SD: Institutions : Technical Consultation on Decentralization. Decentralization and Local Capacity: Some Thoughts on a Controversial Relationship

Updated December 1997

Food and Agriculture Organization of the United Nations United Nations Capital Development Fund International Fund for Agricultural Development German Agency for Technical Cooperation Swiss Agency for Development and Cooperation World Bank

Rome
16-18 December 1997
Technical Consultation on Decentralization
Documentation

Decentralization and Local Capacity: Some Thoughts on a Controversial Relationship

by Ariel Fiszbein
Economic Development Institute
World Bank

The views expressed in the following paper are those of the author and are not to be attributed to any of the TCD sponsoring organisations.

THIS NOTE seeks to present some basic premises and hypotheses for discussion regarding the relationship between policies of state decentralization and the development --or existence-- of the local capacity required to implement such policies. It is largely based on three research projects, two of which involved an analysis of the decentralization process in Colombia (Fiszbein [1997] and Bird and Fiszbein [1997]) and the third one an ongoing project on public-private partnerships in several countries in Latin America (Fiszbein and Lowden [in preparation]). It was prepared as a contribution to the discussions to be held during the Technical Consultation on Decentralization on the issue of local capacity and, in that sense, it is more an attempt to organize my thoughts than a research paper.
It is not unusual to start a paper dealing with the issue of state decentralization with a remark on the many different uses of the term (Rondinelli [1981]). For the purpose of this paper, it suffices to say that the points I will be arguing here are to a great extent valid both for cases of devolution and/or democratic decentralization as well as for cases of delegation. They are, I believe, less relevant for experiences that do not go beyond the deconcentration of state functions to local and regional offices of the center.

Countries initiate decentralization reforms for a variety of --many times overlapping-- reasons. In most cases these reforms have been a response to political challenges emerging from democratization and/or regional conflicts. In other cases efficiency and financial considerations played a role in the design of reforms (Bird and Vaillancourt [1997]).

While the specific form taken by these reforms (e.g. the design of inter-governmental transfers, the specific assignment of expenditure responsibilities or revenue authorities) is likely to be strongly associated with the nature of the underlying factors motivating decentralization, the question of whether local capacity exists or can be developed is likely to be an important one in any case. After all, decentralization means that certain functions previously performed by national bureaucracies will be performed by a given combination of public and private agents at the local level.

From the point of view of policy-makers at the center --and, one should add, international development agencies-- the expected benefits of decentralization are clearly dependent on whether local governments --and more broadly speaking local institutions (North [1990])-- are "up to the (new) job". The "new job" in question might be the delivery of a specific service, collecting a specific tax or a wider combination of typical state functions. But whether the local level has, or can develop quickly enough, the necessary capabilities will remain a critical question in the minds of most policy-makers.

An interesting illustration of the controversies associated with these questions is the parliamentary debate that took place in Colombia as a new law that would create untied fiscal transfers to local governments was being discussed in the early 1980s. The mainstream opinion in Congress was that no real benefit would be derived from transferring funds and responsibilities to local governments if their lack of capacity would not allow them to manage them effectively in order to improve the quantity and quality of services offered to the population. Interestingly enough, the proponents of the law --that would eventually be passed by Congress-- did not try to argue that such capacities indeed existed. Rather, their argument was that only if fiscal resources and responsibilities for service delivery were transferred to local governments would those capabilities develop, as it is only if and when faced with concrete challenges that local institutions would acquire them (Galan [1990]).

In fact, at least for some of the Colombian reformers, the creation of local capacity --understood as the consolidation of democratic state and civic institutions particularly in more than 800 rural municipios-- was an objective rather than a condition for decentralization. Almost a decade later, Bolivia would follow a similar path, and similar discussions can be found in post civil war debates in several countries in Central America.

