May 2001
Luis Llambí
Tomás Lindemann
Part 1 of 3
Also available in Spanish
This paper is about strengthening the institutions that ensure good governance. Strengthening the institutional capacities of local governments is important for: 1) economic reasons (e.g. productive and allocative efficiency), 2) equity reasons (territorial and social equity); and 3) political reasons (e.g. elected officers' accountability to citizens, citizens' participation in decision-making, and democratization of decision-making).
This paper compares the experience of two decades of state reforms in seven Latin American countries (Chile, Colombia, Bolivia, Brazil, Mexico, Peru and Venezuela), in order to analyze the factors that might explain what succeeded and what went wrong; and to draw some lessons for the future.
In all these countries - starting with Chile in 1973 - various types of state reforms have been implemented, sometimes as a component of macroeconomic stabilization programmes, but also independent from them.
State reforms are historical processes, which involve political, legal, administrative and territorial changes. As a result of all these processes, by the end of the 20th century, the highly centralized forms of the nation-state have yielded some of its functions to: a) supranational governance bodies (e.g. MERCOSUR, NAFTA); b) to subnational agencies; c) nongovernmental organizations; d) grassroots organizations, and e) private enterprises.
The transfer of some decision-making powers and responsibilities from central government agencies to subnational (e.g. regional and/or lower) governments has been an important component of some of these reforms in the cases of Bolivia, Colombia, Brazil and Venezuela. This "territorial decentralization" of power in some cases has been preceded by political reforms: such as the popular election of local mayors and/or regional governors. In a few cases, the transfer of decision-making has gone from a central government agency to a 'civil society' or 'grassroots organization'. In a large number of cases transfers have favoured private enterprises, either via the selling out of state enterprises and assets to private firms or via contracting out for services without any shift in public property assets. In most cases, however, these reforms have been accompanied by the dismantling or downsizing of some central government agencies, particularly from the agricultural public sector.
Yet, although there has been not serious attempts at evaluating the differentiated impacts that the reforms have had on the extreme poor, the poor and the middle classes, the Latin American experience also shows that few reforms have reached all their alleged goals (e.g. increased market competitiveness, poverty reduction, environmental sustainability, sound governability), and some have even resulted in serious short circuits and backlashes.
From a normative perspective there are three points that the paper stresses:
However, the capacity of local governments, particularly in rural areas, tends to be limited. Most local governments do not have access to appropriate administrative skills. They are not familiar with the national laws and administrative procedures of the central government. They lack the necessary knowledge to negotiate with other public agencies and with private firms. Their know-how for fund-raising is limited. Their skills for designing, monitoring and evaluating short-term investment projects is also inadequate. Moreover, they have scant information about their political economic environment, and therefore to develop long-term economic local development strategies. For all these reasons, institutional building at the local government level should be a major component of any decentralization programme.
This report has been written by Luis Llambí and Tomás Lindemann, following the Terms of Reference of Luis Llambí as Visiting Professor in FAO's Academic Partnership Programme. The goal was: 1) to develop a conceptual framework aimed at evaluating the institutional reform processes in the agricultural public sector, focused on the devolution of functions from the Ministries of Agriculture to local levels; and 2) to provide a comparative critical review of the empirical literature on institutional reforms in Latin America.
We would like to acknowledge the large amount of publications already produced by FAO, on which this current report is based. 'The Proceedings of the Technical Consultation on Decentralization for Rural Development' (Rome, November 1997), under the leadership of Gustavo Gordillo de Anda, Santiago Funes and Jennie Dey-Abbas, were extremely helpful. It was also helpful to review the "Decentralization and Rural Development" documents published by the Rural Development Department, the RED-IFO Decentralization Model prepared under the supervision of Mohamed Nadir, Tomás Lindeman and Jean Bonnal, and FAO's Online Sourcebook on Decentralization and Rural Development.
