1. Responses to Globalization. Differentiated Policies and Interventions
2. Poverty reduction Programs
3. Institutional Reforms and Decentralization
4. Institutional Reconstruction: The Transition from Public to Private Institutions
5. Sequencing of Reforms
In recent years, considerable emphasis has been placed on the implications of the heterogeneous nature of the rural population for the analysis of the differentiated impact of policy reforms and for the design of differentiated policies and interventions. Thus, while there has been an international process of homogenization of macropolicies to achieve stabilization and adjustment promoted by the Bretton Woods institutions and the "Washington Consensus", there have also been increasing, although yet mild and incomplete, attempts at identifying the differentiated microeconomic effects of these policies on heterogenous populations and at designing differentiated complementary micro policies and interventions. This emphasis on heterogeneity has its origin in several phenomena that are creating increasing demands for more differentiated government interventions:
(i) Responses to the homogenization of macropolicies and to the associated process of globalization have been highly uneven. Some rural groups have expressed strong opposition to the process, even when the national economic impact of the reforms had been positive. Indeed, local responses to globalization is one of the currently most divisive issues for the social science profession, with economists usually arguing in favor of the global efficiency gains created by specialization and trade and other social scientists and NGOs arguing for greater equity in the distribution of benefits, preservation of access to place (rurality, rootedness), defense of local control over community welfare, and reproduction of cultural identity.4 Negative local responses to the global policy reforms have threatened the political stability, and hence the sustainability, of the economic recoveries, sometimes triggering, as in the case of Chiapas, global economic recession and political destabilization.
(ii) Democratization and decentralization of governance have given heterogeneous constituencies more direct access to policy makers, allowing local governments to better respond to the differentiated demands for public interventions.
(iii) The rapid spread of non-governmental and grassroots organizations has increased the bargaining power of heterogeneous groups, allowing them to press more powerfully their demands for differentiated government interventions.
(iv) The rapidly increasing availability of detailed household-level survey data has allowed better characterization of the heterogeneous nature of the rural population. Analysis of these data helps identify the differential causes of poverty as well as the variety of options for exiting from poverty.4 Several international fora have addressed this issue, in particular the International Forum on Globalization, "The Social, Ecological, Cultural, and Political Costs of Economic Globalization" (George Washington University, May 10-12, 1996) and the 6th annual conference of the International Association for the Study of Common Property, "Voices from the Commons" (University of California at Berkeley, June 5-8, 1996).
How are governments able to respond to these demands for more differentiated interventions? Under ISI and before the debt crisis, the old policy tools were:
- Blunt: real exchange rate appreciation, price interventions, and restrictive monetary policies have broad macroeconomic effects.
- Cheap: import tariffs and export taxes generate government revenues. These government revenues in turn allowed to fund compensatory subsidies for agriculture and soft budget constraints in parastatal services to agriculture.
- Extensive: massive price and interest rate distortions compared to FMFT equilibrium, both through indirect (exchange rate overvaluation and industrial protection) and direct interventions (trade policy, forced procurement) (Krueger, Shiff, and Valdés).
- Allowed for differentiation through macro-policy instruments: multiple exchange rate regimes, targeted interest rate subsidies to selectively compensate from price distortions, quantity restrictions, multiplicity of tariff levels.
- Exposed to rent seeking: targeted compensations for price distortions were generally appropriated through the forces of the political economy, for instance credit subsidies, creating major inequities.
Under stabilization and adjustment policies as well as in the context of democratization and decentralization of governance, the new policy instruments are:
- Constrained by loss of control over the traditional trade policy instruments (due to membership to GATT, NAFTA, and/or regional integration schemes) and severely limited by restrictive monetary policies (due to conditionality lending or self-imposed orthodoxy in controlling inflation and avoiding the crowing-out of financial markets by government borrowing).
- Constrained by stringent budget restrictions, in part due to the loss of revenues from trade taxes, limiting the scope of government interventions, particularly through subsidies.
- Better prone to co-production or co-management between government and heterogeneous constituencies as a consequence of increased participation and local definition associated with decentralization, democratization, and the rise of NGOs and GROs.
The key question regarding the usefulness of differentiated policies is whether they can be:
- More efficient as policy tools than non-differentiated interventions. A positive theoretical answer would be as follows. For efficiency purposes, government intervenes in response to market failures (e.g., economies of scale for ISI). If market failures are differential across economic agents, then the nature of compensatory government interventions should be correspondingly differentiated. Government interventions in compensation to market failures could thus be made more efficient.
