political economy of federalism in Russia, The

The debates on the optimal size of a government bureau, the efficient organizational pyramid of the bureaucracy, the political economy of rent-seeking activities, and income redistribution are inextricably linked to the study of federalism in the context of the political economy and the economics of property rights. In this article, I will concentrate on the problems of implementing a federal structure in a command economy undergoing a dual transition-to the market and to democracy-in Russia. By analyzing Russian federalism, I will attempt to shed more light on one of the root causes of many of Russia's economic and political problems: a lack of bureaucratic reform, with an emphasis on the consolidation of the rule of law.

As a guideline to this analysis, it is useful to consider two approaches to analyzing federal structures. First, there is the "benevolent government" framework that assumes that the government wants to increase social welfare by reducing transaction costs and providing a desirable level of public goods. The second approach analyzes federalism as any private sector institution that emerges because of unexploited gains that rational agents wish to internalize. There is hardly any direct emphasis on social welfare in this approach, and any marginal improvements in aggregate welfare are results of the welfare-maximizing characteristics of politics conducted in a democratic, market-oriented economy. The government in this approach is viewed as a leviathan.

As an example of the first approach, consider the decentralization theorem. Oates suggested, quite simply, that if a unitary system of governance imposes welfare costs that are higher than those of a federal system, then we could expect a more compelling case for more governance units.' If the unitary government provides an amount of public service that is the average of its major administrative units, for instance, then the welfare cost it imposes on the system will vary positively with the heterogeneity of demand among the units. Therefore, the average amount of service supplied by the federal government would be more acceptable if the distance between the highest demanders and the lowest demanders were less. This insight is very simple in its logic and has the merit of being easily testable. A further benefit is the transportability of the argument in analyzing centralized redistribution across goods and services for which a subunit demand exists in the system. The trouble, however, with this line of reasoning is that it loses its value in assessing new proposals, where it must rely only on an ex post analysis.

The elasticity of demand for publicly provided goods and services is vital to this theory, owing to its immediacy to welfare cost calculations. Given the Soviet Union's centralized command economy, with a near absence of private enterprise, this issue is a point of divergence between the analysis of Soviet-type economies and other economies at similar levels of economic development, where price elasticity for publicly provided goods is agreed to be less than -1.2 The economic transition process itself then becomes important in assessing the welfare costs imposed by the country's transition to federalism. This is an important point. Federalism may well be a political institution, but neglecting the private sector economic transition will lead to a biased analysis. To the extent that the private sector reduces dependency on the state, a positive analysis of how best to achieve a federal structure can be undertaken without any compunction.3

It is also generally acknowledged, under the framework of the benevolent government paradigm, that macroeconomic stabilization policies are better managed from the center. We can arrive at this conclusion by considering, for example, the Tiebout hypothesis. The more differentiation is possible among the characteristics of various localities, the more localities there can be in number. Then, given fully open policies, individuals or households will "vote with their feet" by moving to the locality that best suits their preferences and tastes. Although this appears to be unquestionable support for decentralization, the problem is that expecting the Tiebout hypothesis to yield a stable equilibrium seems unfruitful.4 However, the assumptions of the hypothesis-zero moving costs, zero externalities of migration, scale economies, and so on, analyzed in Buchanan's theory of clubs5-virtually prescribe that stable equilibria for localities will be hard to characterize unless we consider smaller units like suburbs and inner cities where differentiation through self-selection and explicit control can be more subtle. The point, though, is that with a fluid population and open domestic borders, central control over certain expenditures, primarily redistributive in nature, is likely to be more effective because local government expenditures will have a weaker impact on a desired outcome.6

The second approach that I mentioned above considers the objective of the state or local governments to be not vastly different from that of various other private sector organizations. An example of this comes from Brennan and Buchanan, who suggest that the central government facilitates contracting between lowerlevel governments by acting as an enforcer and seeks compensation by partaking in the benefits that ensue.' In addition, various other local governments can impose restrictions as correctionary measures against a local government that dissents or reneges on a contract. This has come to be called the "collusion theory."