Is lack of capacity a binding constraint for a successful process of decentralization? Does decentralization lead to stronger --more capable-- local institutions (as, for example, some of the Colombian reformers speculated)? What types of external interventions are most effective in promoting the development of local capacity? These are some of the controversial questions underlying the debates on decentralization reforms and, not surprisingly, the ones being presented to us today.

The literature on institutional development (for example, Brinkerhoff and Goldsmith [1992], Cohen [1995], Goldsmith [1993], Uphoff [1986]) fortunately provides us with very useful analytical and empirical references to approach these questions within the context of the international evidence that is being presented and discussed during this Technical Consultation. But, relative to other sub-topics within the decentralization literature (particularly dealing with fiscal issues), my humble interpretation is that this is an area in which conclusions that can be generalized are difficult to come by. I have thus organized the rest of my presentation in the form of "generalizations", with references to some of the empirical evidence coming from a few research projects I have been involved with over the last three years. Whether they will survive the scrutiny of this group remains to be seen but, I hope, they will at a minimum provide material for discussion and criticism.

Capacity and conflicting objectives

Local capacity is hardly an end in itself: it is rather a means to achieve a combination of objectives. That is, capacity (local or otherwise) can be understood as the tools that allow the pursuit of a goal. In that sense, one knows that there is a capacity deficiency when particular organizations or institutional arrangements are under-performing relative to what could be achieved given the available resources (Fiszbein [1997]). What this highlights is that capacity should be evaluated in terms of specific objectives. In other words, when faced with the question of whether local capacity constitutes a bottleneck to successful decentralization reforms, one should ask "capacity to do what?" before attempting to offer an answer.

An important implication is that what appears to some analysts and policy-makers as lack of capacity, might in fact be the reflection of a conflict in the objective function used, on the one hand, by those analysts/policy-makers and, on the other hand, by the local people. In other words, I say you have poor capacity because you are not doing what I would like you to do.

When Richard Bird and I (World Bank [1996] and Bird and Fiszbein [1997]) reviewed the state of decentralization in Colombia ten years after it had begun, we quickly found that what was being characterized as poor planning capacity among municipios was in fact a genuine disagreement between local and national priorities. At that point in time disbursement ratios under the system of matching funds managed by the central government were extremely low. In the view of the central government, this was due to low capacity on the part of local government required to prepare local development plans and project proposals. The reality was, however, that most local governments felt that the conditionality imposed in the use of those funds contradicted local choices on how to allocate resources and manage local services, and thus preferred borrowing funds at market rates to using their resources to match the national grant. Many of those local governments might have unusual or perverse preferences --at least from the national perspective-- but they sure had no lack of capacity to achieve their objectives.

This review also showed that in many areas, national legislation and policies severely limited local capacity to pursue not just local but also national goals. For example, even after a decade of decentralization, Colombia made extensive use --through national laws and regulations-- of earmarking of local resources for specific uses. In addition, over the years, the national government had established a variety of "unfunded mandates" on local governments such as the requirement that each municipality (70 percent of which had populations of less than 20,000) establish a toll-free telephone number to report cases of corruption. When fully accounted the combination of earmarking and unfunded mandates represented for many municipalities more than 100% of the untied portion of the automatic inter-governmental grant they were receiving (World Bank [1996]). Thus, the observation that few municipalities were complying with those mandates (e.g. no toll-free number) was more a reflection on the absurdity of the policy than on local capacity [1].

What this and many other examples in Colombia and elsewhere in the region illustrate is that what in many cases is perceived to be lack of local capacity could in fact be interpreted as (1) a conflict between local and national preferences; or (2) an inadequate design of incentives in a principal-agent relationship.

Scale as a barrier to capacity

It is sometimes assumed that there exists a close correlation between the size of the subnational entity and its capacity. The argument is that small units (local governments, communities) will never have the capacity required to perform certain functions. It is impossible to disagree with this statement at such high level of generality: no one in his or her right mind would suggest, for example, that a municipio of a few thousand in the Bolivian altiplano should be in charge of trade policy or attempt to run a graduate program in physics. The question becomes relevant when specific functions are considered: running schools and health clinics, operating irrigation systems, building and maintaining roads, taking initiatives for local economic development, etc.