- More equitable than non-differentiated policies. Again, from a theoretical standpoint, governments intervene when the welfare consequences of market forces are not desirable, in general because they leave too much poverty or create too much inequality. With greater decentralization and participation, differentiated government interventions can be better adjusted to the demands of local populations, unless they are captured by local interests. According to the nature of local governance, public interventions will be more or less effective to reduce poverty and inequality. In particular, differentiated interventions that are mediated through the local representation of communities (e.g., the Colombian Fondos Municipales de Cofinanciación and the Peruvian FONCODE) are effective to reach poor communities and to adapt the use of public funds to their needs, but they may not be effective to reach the poorest within these communities since it is the community itself that decides on the intra-community allocation of resources.
There are two approaches to seeking differentiated effects through policy interventions:
- One is through single or indiscriminated policies which have differentiated effects. In this case, differentiated effects can be adjusted through the choice of policy instruments in terms of their expected differential effects. For instance, a real devaluation of the nominal exchange rate will raise the price of food, benefiting net-sellers, leaving unaffected self-sufficient households, and hurting net-buyers. Policy can also be made by crops which, after allowing for substitution effects in production and consumption, will differentially benefit those households producing or consuming that crop. Similarly, policy can be made by region which, after allowing for secondary trade, will differentially affect those living in that region.
- The other is through differentiated policies or through differentiated public programs and interventions. In general, a differentiated policy will consist less in the direct use of a differentiated policy instrument (e.g., a price policy targeting a crop or a region) than in the delineation of a strategy of differentiated programmatic interventions (e.g., a policy defining a set of interventions to differentially benefit indigenous populations). The key issue for differentiated interventions is the ability to exclude so that the intervention be effectively differential.
Excludability thus opens the subject of targeting: How to identify the beneficiaries? How to exclude others and avoid capture by non-targeted users? How to establish the optimum leakage from a cost/benefit standpoint given the fact that there are increasing costs from precision in targeting? How to manage the political economy of targeting? How to establish optimum leakages to achieve political feasibility?5
5 See the discussion in Lipton and Ravallion, 1995.
How to design and implement differentiated interventions?
Differentiated interventions are not new, but recent progress in decentralization, democratization, and increased participation of civil society through representative grassroots and corporatist organizations give new potential for differentiated interventions to play more effectively in favor of the rural poor. For the allocation of public resources to the delivery of local public goods and services, decentralized co-financing funds have been introduced. This raises several issues:
Decentralization of the allocation of public funds to the community level through Fondos de Cofinanciación is a powerful instrument for the co-production of public goods between state and civil society that allows to differentiate the production of public goods in response to community demand and to mobilize local resource for co-financing. It has been used successfully in Colombia (DRI), Peru (FONCODE), Mexico (Pronasol), Bolivia (Ley de Participación Popular), and projects funded by IFAD (Community Development Funds) and the World Bank (Social Action Funds). In following this approach, the unit of attention in government programs is shifted from the individual to the rural community. This raises two important issues:
- How to maintain coherence with policy objectives when control over funds is delegated to the community?
- How to reach the poor when the community gains control over the intra-community allocation of resources?
Consistency with government policy can be partially obtained by offering to the community different rates of co-financing according to types of programs, where rates increase with government priorities. In Colombia, the percentages of co-financing vary by types of projects, and these rates also vary with the Index of Municipal Development that characterizes every particular municipio. In the end, municipios receive three types of transfers: fiscal funds from the center, funds for the co-financing of projects, and subsidies for specific purposes. Yet, it is the case that projects tend to be very dispersed with little overall strategic vision, even though they must be part of a municipal plan and need to be approved by the Consejo Municipal de Desarrollo Rural and by the governor of the state. In addition, resources tend to concentrate toward the municipios with more managerial ability, which tends to be the already better-off municipios. The result is a process of acceleration in the differentiation of municipios regarding the quality of public goods and services. Counteracting this process, the availability of discretionary resources at the community level induces communities to formulate projects and hence to increase demand for co-financing. While, in a first stage, better off communities are more effective in submitting projects for co-financing, in a second stage, learning to organize, cooperate, and formulate projects should allow other communities to become more effective in competing for funds over time. This ability to learn can importantly be assisted by NGOs.
Targeting within the community is in principle impossible when the allocation of funds is made the responsibility of the community. As a consequence, while poorer communities, characterized by a low Index of Municipal Development (Colombia) or a high Municipal Marginality Index (Mexico) can be differentially targeted, whether the poor within these communities are reached or not depends on the local power structure of the communities. And this local power structure may be more regressive than desired by the central government. There are several mechanisms that can be called upon to improve the likelihood that the poor will be reached:
- NGOs can be used to help the poor organize and formulate projects that can be submitted for funding. Important is to assist the poor to gain due representation in the Consejos Municipales de Desarrollo Rural or other decentralized entities that allocate funds competitively.
