estate collateral. These gaps have important economic and social implications. The lack of longer term loans limits the ability of the agricultural sector to make the investments required to modernize farming and to compete in international markets. The limited access that food-producing farmers have to formal finance has long-run implications for future food prices and import requirements. Although financial markets have proved to be weak instruments for directly addressing poverty, the social implications of what is done in these markets should not be ignored. Peru was wracked by three decades of turmoil that had its roots in rural poverty. Unless the well-being of many more poor people living in the interior of the country can be improved the probability of that turmoil re-erupting is substantial. Efficient financial markets should provide attractive deposit services to rural people that are paths out of poverty through saving more. Likewise, an efficient financial market could provide an increasing number of creditworthy people in rural areas--both farmers and non-farmers-assistance in expanding their businesses and incomes. 17 PERU: An Agricultural Development Stragy 18 Annex 3 POLICY OPTIONS Experience in other countries suggests that rebuilding rural financial systems takes time and patience, including building a large measure of faith in free markets.22 It also involves setting clearly defined goals, avoiding actions that weaken these markets, and stressing policies that encourage the system to become more efficient and to expand coverage. In many cases, the most challenging thing is to resist actions stemming from direct-credit addiction that previously undermined efforts to build efficient and durable financial infrastructure in rural areas. This addiction includes political leaders, congressmen, and even donors trying to ameliorate a wide variety of problems with subsidized and directed credit: attempting to use subsidized credit as a second-best alternative to offset the adverse effects of other policies, and using cheap credit to treat poverty. Policy makers must deflect efforts to return to the days of directed credit, enhance the macroeconomic environment, treat rural areas fairly, and assure that formal financial markets are prudentially regulated and supervised. Setting goals In large measure, the options that are selected to address problems in rural financial markets depend on how problems are defined. Traditionally, rural financial market programs and policies were based on market-imperfection assumptions. Over-reliance on collateral, too few loans for poor people, too few loans for farmers, too few medium- and long-term loans, too few loans for women, asymmetric information problems between lenders and borrowers, and exploitation by informal lenders were often cited as signs of these imperfections. These alleged imperfections were then commonly used as justification for government or donor programs of directed and subsidized credit to overcome these market short-comings. Those attempting to provide directed credit commonly overlooked the effect of these programs on the vitality and durability of financial infrastructure, on the incentives in the system to mobilized deposits, as well as the substantial imperfections that are often present in government- and donor-sponsored efforts. More recently, concerns about risks and high transaction costs have increasingly replaced the market-imperfection preoccupation. Those supporting these new views argue that rural finance is inherently costly and risky and that these factors--not market imperfections-- block financial markets from embracing more rural clients. The policy prescriptions that logically follow from these new views focus on reducing the costs of doing rural finance and dealing more effectively with the associated risks. Supporters of these new views argue that subsidized and direct credit usually increases, rather than decrease these costs and risks. In contrast to the traditional approach, the vitality, efficiency, and durability of the financial infrastructure is of primary concern to those promoting the new views. ' A highly successful reform and rebuilding of a formal rural financial system in Indonesia, for example, took approximately 15 years. 