We also recognize the insights and empirical material provided by the documents prepared by the Grupo CEO of Argentina (Martín Piñeiro, Roberto Martínez Nogueira, Eduardo Trigo, Filemón Torres and Eduardo Manciana), under the auspices of the Interamerican Development Bank. In addition, we are grateful for the documents prepared by María Elena Cruz (Chile), Edelmira Pérez (Colombia), Jorge Romano (Brazil), and Luis Fernando Fernández (Costa Rica), for the Workshop on Decentralization held at Río in November 1999, sponsored by REDCAPA (the Latin American and Caribbean Network on Economics and Agricultural Policy Training).
During Luis Llambí's assignment he also had the opportunitiy to visit Bolivia as part of a FAO mission to assist the Bolivian Government on the preparation of a follow-up project of the Special Programme for Food Security. Visiting a country which led Latin American efforts on decentralization in the early 1990s, added new insights to this report.
We are most grateful for all the support and assistance given to us by different people in FAO's Rural Development Division, and especially to Jean Bonnal and Norman Messer. We, however, assume full responsibility for any errors or omissions in the report.
"Decentralization must be fitted into the broader process of state reform (…) It is essential to bear in mind the broader historical process of the formation and construction of nation states." (TCD. Proceedings. 1987:69).
"Decentralization is such a broad interdisciplinary problem, so it is not surprising that much of the work done on decentralization by social scientists (…) is weak in terms of theoretical categories. A lack of theoretical grounding makes both the positive analysis of decentralization (how to derive lessons from experiences, particularly via monitoring and evaluation) and the normative design of why and how to decentralize, unreliable. For both of these purposes, a solid grounding in analytical categories is needed. This should start with an inventory of the theoretical constructs that identify why there are benefits and costs to decentralization… To be scientifically valid, every effort must be made to ground this research in rigorous theoretical categories" (A. de Janvry, TCD Proceedings 1987: 72-73).
The case for decentralization is often based mainly on political and equity considerations. Yet, the political objectives to increase accountability, citizens' participation, and democratization may coincide with the economic objectives of better allocative decisions of public resources and increased willingness to pay for local services.
The two main forces are currently determining the decentralization of nation-states, together with the emergence of new governance supranational institutions. On the one hand, globalization, which is not only related to the emergence of highly competitive global markets and rapid technical change, but also by the search for new rules of the game. On the other hand, an increasing grassroots' struggle for democratization.
State reforms are historical processes. They involve political, legal, administrative and territorial changes. As a result of all these processes, by the end of the 20th century, the highly centralized forms of the nation-state have yielded some of its functions, and part of its authority and power. Some were yielded to multinational and supranational institutions, while others were transferred to local governments, non-governmental organizations (NGOs) and to private enterprises. However, to our knowledge no in-depth evaluation has yet been undertaken to assess the pros and cons of decentralization.
At the national level, the general directionality of the process is captured by the term decentralization. Being historical processes, these reforms are also subject to political backlash, short-circuits and mixed results.
One of the problems is that decentralization has become such a slippery concept in the development literature. According to one of its critics:
"It is empty enough that one can fill it with alsmost anything. A concept rendered meaningless if we are not carfeful in its use. Many reforms gain the label 'decentralization' when it is not entirely clear that the term is appropriate" (Gershberg 1998:407)
The paper's goals are:
Rondinelli's typology of decentralization is usually the conceptual framework for most documents on the subject. He distinguishes three possible outcomes of administrative state reform: a) the transfer of CG functions to parastatal agencies, which he called "delegation"; b) to local governmental agencies, which he calls "deconcentration"; and c) to semi-autonomous levels of government, which he calls "devolution" (cfr. D. Rondinelli 1981 "Government Decentralization in Comparative Perspective: Theory and Practice in Developing Countries", International Review of Administrative Science 47, as quoted by Manor 1997:4). There is an underlying value-ladden assumption in this conceptual framework: the perception that there is some kind of progression when the process goes from one category to the next, with being devolution the preferred outcome.
As with any other taxonomy or typology, Rondinelli's three conceptual boxes have their merits. The typology clearly defines three possible outcomes of state reform processes. One problem, though, as with other ideal-type categorizations, is where to draw the line when you are comparing different historical cases of reform. In fact, most state reform processes usually exhibit accomodating elements of all of these different categories. Besides, the assumption of a uniform, unilinear, ahistorical process of state reform, is misleading.