From that perspective, the argument should be that certain dimensions of capacity require a certain scale before they can fully develop [2]. For example, when conducting field work in Colombia to analyze the development of local capacity to deliver services in three sectors (roads, water and education) we considered three dimensions of capacity: labor (quality of staff), capital (equipment, buildings) and technology (internal organization and management style). The evidence collected does not support the view that scale is an intrinsic barrier along these three dimension (see World Bank [1995]).

It's not that we didn't find areas in which scale presented a problem in terms of the municipio's capacity to deliver these services. Not surprisingly we found that most rural municipios had difficulties in hiring a cadre of professionals that could effectively deal with the technical challenges presented by the variety of areas in which they had service delivery responsibilities. But, interestingly, we found that many of them were coming up with "market" or "bottom up solutions". The most typical one was the constitution of associations of local governments for specific purposes. It is not unusual to see half a dozen small rural municipios sharing construction equipment and professional staff, for example.

Even so, we found that some very small and isolated municipios simply found it impossible to, for example, hire even a part-time supervisor in charge of the local schools that would satisfy national professional standards. But, what we also found is that, before decentralization, this community de facto did not have the benefit of that service either. Formally, the regional deconcentrated office of the Ministry of Education provided that service but in reality the supervisor hardly ever showed up. In fact, the main change since decentralization was that the sole white colonist leaving in the community ceased to be the local teacher and was replaced by several indigenous people trained in bilingual education by an NGO [3].

After we completed our research, another team conducted further field work in rural municipios. The evidence collected, led them to argue that when it comes to local capacity "small might be better" (Rojas [1996]). Their paper suggests that, in fact, there appears to be some advantages in being small from the point of view of identifying community needs and establishing partnerships with non-governmental actors. To this last aspect I will now turn my attention.

Partnerships as a strategy for capacity enhancement

If the ultimate objective of a strategy of decentralization is achieving developmental impact (or effectiveness in the delivery of services) what matters is not just whether the local government will have the required capacity but the degree to which local actors (public and private) have or can develop it. This brings up the role of partnerships between the state and the private sector (business, NGOs, community groups) as a tool to enhance local capacity, the subject of an ongoing research program in Latin America I'm involved in (Fiszbein and Lowden, [in preparation]).

When we talk about partnerships, we are in fact recognizing that as a result of the trends towards decentralization, democratization and market reforms in most countries in Latin America there is a de facto re-definition of the boundaries and relationships between the state and society, particularly at the local level (Fiszbein and Crawford [1996]). Some people in Latin America are now insisting that one should differentiate very clearly between the "state" and the "public sector", where the latter involves a combination of state, civic and commercial interests that manifest themselves in public ways.

What we have learned over the course of an action-research program in six countries [4] is that state-business-civic partnerships have the potential to dramatically expand local capacity by (i) making more human and technical resources available for development initiatives; (ii) increasing the productivity of those resources; and (iii) creating new social assets that facilitate further partnerships at lower costs.

I'll present two examples that I hope will illustrate how these partnerships can radically alter local capacity. The first case refers to the municipio of Palpalá (population 40,000) in the northern tip of Argentina, where as a result of the privatization of the largest local enterprise approximately 25% of the labor force became unemployed. Several other small communities in the country have experienced similar shocks. In most cases, local governments did not have the capacity to deal with the challenge by themselves and waited in a passive manner for provincial or federal assistance. Palpalá, in contrast, faced this challenge starting from the recognition that what was required was to leverage the limited capacity in the hands of the municipal government with that of other non-state actors (Arroyo and Estebanez [1997]). A recently created municipio (as a result of a decentralization reform in the province) the municipal authorities belonged to a dissenting political group within the majority party in the region and could not thus count on provincial support. Their goal was the re-creation of a local economic base through micro-enterprises seeking a strong integration among them and between them, the municipal administration and the local schools in order to achieve high levels of competitiveness. The municipal government mobilized its limited financial, technical and professional capacity to promote this approach helping establish a network of partnerships and avoiding the dramatic social costs experienced by other municipios in similar circumstances.