- Conditions can be placed on communities to qualify for funds under more attractive conditions according to their ability of reaching the poor. For this purpose, close monitoring of poverty needs to be put into place so that progress in some indicators can be used for these conditionalities. Monitoring should be organized as a joint process between central state and communities. All too often, monitoring is largely absent or ineffective.
- Decentralized interventions can be combined with other dimensions of social policy that eventually seek to compensate for what decentralized allocations do not achieve.
Market liberalization has meant the end of blunt, economy-wide poverty programs such as staple food crop price subsidies (both higher sale prices and reduced purchase prices compared to FMFT prices), subsidized credit, and subsidized inputs, especially fertilizer and water. These price policies were deemed failures because they distorted incentives leading to an inefficient allocation of resources, encouraged rent seeking behavior, and had high degrees of "leakage" of benefits to the non-poor. Targeted food subsidies and social funds attempt to minimize economic distortions and government fiscal burdens, while maximizing the beneficial impact of the program on the selected group.
Targeting. Increased attention has been placed on targeting as a means of improving the cost-effectiveness of programs, and reducing economic distortions (Gill, Indermit, Jimenez, and Shalizi, 1990; Grosh, 1994; Haddad and Kanbur, 1991; Ravallion, 1993). Targeting can reduce the fiscal costs of a program by directing resources towards a selected group. However, there are costs associated with administering a targeted program that increase with precision in targeting (Besley and Kanbur, 1990). Economic distortions and inefficiencies result from most types of poverty programs, including targeted programs. There will be price distortions in the markets associated with the goods being provided by the program (e.g., food, health care, jobs), there will be income effects that distort labor-leisure allocations, and there will be incentives to relocate to areas providing the programs. Targeted programs, because they are more concentrated in scope, can impose a reduced distortionary impact on the economy. In implementing targeting, policy makers must determine what type of targeting method (individual, group, regional, or self-targeting) is most efficient in terms of costs, and in terms of the optimal tradeoff between minimizing overcoverage (incorrectly classifying a non-poor person as poor) and undercoverage (incorrectly classifying a poor person as non-poor).
Social funds. Social funds are institutions set up to fund small scale development projects. The projects usually include construction or rehabilitation of infrastructure using labor intensive methods. Projects include construction of schools, clinics, roads, water, and sanitation facilities. The idea is to create employment in the process of construction as well as benefits from the finished project (Grosh, 1995). Social funds offer advantages over traditional projects: 1) they are demand driven and hence responsive to the needs of the beneficiaries, 2) they operate to a large extent outside the realm of public bureaucracy, and 3) they employ low income community members.
An important political problem in the management of social funds is whether to jump over intermediate levels of government to reach directly the institutions and individuals engaged in the management of the social fund. In many situations, this is a tempting option, particularly for the short run. Intermediate levels of government may be non-representative and seriously under-equipped to fulfill their new functions. From a long term perspective, however, jumping over the intermediate levels of government is not desirable. Alienating these levels of government is not favorable for the integrity and the continuity of the program. And, for the long term purpose of democratic construction, involving these levels of government is fundamental. Thus, short run expediency (unless the program is clearly seen as a one-shot relief operation) needs to be clearly weighted against broader long term gains in deciding to involve or not these levels of governance.
Transitory and chronic poverty. In the context of the debt crisis and implementation of stabilization and adjustment policies, attention has been given to transitory poverty in order to manage the political feasibility of the reforms. The main instrument for this purpose has been the use of social funds and implementation of safety nets. If economic shocks are recurrent, as in agriculture where incomes are exposed to climatic fluctuations in addition to policy shocks, reducing transitory poverty requires providing the rural poor with access to risk coping instruments so they can reduce the need to engage in costly risk management (Alderman and Paxon). Rural insurance schemes are, however, hopelessly exposed to moral hazard problems due to the difficulty of verifying insurance claims. These moral hazards can be effectively controlled through mutual insurance, but the ability of the community to effectively insure is limited by high local covariation of incomes. Again, the solution to this dilemma may be in the integration of local/traditional institutions with regional/modem institutions, thus combining the advantages of information and portfolio diversification for successful insurance.