However, its main demerit seems that it hides the basic distinction between the political and the administrative aspects of state reforms. Therefore, when one tries to apply these distinctions to any particular country, the analysis becomes blurred. As with all ideal-typical definitions, the risk is that, with time, they become void of content and become instead a strait jacket for both the analyst and the policy-maker.
The term "decentralization" has also become a slippery concept both in real life and in the literature. The term is given various and variegated contents when used by different government officials, academic analysts, and other development practicioners. According to its different users, a myriad of quite different reforms have been labelled decentralization. Thus, for some analysts:
"(the term) is empty enough that one can fill it with almost anything (…) A concept rendered meaningless if we are not careful in its use. Many reforms gain the label "decentralization" when it is not entirely clear that the term is appropriate." (Gershberg 1998:405-407)
To avoid such misunderstandings, the paper proposes: firstly, to provide a brief though precise definition of the nation-state and its historical forms. Second, to distinguish between the political and the administrative aspects of the state, and therefore between the political and administrative dimensions of state reforms.
The emergence of the nation-state was a historical product of political centralization, worldwide. It was related to longstanding nation-state formation processes, as well as to the construction of nationhood. Centralized nation states in Western Europe were thus an historical product. But also elsewhere. The three centuries from the Westfalia Treaty (1648) to the Bretton Woods agreement in 1945, gave rise to different forms of centralized nation-states not only in Europe, but also to the gradual extension of the centralized nation-state model to the rest of the world through colonization, first; and, then, via independence (Held 1995).
Governance structures, in general, and historical nation-states, in particular, are always the result of political as well as economic/administrative processes. All theories of the state recognize this double-edged nature of the state.
Current nation-state restructurings are also the result of a worldwide process. The demise of the Bretton Woods agreement in 1973, the fall of the Berlin Wall in 1989, and the beginning of the Uruguay Round of trade negotiations in 1986, determined the search for new political and economic institutions worldwide. As a result of all these changes, by the end of the 20th century, the highly centralized forms of the Latin American nation-states have yielded some of their functions to: a) supranational governance bodies (eg. WTO, NAFTA, MERCOSUR, etc.); b) subnational governments (e.g. provincial or state governments and municipalities in Latin America); and c) different sorts of non-governmental organizations, grassroots' associations, and private enterprises. Parallel to these processes, there is also a trans-nationalization process of governance functions related to the trans-nationalization of capital worldwide.
Yet, within each nation-state, the reforms are the result of unending processes of negotiation between organized stakeholders at all levels, more than of technical considerations.
1.2. The three-stage unilineal evolutionary framework
Closely following the relative success of the Chilean reform process, a three stage model for interpreting the reform of the state in Latin America has also been proposed by some analysts (Martínez Nogueira 2000; Piñero et al 1998). The first stage, macroeconomic adjustment, is basically about the devolution to market mechanisms of the allocative functions which were assumed by the nation-state during the whole inward-oriented growth strategy. The second stage, usually referred to as structural reform, is basically about creating the new institutional architecture of the state in tune with the market-led growth strategy that characterized the first stage. Whereas allegedly the first stage was mainly based on market deregulation, economic opening, and privatisation; the second was based on institutional engineering. The missing links, however, and this is why a third stage is called forth, are about the most difficult tasks of: on the one hand, constructing the social and political pact for pursuing reforms; and, on the other hand, consolidating the necessary cultural and value changes which should accompany reforms efforts to be successful. In addition, although this paper is not geared to deal with this problem, the third stage of reforms should also look at the consolidation of supranational structures.
Again, as our case studies will hopefully illustrate, the history of Latin American state reforms is much more variegated and complicated than that. Not only have the reforms not always followed the same stages, but their sequences have seldom followed very different paths. Even in the case of Chile, which for some authors has become paradigmatic, neither the number of stages nor their content and sequence have followed the schematic path proposed by these interpreters. Therefore, there is no other option than to return to analysing in detail the different paths followed by each reform process, and only then make an attempt to understand the threads underlying each case to finally establish if there are some commonalities, or if each case followed its own particular logic according to each country's specific past history and current historical conditions.