18 PERU: An Agncultural Development Strafegy 19 Annex 3 Through their actions it appears most policy makers in Peru support these new views If this support is sustained, which is ought to be, the government has two primary options to deal with problems in rural financial markets. The first option is to pursue a policy of benign neglect based on the hope that unfettered private actions will eventually fill most of the gaps that currently exist in rural financial markets. The other option is for the government to take concerted actions to stimulate building of additional rural financial infrastructure, along with other activities that reduce the transaction costs and risks associated with rural finance. The Option of Benign Neglect Other things being equal, commercial banks, Cajas Municipales, and possibly Cajas Rurales will continue expanding their agricultural lending. This expansion, in turn, will most certainly result in further expansion in the amount of informal lending done in rural areas, and at least some of this informal lending will be to the rural poor.2 This cascading of loans from formal to informal through marketing intermediaries, however, is unlikely to ever fill all of the important gaps that are present in Peru's rural financial markets. Informal finance typically does not provide relatively large loans, medium- and long-term loans, or attractive deposit opportunities. It will likewise be impossible for semi-formal and formal forms of finance to provide loans directly to a much large number of farmers until more financial infrastructure is built in rural areas. It may be appropriate for government to explore ways to accelerate that building through carefully selected actions. Another concern with the benign neglect approach is that there is no formal mechanism to protect the rural financial system against recidivism to directed and subsidized credit during elections or in times of natural disaster. The Option of Concerted Actions Five major elements might be included in government efforts to accelerate improvements in rural financial markets. The first to assign responsibility to someone for fostering the well-being and growth of these markets. The second is to elicit donor support for elements of a rural financial markets development program. The third is to focus on strengthening and possibly rebuilding some financial institutions in rural areas. The fourth element is an organized effort to reduce transaction costs in rural finance. The fifth element is a parallel effort to reduce the risks of rural lending through legal and court reforms aimed at easing loan collateral problems. 2 The informational advantages that informal agents have over banks allows them to screen potential borrowers for creditworthiness more cheaply than can more distant formal lenders. These informal lenders may also employ loan recovery techniques that are more effective than the techniques available to most formal lenders. For example, a merchant who sells fertilizer to a client on credit may be able to deny that client access to chemical fertilizers in the future if loans are not repaid. 19 PERU: An Agricultural Development Strateg 20 Annex 3 Leadership and Responsibility One of the results of the liquidation of BAP is that no one in Peru feels responsible for representing, maintaining, or fostering rural financial markets. Through efforts of the President and others in Peru--with support from the IDB--there is a more clearly defined support group that promotes the microcredit industry in Peru, for example. In several countries, including the Philippines and Uganda, an inter-agency committee has the responsibility for tending rural finance. Ideally, this committee is chaired by a Minister whose portfolio includes responsibility for maintaining the well- being of financial markets. In many cases this is the Minister of Finance. Other members of the committee might include representatives from the Ministry of Agriculture, the superintendent of banks, commercial banks, Cajas Rurales, Cajas Municipales, Credit Unions, COFIDE, and possibly even NGOs. In addition to defending the interests of rural financial markets in political arenas, the committee should also be responsible for formulating and carrying out a development strategy for rural financial markets. Segments of this strategy could then be offered to donors who wished to participate in the development of these markets. In addition to fostering more financial infrastructure, the committee might also focus attention on other important issues in rural finance such as how to lower transaction costs and how to reduce lending risks. The committee's responsibilities might likewise include fostering training and research that supports the development of efficient rural financial markets. Donor Coordination The committee would also be responsible for assuring that donor efforts were coordinated and that only practices and