Two of the Colombian municipios studied in World Bank [1995] appeared --by coincidence-- in a new research project dealing with partnerships allowing us to see how capacity had continued expending over a period of two years. In one of the municipios --Versalles-- our original study showed that, due to the traditional clientelistic style of the local administration, the municipal government had an extremely weak capacity compared to community organizations. Two years after the initial study, it was brought to our attention the existence of a local development corporation that is undertaking a number of income generating and social initiatives. This corporation involves a partnership between the local government, a private sector foundation operating from a city in the same region and a community organization. Over the previous two years, the community organizations continued expanding their capacity in a way that when there was a change in the local political leadership, the process of constituting the corporation was relatively easy.

These examples, and others similar in nature, suggest a link between decentralization, partnerships and enhanced local capacity. Our analysis is leading us to emphasize the enabling role played by empowered local governments in the conformation of these partnerships (with the consequent impact on capacity). It is not, I believe, simply a question of attitudes. Local governments in Latin America face a set of incentives that position them better than national governments for partnerships (particularly those implicit in local development approaches such as those I referred to above). Namely, they face lower costs and higher benefits from adopting a partnership approach than their respective national governments.

Even a motivated government willing to involve organizations of civil society (NGOs, CBOs, private sector groups) in projects leading to improvements in the living conditions of communities requires information on who are the relevant groups that should participate. That kind of information is easier to be obtained at the local level. Similarly, the benefits of such partnerships-that is, of a process of local development-are felt mostly locally or regionally but very rarely nationally.

In most circumstances, achieving the level of inter-sectoral cooperation implicit in a model of partnerships involves a drastic change in the way services are delivered or decisions are being made. In that sense, some interest groups are likely to be affected. Local governments, by virtue of starting from an "outsiders" position tend to be more inclined to challenge those interests than national governments that have developed allegiances and dependencies over many years of running national programs.

But whether such partnerships develop seems to be closely associated with political factors, as the cases of Palpalá and Versalles suggest.

Political competition and capacity development

The political dimension of decentralization is the subject of another working group in this technical consultation and I will not attempt to deal with this important issue in any systematic way in this note. There is, however, one angle to the capacity development question that forces us to consider the relationship between decentralization and political reforms.

It is a fact that in Latin America decentralization has been associated with broader political reforms that, in a variety of ways, have attempted to expand democracy and citizen participation. In the late seventies less than half a dozen countries in the region had elected national governments and among them only a couple had elected mayors and/or governors. Today, there are over 12,000 elected sub-national governments in the region (Campbell [1995]). Several countries have incorporated the promotion of citizen participation as a goal in their constitutional and legal frameworks (e.g. Bolivia, Brazil, Colombia, Venezuela). And the resolution of some long-standing armed conflicts in Central America is also opening the doors to more active collaboration between the state, NGOs and private sector actors.

While this remains a largely controversial topic even within Latin America, over the last few years we have seen an accumulation of evidence suggesting that political competition at the local level --together with a greater involvement by non-state actors in social programs-- is a strong motivator for local capacity development. Our work in Colombia, for example, offered strong evidence of the existence of political reform/capacity development link. Namely, we found that among Colombian municipios, competition for political office opened the door to responsible and innovative leadership that became the driving force behind capacity building. The combination of added responsibilities for service delivery, more fiscal resources and political reforms that increased competition seems to have created an environment conducive to more effective local governments.

Some very recent evidence from Central America suggests that an increase in the degree of competition in local elections is having a positive impact on the development of local capacity by motivating local leaders to invest financial, human and political resources to improve the effectiveness of public action in their communities.