1.3. From centralized to decentralized nation-states
As a result of democratization processes, in some Latin American countries autocratic centralized regimes have been replaced by elected governments. The election of mayors and municipal councils through uniform suffrage, together with the emergence of mechanisms of popular participation (e.g. town meetings or "cabildos", referendum, etc.) are good examples of this. On the other hand, democratization has also created the enabling environment for a proliferation of grassroots organizations and non-governmental organizations, as well as their vertical and horizontal integration into second-and third-tier level organizations that impact policy decision-making processes.
The different layers of government. The central government: sometimes referred to as the "national" government in unitary states, or as the "federal" government in federal states (e.g. Brazil, Mexico). The subnational "regional" level, referred to as "state" governments in the Federal States, or as "regions", "provinces", "departments" in the unitary states (e.g. Colombia). The local level: usually referred to as municipal governments but also "districts".
Chile was the pioneer of state and economic reforms in Latin America. The military coup of 1973 ended four decades of state-led inward-oriented import-substitution industrialization as well as the socialist experiment of the democratically elected government in 1970. From 1973 to 1982 a market-led structural adjustment programme with minimum state intervention in the economy was gradually put in place. In 1980, the principles guiding the new roles of the State within the market-led growth model were formalized in a Constitutional Law: the national government would play a "subsidiary" or "complementary" role in economic activities, leaving market forces as free as possible, and extending property rights in order to promote private investments and initiatives. The alleged economic goals of the reforms were to achieve an equilibrium in the balance of payment, stabilize domestic prices, and promote microeconomic efficiency. The means were: fiscal austerity, opening the economy to the global markets, privatization of most public enterprises, domestic market liberalization, financial sector deregulation, and an attempt to strengthen the land market via a reversal of the former land reform law and the establishment of clear property rights for private enterprises. The nation-state would retain its normative and regulatory capacities, transferring some functions to subordinate levels of government, and opening new spaces to private sector enterprises in the provision of public goods.
For the agricultural and rural sector, the 1973-1982 period was also the end of the agrarian reform process begun in 1962, the retrenchment of the agricultural extension systems, and the end of most public investments in the productive sector. Agricultural growth was subordinated to the goals of macroeconomic domestic stability and economic opening to global market forces, without any autonomous agricultural policy. The gradual eviction of the functions previously performed by public agencies determined both the strengthening of the private sector and the gradual dissolution of the peasantry which emerged during the agrarian reform process. By contrast, the emergence of a strong agribusiness and agroindustrial sector filled some of the spaces previously occupied by the agricultural public sector.
In 1982, the social and economic impacts of the debt crisis on this widely open market-led economy forced a revision of some of the measures previously adopted. The agricultural sector was one of the most affected by economic recession, impairing its competitiveness in the newly emerging global market economy. To avoid social disruptions, some protecting measures had to be introduced into the model: import tariffs were again raised, controls to interest rates were restablished, the exchange rate was again manipulated, subsidies to "vulnerable" farm products were restored, and new state interventions such as price bands to vulnerable agricultural products were put in place.
This first stage of state reforms were not based on any rigorous conceptual statement of organizational and institutional change. In fact, most of its theoretical underpinnings were only developed a posteriori. It was based on the assumption that public agencies were able to mould according to political decision-making; that freed of most public interventions competitive market forces would allocate resources transparently; and that private enterprises would adapt rapidly and efficiently.
1990-1994. The second generation of reforms. At the beginning of the 1990s the agricultural public sector was extremely weak. A reduced budget and a limited number of policy instruments, only those deemed to ease the external risks or to maintain some sanitary controls, to manage forestry and irrigation subsidies. A technical transfer programme which was operated through private enterprises of technical assistance. Therefore, in 1990 the goal of the new democratic government was to strengthen and modernize the sector's public institutions and to promote equity considerations. In addition, the opening to new commercial agreements in order to strengthen the position of the country in the global markets, but the basic traits of the market-friendly approach to economic development remained almost untouched. A mayor emphasis was put on public spending to implement social policies. From 1992-1996, social spending more than 65% of public spending, which compares favourably to the 55% of the 1985-1989 period.