policies that enhanced the overall performance of rural financial markets would be allowed. This might include insisting on cost-recovery rates being charged on loans, reporting and evaluation requirements that do not substantially increase transaction costs in these markets, and pricing of funds such that deposit mobilization is encouraged. Ideally, the committee would be proactive in seeking donor support and setting the boundaries for that assistance. Building or Strengthening Rural Financial Infiastructure It is readily apparent that more bricks and mortar are needed to expand rural finance in Peru. Many of the villages and communities that once had a BAP office or a credit union remain without formal financial services, especially access to deposit services. With proper encouragement, donors may be willing to help with building or rebuilding some of this vital infrastructure. For example, the Germans might be invited to help strengthen Cajas Rurales, much like they did earlier with the Cajas Municipales; USAID might be invited to help rebuild rural credit unions to assist with deposit mobilization in rural areas; and the IDB might be invited to expand its microlending efforts to extend more loans to non-farm rural enterprises. The World Bank might be encouraged to assist with court and legal reforms, with communication enhancements that lower transaction costs, and with providing support to the Inter-Agency Committee. 20 PERU: An Aricultural Development State 21 Annex 3 Reducing Transaction Costs As mentioned earlier, rural finance is costly, partly because many of the transactions are small. All financial intermediaries are impeded from doing more rural finance by these costs. New practices and state-of-the-art electronics is needed to drive down these costs throughout the rural financial system. In Malaysia, for example, a reformed agricultural development bank rapidly computerized and tied its communication together through satellite up-links to dramatically reduce transaction costs. In other countries credit cards and smart cards are being used to lower transaction costs involved in microlending. Similar innovations are needed in Peru and an inter-agency coordinating committee might give priority to promoting such innovations. Properly packaged these efforts might attract donor support. In addition to directly addressing transaction costs, the inter-agency committee should monitor and deflect government and regulator actions that boost transaction costs in rural financial markets. Reducing Lending Risks Understandably, lenders in Peru are highly risk averse. They prefer to lend short because of their experience with inflation, and lend against real estate collateral because of serious loan recovery problems during the 1980s and early 1990s. These bitter memories will only fade with time. Memories are reinforced by legal and court problems, however, that strengthen these conservative tendencies regarding collateral. Accelerated land titling, changes in laws that would allow banks to accept a broader range of collateral--cattle, machinery, and inventory, for example--along with more efficient courts would be steps in easing collateral difficulties. An Inter-Agency Committee that represented rural financial markets might crystallize and focus attention on these problems. In some countries these collateral problems are ameliorated through leasing arrangements, especially of equipment. These and other financial innovations ought to receive attention by the group who has responsibility for tending rural finance. Fostering leasing agencies may be one way of overcoming the lack of medium- and long- run lending for farmers in the near future. Loan guarantee programs are another popular form of dealing with lending risks. Serious questions, however, have been voiced about the costs and benefits of these programs (Llisterri and Levitsky). Critics argue that supporters of these programs have provided little evidence showing that loan guarantee schemes cause much additionality in lending in directions desired by policy makers. Critics also note that most of these schemes require periodic subsidies to keep them afloat, that most of these subsidies are captured by lenders, and that these programs inevitably increase, rather than decrease, 24 See World Bank 1997 for a fuller discussion of these issues. 