Posted February 1998
| 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
THE DECENTRALIZATION DEBATE is both broad and often frustratingly imprecise. Arguments for and against decentralization frequently assume the character of sweeping, cross-disciplinary claims about the effects of administrative measures on the quality and efficiency of both government and social interaction. Partly as a result of this, the economic and political literatures on decentralization and whether it increases or decreases social welfare and efficiency are still very much inconclusive. This paper reviews some of the more important rationales for decentralization to date, examines their theoretical underpinnings, and then goes on to describe the elements of a new theory of decentralization which might take us far in our understanding of its effectiveness and implications. Lastly and most importantly, we review a large amount of new data on the effects of decentralization in Bolivia. We find strong evidence that devolving resources and power to local government has (i) increased the allocative efficiency of the Bolivian public sector, (ii) led to greater stability in cross-sectoral public investment patterns, (iii) led to more responsive, needs-oriented government locally than was ever achieved by the center, and (iv) very likely increased the cost-effectiveness of the public sector.
We begin providing a clear and analytically rigorous definition of decentralization which, because it is more restrictive than many in the literature, lends itself more readily to theorizing. We then move on to the political science literature on decentralization, concentrating on modern arguments of how decentralizing the state can lead to better and more efficient government. The fundamental objection to these arguments, and one which is generalizable for this literature, is that it simply assumes that central government will produce more standardized, less-differentiated outputs less suited to local preferences than local government. Although this is intuitively appealing, the lack of an explanation for how this comes about amounts to assuming away the problem. We can easily draw up a model where central government installs agents in each community to gather information and detect local needs and preferences, which data is then relayed at low cost back to the center. The paper outlines such a model and explains why, in theoretical terms and with moderate assumptions, the optimal governmental architecture in this context is not decentralization but rather hyper-centralization.
We then turn to the economics literature, beginning with the Tiebout model of "voting with your feet". We comment that the approach of a competitive equilibrium in locational decisions - though in some ways appealing - would seem to model the wrong dynamic. We note that a more useful mechanism would incorporate regular elections, and hold voters fixed relative to local administrations which change periodically. We then discuss information - the economics of which is still a fairly new field - as the central element in the central vs. local government debate. But we are wary of the potential of information as a rationale for decentralization. We explain that the main features of this approach do not concern information per se, but rather how it is put to use. As such, this argument is really about the incentives that governing agents face, and the relationship between this and the administrative architecture of government. In this vein, it can be shown that local government will enjoy advantages in allocative efficiency over central government to the degree in which local officials' professional incentives are more in line with the interests of the local population than the incentives of central government officials. This condition will obtain where local officials are fully accountable and responsible to the local population, and where electoral representation is sufficient to ensure that all groups have a voice in local affairs. We describe the model in detail and discuss the assumptions necessary for it to work.
The paper then turns to new evidence from a dramatic case of decentralization which is particularly suitable for empirical work - Bolivia. We outline the important features of Bolivian decentralization, and move on to the data. Devolving power and resources to independent local governments does seems to change the composition of public spending and investment considerably. Local governments significantly increased investments in Health, Education, and Civil Works projects compared to the central government before the reform, while investing less in Industry & Tourism, Hydrocarbons and Communication. We note the important role of smaller, poorer and more rural municipalities in bringing about this result. We then examine recent municipal spending in several sectors compared to objective social and demographic indicators of need, and find that municipalities treat social expenditures as investments in human capital, with the familiar characteristics of decreasing marginal returns. Taken together, these results point to a highly rational model of local decision-making where the fundamental criterion is need, and governmental outputs are conceived of as the equivalent to productive capital, with the marginal investment going to those municipalities where return is highest. This directly contradicts claims that local government is too poor, too ignorant, or too prone to interest-group capture to operate efficiently, necessitating the guiding hand of national government which is technocratic, capable, and generally knows what to do. Here we begin to see evidence of the opposite: local government has a deep understanding of its task, and has the capability and the incentive structure to produce the public outputs that people want.
Lastly, the paper examines data from over 350 hours of interviews carried out by the author in a series of case studies in nine Bolivian municipalities. We focus on the interactions between social, political, and institutional variables in a dynamic context, with a richness of detail which econometric studies cannot approach. We examine a number of the tenets of conventional wisdom concerning local government, only to find that they are directly contradicted by the evidence, and examine possible explanations for this. We conclude that this data supports a view of local government in full accord with the previous findings, and with the theoretical model based on representation and accountability which is outlined above.
In its conclusion, the paper proposes an ordinal ranking of the most important problems facing local government, lists the minimum conditions necessary for decentralization to work, and makes general suggestions for reorienting aid programs aimed at institution-building.
|