1995-1999. The start of the third generation of state reforms in the agricultural public sector. The priorities of public spending: education, health, housing, and focused social interventions (safety nets) to help the most vulnerable social and productive sectors. The agrarian policy had been restructured: irrigation, technical modernization, natural resources, access to credit. A dualist view of the sector. The divorce between agricultural and rural development. The distinction between the promotion of agricultural competitiveness, and rural poverty alleviation programmes. The promotion of agricultural competitiveness aims at a sustainable role for agricultural production within the scope of an open economy. Whereas the social policies have the objective of ameliorating the quality of life, with special emphasis on rural poverty. The former was a role attributed to the Ministry of Agriculture. The latter was divided between different ministries: planning and social solidarity, education, health, housing.
The social segmentation of agricultural institutions and policy. The distinction between so-called entrepreneurial agriculture and the "peasant" sector, and within this latter sector between those economically "viable" and "unviable".
Therefore, once the first stage of reforms has been achieved, the goal is to generate the institutional reforms that are needed to make the operation of markets more efficient. The second stage is not concluded. The need for a third reform: the internalization of the new rules to the managerial style of public and private agents.
Decentralization has never been a priority on the agenda of state reforms, if we consider decentralization to be something different from the various forms of privatization which have taken place in the relations between the public and private sector. Three forms of privatization of former public sector agencies have been attempted: a) selling out state enterprises and transferring assets to private firms: b) contracting out to private enterprises for services formerly provided by public agencies; and c) the administration of public agencies according to competitive market principles.
Contracting out has been one of the preferred forms of privatization, particularly in the provision of technical assistance and agricultural extension services to small- and medium-scale farmers. The CG has retained a public function, but has: a) mobilized groups of farmers to draw up specific demand-driven extension agendas; b) financed private sector extension service providers, rather than those directly providing those services; c) contracting out to private professionals and requiring clients' participation as part of contractual arrangements (cfr. TCD:176). The basic principle underlying several programmes was to develop a market for agricultural extension services, by promoting the creation of small service enterprises among the ranks of former civil servants.
The results of these programmes have not been encouraging. Most of the services provided to the farmers have been considered of low quality, and therefore the expected productivity gains never showed up. One of the reasons was the lack of an adequate supervision and monitoring functions by the CG's agencies which were supposed to do so. The other reason was that the farmers never were willing or able to provide the increasing funds to finance these services, as was programmed by the co-financing mechanism under which the programme was based.
Territorial decentralization, in particular, has not been very high on the reform agenda. The country is divided into thirteeen regions, each one headed by an Intendente who is nominated by the CG. The municipal authorities are elected. Some state functions, particularly urban public services, have been transferred to municipalities, whereas others such as health, education, the protection of the natural environment, etc. are shared with other levels of government. Yet, local authorities have no hierarchical authority over any of the national institutions operating within the municipality. Different co-financing mechanisms have been attempted with the aim to provide transparency to the allocation of public funds and to develop competition between public agents in pursuing investments and the access to public services.
Chile's case cannot be considered as a blueprint for state and economic reforms. As in any other case, both its sucesses and failures have to be explained as a rather specific historical experience.
Since the 1891 constitution, Brazil has been formally defined as a Federal State government, comprising 26 (subnational) states, and about five thousand municipalities. From the 1950s to the late 1980s, however, based on the country's relatively huge domestic market, a rather centralist and interventionist state was put in place.