21 PERU: An Agicultural Development Sraegy 22 Annex 3 transaction costs. Until more empirical evidence on the benefits and costs of these programs is available, Peruvians should move slowly in using loan guarantee schemes in attempts to deal with the risks of rural lending. CONCLUDING COMMENTS Government and donor efforts to support rural finance under the new paradigm must be more subtle, innovative, and done with more discipline than was the case with support under the old directed-credit paradigm. Under the old paradigm, perceived gaps in the rural financial system were filled by directing large amounts of outside money at these gaps. It was relatively easy for political leaders to announce a large infusion of funds into the system to address pressing problems, or for donors to package relatively large amounts of money into projects aimed at overcoming some market imperfections or filling some gap in rural financial markets. Unfortunately, these practices have proved to be inconsistent with building durable and efficient rural financial systems. It is increasingly recognized that sustained access to attractive financial services is much more important for the clients of these services, than are any short-run subsidies that might be attached to these services. Like education systems, power grids, and roads, financial markets are comprised of infrastructure whose maintenance is important for the well-being of Peru's economy. It is paradoxical that efforts to improve rural financial markets under the new paradigm (that stresses using market forces), must be more carefully planned than was the case with the old paradigm (that was often associated with central planning). Ad hoc programs that insert additional outside money into the system will discourage deposit mobilization, increase transaction costs, and deflect rural financial markets from filling important gaps. Peruvians and donors alike must be innovative in the ways they deal with these markets. Large donors must be particularly creative in repackaging their assistance so they can make relatively large loans, but at the same time design these activities so they do not weaken incentives that are critical in the development of rural financial markets. Including activities aimed at fostering rural financial systems in large program loans aimed at the overall agricultural sector may be one way of doing this. 22 PERU: An Aicutura Development Straegy 23 Annex 3 BIBLIOGRAPHY Alvarado, Javier. 1991. Ahorros en la Peguena Agricultura. Debate Agrario No 12, Centro Peruano de Estudios Sociales (CEPES), Lima, Peru. Alvarado, Javier. 1993 Cajas Rurales y Fondos Rotatorios: Soluciones o Mitos para el Financiamiento Rural en El Peru. Debate Agrario No. 16, Centro Peruano de Estudios Sociales (CEPES), Lima, Peru. Alvarado, Javier. 1988. "Credito Agrario: Limites de un Instrumento Financiero." In Debate Agrario, October-December. Alvarado, Javier. 1995. La Innovacion en las Tecnologias Crediticias. Debate Agrario No. 21, Centro Peruano de Estudios Sociales (CEPES), Lima, Peru. Alvarado Guerrero, Javier. 1997a. "El Financiamiento Rural Peruano: Experiencias Y Propuestas Para La Constitucion de Un Sistema Formal." Unpublished study prepared by the Centro Peruano de Estudios Sociales (CEPES), Lima, Peru, February. Alvarado Guerrero, Javier. 1997b. "La Technologia Crediticia de Los Prestamistas Informales en El Ambito Rural." Unpublished study prepared by the Centro Peruano de Estudios Sociales (CEPES), Lima, Peru, July. Alvarado, William Diaz and others. 1996. "Los Mercados Financieros Rurales en El Peru Y Posibilidades de Su Promocion." Unpublished study prepared for GTZ by Luso Consult, Hamburg, Germany. November. Apoyo Consultoria S.A. 1997. El Sistema Financiero en El Peru: Situacion Actual y Perspectivas." Unpublished report prepared by Apoyo Consultoria S.A., April. Aznaran, Castillo G. 1991. "Cajas Rurales de Ahorro y Credito." In Avance Economico, September. Aznaran, Casitilo G. 1994. "Que Pasa con Las Cajas Rurales?" In Avance Economico, February. Banco Agrario del Peru. Memoria, various years, Lima, Peru. Bonicelli, Carlos Otero y Hugo Wiener Fresco. 1996. "Financiamiento de La Pequena y Microempresa: Una Propuesta Alternativa. Unpublished paper, Instituto Peruano de Economia, Lima, Peru. Bravo, Adolfo. 1991. "Financiamiento a La Microempresa A Traves de Los Grupos Solidarios: La Experiencia de Accion Comunitaria del Peru." In Pequena Empresa 23 PERU: An Apicultural Development Strateg 24 Annex 3 Desarrollo Moderno, edited by Baltazar Caravedo and Fernando Villaran. Lima, Peru: Pemtec Insotec Comunidec Sase. Burkett, Paul. 1981. "Savings Mobilization in The Third World: A Case Study of Peru." Unpublished paper, Department of Economics, Syracuse University, August. Burnett, Jill. 1996. "The Cajas Municipales of Peru." Unpublished paper prepared for the World Bank, Washington, D.C. January 22. Carranza, Octavio. 