In recent years, with the restoration of growth high on the policy agenda, chronic poverty has been badly neglected, leading to eventually explosive responses to the frustrations of marginalization and social exclusion in the context of globalization (viz. the Zapatistas and the EPR in Mexico, rural violence in Colombia, and guerrilla movements in Peru). For Latin America, and particularly for agricultural and rural policy makers, devising schemes to reduce chronic poverty is probably the greatest challenge ahead (Gore, 1995; The Economist, 1996). This implies increasing the productivity of the poor. Doing this requires achieving four policy objectives:
(i) Increasing the poor's access to assets: assets are multidimensional including most particularly agricultural capital (land and water), microenterprise capital, human capital (education, health), social and organizational capital (access to credit, membership to organizations), political capital (access to the state and to public goods and services), and migration capital (kinship and community networks of migrants).
(ii) Increasing productivity in the use of these assets: this is the objective of rural development programs. Technological change and the diffusion of innovations are the key instruments to achieve this goal.
(iii) Improving the context that gives value to the use of these assets: favorable prices for the products and factors sold and a favorable investment climate.
(iv) Improving the linkages between the poor and markets and public services by reducing transactions costs to enhance the value (effective prices paid and received, relaxation of constraints in access) derived from productive use of the assets they control.
Decentralization programs that transfer responsibilities from the central government to the sub-national government have been enacted in many countries of Latin America, including Bolivia, Brazil, Colombia, Ecuador, Guatemala, Mexico, and Venezuela (Ruflán, 1993). Decentralization can take a variety forms (administrative, political, economic) and can be carried out to varying degrees: deconcentration, which occurs within a given level of government; delegation, when a higher level of government passes some authority along to a lower level, but maintains the ultimate decision making power; and devolution, when full responsibility to determine and execute policies and projects is transferred to lower levels of government. These programs are motivated by the quest for increasing the efficiency and effectiveness of the public sector.
In terms of criteria for what should be decentralized, functions that can decrease cost, increase quality, or increase participation should all be candidates for decentralization. Specifically regarding rural development, administrative and political decentralization (preferably devolution) can lead to a number of benefits related to participation, responsiveness, efficiency and effectiveness of policies, and ultimately greater local economic growth and poverty reduction (Chiriboga, 1994). For example, municipal governments are likely to be in a better position to identify local needs and demands, and can thus design a more appropriate set of rural development policies than central governments. In terms of incentives, the more local actors are involved, and the more they feel that the money and the projects are their own, the more they are likely to care about the efficiency with which the money is spent, and, perhaps, the less likely they are to evade taxes. Politically, if decision making authority comes from fair and open elections at a local level, then decentralization can contribute to overcoming problems of legitimacy that many authoritarian governments have confronted. And, economically, the closeness of actors (government, business, GROs, universities) can contribute to an interactive process, leading to more creative solutions and greater local economic dynamism. There are, however, risks associated with decentralization such as local control by powerful minorities and reproduction at the local level of urban biases. Nevertheless, when all of these factors are combined, a strong argument can be made for the potential superiority of local government in the design and management of rural development projects.
The extent to which decentralization can improve efficiency will depend to a large degree on the local public choice mechanisms. Systems that have a democratic component - where voters decide specific tax-expenditure measures or indirectly elect government officials - provide a more efficient mechanism for the population to express its preferences and achieve accountability. Accountability, in turn, depends on the availability of public information such as financial statements, and evaluations of cost and performance of public programs (Winkler, 1994). Increasing efficiency through increased accountability requires that local governments have real authority. In particular, local governments must have the authority to raise local revenues and set expenditure levels to meet the needs of the local populace. They must also have budgetary discretion in regards to personnel, contracting, and determining local expenditures. The expectations and obligations of local governments must be clearly specified and their jurisdictions clearly delineated.
Decentralization involves, to various degrees, both administrative decentralization-increased regional authority over policy making -, and fiscal decentralization - increased regional autonomy in terms of tax and spending decisions. The balance between, and extent of administrative and fiscal decentralization depend on tradeoffs between administrative and allocative efficiency, and economy-wide redistributive efficiency (Musgrave, 1991). While increased administrative decentralization can increase the responsiveness of the state to local needs, fiscal autonomy can inhibit redistributive transfers from more to less prosperous regions. That is, poor regions that lag behind more prosperous ones will stand to lose federal financial resources for development with increased fiscal decentralization. Inequitable interjurisdictional differences may also have efficiency costs. If similar individuals receive different services depending on which jurisdiction they fall, then there is an incentive for individuals to relocate resulting in possible economic inefficiencies.