In 1967, during the military regime (which lasted from 1964 to 1986), a first administrative reform was decided. Its goal was to build a new institutionality based on semi-autonomous public agencies that were variously linked to Federal government's Ministries, parastatal foundations, public enterprises, and joint (private/public) ventures. Allegedly this was an attempt to contain the growth of the civil service by introducing labour market flexibility into the public service. For instance, the Federal government created "State Commissions of Agricultural Planning", together with "Special Programmes" for less-developed states. In general, these (subnational) state agencies enjoyed considerable autonomy, mainly due to the lack of capacity manifested by the Federal government agencies. Yet, despite these administrative measures, the fiscal crisis of the interventionist state did not recede. In fact, rent-seeking was even increased due to the clientelist practices of the military regime.
The 1986-1990 period was a time of democratic transition. Economic growth with increased income distribution were again top in the public agenda, but this again supposed raising real wages and increasing public spending. In 1998, a Constitutional reform called for the return to a "highly centralized public administration", in order to avoid the uncontrolled public spending associated with the huge parastatal administration created during the years of military rule. The 1989 Constitutional reform assigned new functions to the (subnational) state governments. Yet, given the budgetary constraint, subnational governments were given scant resources to carry out their new responsibilities.
During the administration of President Collor de Mello (1990-1992), the Federal government installed an ambitious economic stabilization programme. This was a minimalist and gradual attempt to reduce the public deficit through exchange rate manipulations, and a state restructuring programme to shrink all levels of government. In 1992, an agreement was signed with the IMF, which opened the way later for an external debt restructuring agreement to be signed with foreign banks. In July 1994, a new currency, the real was put in circulation to fight inflation. Unlike the cruzado, the former currency, the real was not subject to rapid devaluation. Since its introduction, the inflation rate lowered considerably. The National Congress was given the role of controlling the public budget, not only at the Federal government level, but also at the (subnational) state and municipal levels.
After 1988, there was a gradual opening of the economy, and the creation of the Southern Cone Common Market (MERCOSUR), and the establishment of a Common External Tariff between Brazil, Argentina, Paraguay and Uruguay. A strategy of increasing the bargaining power of "southern countries" vis-à-vis the major trading blocks led by the US and the EU. So far, however, the attempts to foster the harmonization of the national policies of its current members, as well as its enlargement to include other Latin American countries, have not been highly successful.
In October 1994, Fernando Henrique Cardoso was elected president. The short-term goals of his administration were: macroeconomic adjustment through anchoring the exchange rate, tributary reform and a fiscal austerity programme. The ultimated alleged goals were to create a Social-Liberal state, backed up by a "strategic managerial" public administration.
Within the public agricultural sector, a particular arrangement exists within the Federal government and the states. Economic planning and agricultural programmes are still Federal (national) prerogatives, and authority flows downward. For most agricultural support services (e.g. training, extension, research, irrigation), policy decision-making is the exclusive domain of the Federal level, with the exception of the provision of inputs in which (subnational) states have decision-making authority.
Federalism and fiscal crisis. The persistent fiscal crisis of the state within the context of a political administration still combines clientelistic, corporativist, and universalist principles. One way out: restrictions to the indebtment of (subnational) states and municipalities. Another would be contracting out public services.
One interesting case study is provided in the literature by Judith Tendler's 1997 book on the Brazilian Rural Poverty Alleviation Projects in ten north-east states. According to her, these projects provided matching grants to rural community associations to finance small-scale sub-projects that they themselves identified and subsequently operated and maintained. Although the Councils were located at the municipal level and were chaired by the mayor, they were not formally part of the local government. The Councils promoted local-level consensus through open "town meetings" (cabildos), and reviewed sub-project proposals for subsequent approval and financing. Tendler also narrates the successful decentralization of health-care programmes in Ceará, Northeast Brazil. Yet, other decentralization efforts such as that of agricultural support services, which revealed the low administrative capacity of the state, were not so successful.
The need for structural reforms. Agrarian reform and the opening of the land market. This was pushed by the rise of grassroots' organizations which reappeared during the mid-1980s: the Landless Movement (Movimento Sem Terrra), the movement of extractivist farmers in the Amazonia (e.g. the Seringueiros), and the movement of migratory farm workers (cold lunch or boias frias) in the southern states, etc.
|