Whether these trends constitute a sufficient --or even a necessary condition-- for the development of local capacity in other countries in Latin America and other parts of the developing world remains an open question which I hope we will be able to address during these consultations.

Notes

1. In conducting field research for World Bank [1995], we identified the case of a small rural municipio of about 7,000 that, unable to connect itself to the national telephone network, had established a sophisticated alternative communication system. The municipal government paid for a regular phone line in the nearest city, and bought an inexpensive radio equipment that allowed citizens (paying a per-minute fee) to connect to that line. I seriously doubt whether this municipal government established the toll-free number although they clearly had the potential capacity to do it.

2. While poor fiscal capacity will certainly limit what a local government or community can achieve,the way I understand institutional capacity is in terms of what outputs can be achieved with a given resource base.

3. There is growing evidence coming from Bolivia indicating that the 1994 Popular Participation Law is facilitating the development of innovative and effective locally-initiated programs to expand access and quality of education in rural areas. For an example where the beneficiaries were girls, see Montaño [1997].

4. The six countries are in alphabetical order Argentina, Bolivia, Colombia, El Salvador, Jamaica and Venezuela.

Bibliography

Arroyo, D. and M.E. Estebanez [1997], 'El caso del municipio de Palpalá', FLACSO, Buenos Aires, May.

Bird, R. and A. Fiszbein [1997], 'Fiscal decentralization in Colombia: The central role of the central government' in R. Bird and F. Vaillancourt (Eds.) "Fiscal Decentralization in Developing Countries" Cambridge University Press (forthcoming).

Bird, R. and F. Vaillancourt [1997], 'Fiscal decentralization in developing countries: An overview and perspective', Discussion paper No.11, International Centre for Tax Studies, University of Toronto, June.

Brinkerhoff, D.W. and A.A. Goldsmith [1992], 'Promoting the sustainability of development institutions: A framework for strategy', "World Development", 20, 3.

Campbell, T. [1995], 'Decentralization and municipal governance: The quiet revolution in Latin America', Interplan, American Planning Association, No.51, July

Cohen, J.M. [1995], 'Capacity building in the public sector: A focused framework for analysis and action', "International Review of Administrative Science", 61, 3.

Fiszbein, A. [1997], 'The emergence of local capacity: Lessons from Colombia', World Development, 20, 7

Fiszbein, A. and S. Crawford [1996], 'Beyond national policies: Partnerships for poverty reduction', Paper prepared for the World Bank's Annual Bank Conference on Development in LAC, Bogotá, July 1996.

Fiszbein A. and P. Lowden [in preparation], "Partnerships for poverty reduction: Cross sectoral collaboration for change in Latin America and the Caribbean", Economic Development Institute, The World Bank.

Galan, L.C. [1990], "Los municipios: Dueños de su propio destino", Bogotá: Procomún.

Goldsmith, A.A. [1993], 'Demand, supply and institutional development in Africa', "Canadian Journal of Development Studies", 14, 3.

Montaño, C. [1997], 'Propuesta pedagógica alternativa: Una experiencia en Mizque', CEDAGRO-UNDP, October

North, D. C. [1990], "Institutions, Institutional Change and Economic Performance", New York: Cambridge University Press.

Rojas, F. [1996], 'Colombia: Capacidad de gestión en un municipio pequeño semi-rural', Bogotá (mimeo).

Rondinelli, D.A. [1981], 'Government decentralization in comparative perspective: Theory and practice in developing countries', "International Review of Administrative Sciences", 47, 2.

Uphoff, N. [1986], "Local institutional development : an analytical sourcebook with cases", West Hartford, CT : Kumarian Press.

World Bank [1995], "Local Government Capacity in Colombia: Beyond Technical Assistance". The World Bank, Washington D.C.

World Bank [1996], "Reforming the decentralization law: Incentives for an effective delivery of services", Report no.15298-CO, The World Bank, Washigton D.C.