Potential efficiency and equity costs to decentralization provide a rationale for a central government's role in equalizing the fiscal capacity of jurisdictions. Brazil has both a state revenue sharing fund and a municipal revenue sharing fund which partially equalize fiscal capacity. Chile's municipal common fund redistributes property tax revenue among municipalities (Winkler, 1994). Disparities are particularly important in the provision of services, such as education, health, or nutrition, where the government wishes to maintain some minimal standard of support. Equity can be maintained in these areas via program specific intergovernmental transfers.
In practice, there are a number of obstacles that inhibit success and point to decentralization as a gradual process that requires the continued involvement of the central government. In a review of rural development decentralization experiences in Latin America, Chiriboga (1994) suggests that two of the most serious constraints have been the limited financial, technical, and human resources of local governments, and the traditional forms of exercising local political power, based on clientelism, personal relations, and elite domination. Both of these issues point to the need for modernizing municipal government and for a crucial role for central government in adequately training local administrators. There is thus a need for identifying the appropriate complementary roles between different levels of government. In addition, devolution without a corresponding transfer of revenues has often led to deterioration in the quality and efficiency of services (an indication of decentralization as part of state contraction and government failure, rather than as part of a redefined functional role).
Privatization of parastatals and descaling of the direct role of the state in the economy have removed many of the public institutions that served agriculture. In the place of state institutions, civil institutions are beginning to emerge. The transition away from public and towards private institutions has impacted different segments of the population disproportionately. In general, larger-scale commercial agriculture and producers of export crops have more successfully adapted to the use of private institutions than has the smaller-scale, food-crops oriented traditional sector. The commercial sector has been able to utilize civil institutions such as commercial banks, agro-industrial contractors, private merchants, producers' organizations, and private technical assistance providers. In contrast, the small to medium scale producers, and producers in less favored regions, have been less able to gain access to or create private sector replacements for the services that had been provided by public rural development programs, leaving for them serious institutional gaps.
Civil institutions that serve the smallholder economy have been slower to emerge, placing smallholders at a competitive disadvantage during the transition period. These institutions are key to enabling smallholders to reduce transactions costs in production, finance, input acquisition, and marketing. Institutions that can lower transactions costs for smallholders may be slow to emerge naturally because they require more organization and cooperation. Many local, often traditional, institutions had been suppressed by preponderance of the state, particularly where the state had been more forceful in managing agriculture such as in Mexico and Brazil. There is now a role for the state to encourage the creation of these institutions during the short run and to facilitate their functioning during the long term.
Institutions that can successfully substitute for the former government services should capitalize on the unique information and enforcement advantages of local institutions. These institutions can use local information and social sanctions to control for adverse selection and moral hazards in contractual relations. They can lower enforcement and monitoring costs and enforce cooperative solutions by employing interlinked transactions, social pressure, reputation, and repeated games with no exit option. Civil institutions that have been able to harness local information include group lending schemes, credit unions, savings and loan associations, financial NGO's, marketing cooperatives, and community storage organizations. Other local institutions include community based organizations that interface with government social funds for the construction and maintenance of public infrastructure, and common property resource management, including pastures, forestry, and water resources. The revitalization of local institutions has thus an important efficiency purpose. Revitalization goes through the revalorization of local culture, the consolidation of community social relations, and preservation of the concepts of rurality and place. Linkages between local, regional, and national institutions is fundamental to allow for diversification of risks and access to broader markets. In a sense, this linkage between the local and the global should be an important dimension of the response to globalization since it gives to local institutions a fundamental role in the competitiveness of regional economies, and hence also an economic logic for their preservation and consolidation.
The differential impact of policy reforms can have important long run equity and efficiency implications in the context of land market liberalization and agrarian reform. As land markets become liberalized, land prices will eventually reflect the present value of the stream of profits generated by the land. Over time, land will become concentrated in the hands of producers earning the highest profits (Carter and Mesbah, 1993). We argued above that the institutional gaps in the transition period have caused small and medium sized landholders to be at a competitive disadvantage to commercial farmers. Hence, with liberalized land markets and undeveloped smallholder institutions, the forces are in place for landholdings to become concentrated in the hands of the larger commercial classes, even if they are not the most efficient producers. An alternative scenario, that delays the liberalization of the land market and encourages the emergence of smallholder institutions would be superior both in terms of equity and efficiency. A sequencing of reforms that aims at preserving a thriving smallholder class would thus proceed first with reconstruction of the agrarian institutions supportive of the competitiveness of smallholders before engaging in a wholesale liberalization of the land market.