1973. "Small Farmer Supervised Credit in Peru," in Small Farmer Credit in South America, A.I.D. Spring Review of Small Farmer Credit, Vol. 3, Washington, D.C.: Agency for International Development, February. Chaparro, J. Javier. 1995. "El Financiamiento Rural en El Peru: Limitaciones y Perspectivas." Unpublished paper prepared for the World Bank by Ecological Services for Planning Limited, Guelph, Ontario, Canada, August. Daly, J.L. Inc. 1995. "Informe Final: Plan de Accion para el Fortalecimiento del Sistema de Cajas Rurales de Ahorro y Credito." Unpublished paper prepared for the Peruvian Government, Lima, Peru, October. Falconi, Cesar. 1997. "Nota Sobre El Financiamiento Rural en El Peru: Situation Actual y Alternativas." Unpublished report prepared in COFIDE, Lima, Peru, June. Gadway, John. 1979. "El Impacto de La Inflacion sobre La Cooperativa de Ahorro y Credito, Constitucion Ltds. No. 51: Analysis y Recomendaciones." Unpublished report prepared for USAID Mission, Lima Peru, November. Gadway, John. 1980. "La Inflacion y La Politica de Prestamos a Bajo Interes." Unpublished report prepared for USAID Mission, Lima Peru, January. Gadway, John. 1979. "La Situacion Actual de La Cooperativa de Ahorro y Credito Protectora Ltda. 261: Analisis y Recomendaciones" Unpublished report prepared for USAID Lima, Peru, October. Gonzales. A. 1989. "Subsidios Financieros a La Agricultura en El Peru 1980-1988." In Debated Agrario, April-June. Gonzales de Olarte, Efrain. 1994. "En las Fronteras del Mercado: Economia Politica del Camesinado en el Peru." Unpublished study done by the Instituto de Estudios Peruanos, Lima, Peru. Government of Peru. 1968. "Banco de Fomento Agropecuario," Ley Organica: Coordinada con Ley de Bancos, Decreto Supremo No. 355-68-HC de 16 de Agosto. Government of Peru. 1963. "Normas Para el Funcionamiento del Banco de Fomento 24 PERU. An Ancultura Developmet Staeg 25 Annex 3 Agropecuario del Peru," Decreto-Ley No 14509, June 14. Hamann de Vivero, Jacinta and others. 1996. "Informe Sistema Financiero Nacional, Segundo Trimestre 1996." Unpublished report by COFIDE, Lima, Peru, Huaman, Garibay M. 1993 "Cajas Rurales No Son Suficientes." In Actualidad Economica del Peru, September. Gadway, John. 1979. "BANCOOP in Huancayo Despues de 10 Meses: Algunas Observaciones." Unpublished report prepared for USAID Mission, Lima, Peru, November. Heath, John and Surajit Goswami. 1996. "Peru: Strengthening Rural Financial Intermediation." Unpublished report prepared by the World Bank, Latin America and the Caribbean Regional Office, Washington, D.C., April 24, 1996. Heywood Fleisig & Nuria de la Pena. 1996. "How Problems in The Framework for Secured Transactions Limited Access to Credit in Agriculture." Unpublished report prepared for the World Bank, Washington, D.C. Instituto Peruano de Economia (IPE). 1996. "El Finaniamiento a la Microempresa en El Peru y Las Cajas Municipales de Ahorro y Credito." Unpublished manuscript prepared by IPE in Lima, Peru, September. Inter-American Development Bank. 1983. "Banco Agrario del Peru: Evaluation of Global Agricultural Credit Programs." Unpublished report prepared by the Operations Evaluation Office, Office of the Controller, Inter-American Development Bank, Washington, D.C., June. Inter-American Development Bank. 1995. "Proposal for A Loan for A Globabl Microenterprise Credit Program." Unpublished report prepared by the Inter- American Development Bank, Washington, D.C. September 26. Interdisziplinare Projekt Consult (IPC). 1994. "Evaluacion de las Institucions Proveedoras de Servicios Financieros al Sector Micro y Pequeno Empresarial en El Peru." Three volumes. Unpublished study prepared by IPC for the Inter-American Development Bank, Washington, D.C. October. Interdisziplinare Projekt Consult (IPC). 1996. "Informe Final de la Asesoria Tecnica en El Area de Creditors a La Microempresa de Las Cajas Municipales del Peru." Lima, Peru" IDP. Larson, Donald W. and Dale W Adams. 1981. "Evaluation of USAID-BANCOOP Pilot Project on The Development of Rural Financial Institutions in Peru." Unpublished study prepared for the Agency for International Development in Lima Peru, May 8. 25 PERU: An Agricultural Development Srateg 26 Annex 3 Lazarte, J. 1996. "Situacion de Las Cajas Rurales de Ahorro y Credito." In Debate Agrario June. Lepp, Anja. 1996, "Financial Products for MSEs--The Municipal Savings and Loan Banks of Peru," Small Enterprise Development 7: 15-25. Llisteri, Juan Jose and Jacob Levitsky editors. 1996. Sistemas de Garantias de Credito: Experiencias Internacionales y Lecciones para America Latina y el Caribe. Washington, D.C.: Inter-American Development Bank. Ministerio de Economia y Finanzas. 1991. "La Banca de Fomento en El Peru: Diagnostico y Soluciones," Unpublished study prepared by the Ministry of Economics and Finance, Lima, Peru, August. North Carolina State University Mission. 1968. "Evaluation Program on Supervised Credit." Unpublished report prepared by the North Carolina State University Mission in Peru for the Ministry of Agriculture, Lima Peru. Palomino Buleje, Luis. 1993a. "Credito Agrario: Una Historia Inconclusa." Agro Empresa, (Revista del Comite Nacional de Productores de Menestras), pp, 28-32. Palomino Buleje, Luis. 1993b. "Dos Decadas de Credito Agrario," Serie Investigaciones. Universidad de Lima, Facultade de Economia, No. 10, September. Palomino Buleje, Luis. 1991 "Fusion de La Banca de Fomento: Reflexiones." Unpublished paper, Faculty of Economics, University of Lima, Lima, Peru, November. PYME: Periodico de la Pequena v Micro Empresa. Various years. Lima, Peru. Riordan, James T. and others. 1994. "Food Security Strategy for Peru." Unpublished report prepared for the Agency for International Development by Agricultural and Rural Development Technical Services Project, Washington, D.C. December. SECREX. 1996. "Credito a la Pequena y Microempresa: Proyecto del Sistema de Seguros de Credito para Pequenas Empresas." Unpublished proposal prepared by SECREX for the InstitutoPeruano de Economia, 25 June. Sotomayor, Narda. 1994. "Credito Rural en Peru." Unpublished paper prepared in the Superintendencia de Bancos y Seguro, Lima, Peru. Superintendencia de Banca y Seguros (SBS). Various years. Informacion Financiera de Cajas Municipales y Caias Rurales. SBS, Lima, Peru. Superintendencia de Banca y Seguors (SBS). Various years. InformacionFinanciera Mensual. SBS, Lima, Peru. 26 PERU: An ApicuurlW Development Suategy 27 Annex 3 Superintendencia de Banca y Seguros (SBS). Various years. InformacionFinanciera Ouincenal. SBS, Lima, Peru. Tello, Maria del Pilar. 1995. "Promesas de Progreso: Cajas Municipales de Ahorro y Credito del Peru." SAGSA, Lima, Peru. Truitt, Nancy Sherwood. 1996. "Microlending to Peru's Poor Is Becoming Big Business" Wall Street Journal September 20, 1996. page. Al5 Valdivia, Martin. 1996. "Del Banco Agrario a las Cajas Agrarias: Pautas para la Construccion de un Nuevo Sistema Financiero Rural." In Boletin de Opinion, April. Vogel, Robert C. 1984. "Savings Mobilization: The Forgotten Half of Rural Finance." In Undermining Rural Development with Cheap Credit, edited by Dale W Adams and others, Boulder, Colorado: Westview Press, pp. 248-265. Vogel, Robert C. and Dale W Adams. 1997. "Old and New Paradigms in Development Finance: Should Directed Credit be Resurrected?" CAER II Discussion Paper No. 2, Harvard Institute for International Development, Harvard University, Cambridge, Massachussets, April. Vogel, Robert C., Dale W Adams, and Juan Jose Marthans. 1997. "Approaches to Rehabilitating Insolvent Government-Owned Banks: Benefits and Costs of Liquidating an Agricultural Bank in Peru." Unpublished paper, IMCC, Washington, D.C., October. World Bank. 1993. "A Review of Bank Lending for Agricultural Credit and Rural Fiance (1948-1992)." Washington, D.C.: World Bank, Operations Evaluations Department. World Bank. 1997. "Peru: How Problems in The Framework for Secured Transactions Limits Access to Credit." Report number 15696-PE, Natural Resources Management and Rural Poverty Country Department m, Latin America and The Caribbean Regional Office, World Bank, Washington, D.C., June 6. World Bank. 1980. "Peru: Fifth Agricultural Credit Project-Loan 933-PE Completion Report." Unpublished report prepared by the Agricultural Division 4, LAC Projects, World Bank Washington, D.C., June 4, 1980. World Bank. 1981. "Project Performance Audit Report: Pem Fifth Agricultural Credit Project (Loan 933-PE)." Unpublished report prepared by the Operations Evaluation Department, World Bank, Washington D.C., May 21. World Bank. 1990. "Project Completion Report: Peru Sixth Agricultural Credit Project (Loan 2302-PE)." Unpublished report prepared by the Agricultural Operations 27 PERU. An Agricultural Development Stnuegy 28 Annex 3 Division, Country Department IV, Latin America and the Caribbean Regional Office, World Bank, Washington, D.C., Zeitinger, Claus-Peter, 1996. "La Crisis Institucional del Sistema de Cajas Municipales de Ahorro y Credito: Causas y Propuestas para un Solucion. Unpublished paper, IPC. 28 CATALOGUERGIFILC FRÄF[ CONFIDENTIAL Report No.: 17270 PE Type: SR