Incentives and Fiscal Independence of Local Budgets

Annotation
Over the last few years, a growing amount of attention has been devoted to the role of government in such areas as property rights protection and contract enforcement, in determining the economic performance of the economies in transition. [See, for example, references provided in Chapter 8 of Roland (2000). Among the papers not mentioned there I would like to mention Polishchuk and Savvateev (1997) and Treisman (2001).] The importance of government performance for the economy can be illustrated, for example, by the contrast in economic performance of Poland and China on one hand, and Russia on the other, during the 1990's. While there were obviously many significant differences between these countries, the fact remains that despite implementing very different economic reform policies, the economies of Poland and China were growing fast, while Russia's economy was contracting. One possible conclusion fr om this observation could be that factors other than purely economic policies play a significant role in determining economic performance during transition. The contrast between Poland and Russia is particularly instructive, because their economies prior to reforms were distorted in many similar ways and both countries used similar economic reform policies. The comparison with China is also important, because although Chinese economic policies often went against conventional economic wisdom, the Chinese economy has performed quite well.

Frye and Shleifer (1997) were perhaps the first to present survey-based evidence of the differences in government performance in Poland and Russia with respect to property rights protection and contract enforcement. They interviewed shopkeepers in Moscow and Warsaw and found that, among other things, the role of organized crime in providing contract enforcement and protection of property rights as well as in extortions fr om businesses was much more important in Russia than in Poland. Also, government regulations in Russia looked much more oppressive than in Poland.

But why does government work reasonably well in one economy in transition and fails in another? In one of the most influential papers on the subject, Shleifer (1997), who concentrated on the difference between government performance in Poland and Russia, considered several potential explanations for this difference. After rejecting such explanations as shock therapy policies, the amount of "social capital" and "trust" in society, and cultural antagonism to capitalism, he suggested that the difference in government performance was due mainly to the following two reasons: (1) the entry of new politicians into Polish local governments versus little change in the local political cadre in Russia and (2) the difference in the incentives of the local politicians in the two countries.

I agree that both of these factors played a significant role. The change of management that introduces new human capital into privatized enterprises has been shown to be important for innovation and restructuring.[ Barberis et al. (1996).] However, the change of human capital does not appear to be sufficient for improving government performance. After all, while the turnover of the local politicians in Poland was indeed significantly greater than in Russia, there is little evidence that government worked better in those Russian regions wh ere such turnover did take place. Also, it appears that the turnover of the local politicians in China, at least early in the reform process, was not particularly large.

Whatever the importance of renewal of human capital of the local politicians, the role of their incentives is clearly crucial. New politicians who do not have appropriate incentives are unlikely to improve government performance. Therefore, in the remainder of the paper, I will concentrate on the explanations related to the politicians' incentives, particularly because these explanations appears to be most often used in the literature.

Shleifer stressed two factors that affected the incentives of the local politicians. First, he emphasized the beneficial role of elections. Elections are obviously important because the politicians who face elections are at least somewhat accountable to the voters and, therefore, have incentives to incorporate the voters' interests in their behavior. In addition, the need to finance election campaigns creates incentives for the local politicians to treat their potential donors (primarily small businesses) favorably. But elections are not absolutely necessary for the politicians to have the right incentives. Even the unelected Chinese politicians appear to have the need to provide their constituencies with the appropriate public goods. As long as the funds for these public goods come from the local tax base, the politicians would have incentives to enhance this tax base.

However, the local tax base is not the only source of funds for the local public goods. There are two other sources. First, local politicians could seek transfers and subsidies from the higher level budget. I will call this a rent-seeking approach. Second, they can raise funds through illegal means such as by extorting bribes from local businesses. These three alternative ways of raising funds have different characteristics in terms of social efficiency.

The rent-seeking approach is clearly inefficient from the social point of view. Lobbying higher budget levels for budgetary funds is costly and it is a purely redistributive activity. Even if the allocation of funds from the upper level budget is relatively non-discretionary and the amount of lobbying is therefore limited, the fact that funds come from elsewhere creates no incentives to develop the local tax base. In fact, it can even push the politicians to hide it in order to enhance the apparent need for transfers and subsidies.

Extorting bribes from the local businesses and other economic actors can at first appear to give the politicians incentives to foster the bribe-giving ability of these actors. However, corruption is associated with significant social costs. Shleifer and Vishny (1993) emphasized the need for secrecy that is costly by itself and could easily result in significant distortions if it is easier to extort bribes from some economic activities than from others. Also, decentralized corruption can be particularly harmful as independent bribe-takers ask for bigger than socially optimal bribes even if secrecy is not a significant factor. Moreover, the need for secrecy prevents the efficient administration of bribe-taking. For example, bribe takers could hardly do audits or base bribes on more or less sophisticated indicators such as profits that require detailed record-keeping. As a result, bribes are often based on gross sales, which is a highly distortionary way of collecting funds from businesses. Wei (1997) stressed the uncertainty created by the presence of corruption. This factor is particularly important in limiting the new investment by entrepreneurs who may not be familiar with the rules established by the corrupt local officials. In addition, in order to enhance their ability to extort bribes, the local officials often impose regulations that make it difficult for the law-abiding businesses to operate. The bribes can then be charged for getting around these regulations. Meanwhile, the regulations impose costs on those businesses that prefer to operate completely legally.

Other things being equal, raising funds from the local tax base via reasonable taxes is (at least on the margin) the most socially efficient method of obtaining resources for the provision of local public goods. Most important, this method creates strong incentives for the politicians to develop their tax base. This method also does not suffer from the aforementioned inefficiencies associated with bribes.

Which of the three methods of fund raising is employed by the local politicians is largely determined by the degree of fiscal independence of the local budget. I define the degree of fiscal independence as the extent to which the budget of a given level is the residual claimant of that portion of the tax collections which is sensitive to the efforts of the local officials. Notice that the tax collections depend both of the efforts of tax collectors and on the size of the tax base. The local authorities in Russia do not collect taxes themselves even though they can affect the efforts of locally based federal tax collectors. However, the size of the local tax base is strongly affected by the actions of the local officials.

In the absence of fiscal independence, the changes in the size of local tax collections have little effect on the funds available to the local politicians. Instead, the increases in collections accrue to the upper level of government, while the declines are compensated by the increase in transfers from the upper level budget. In this environment, the local politicians are better off trying to obtain transfers from the upper government levels rather than fostering their own legal tax base. In addition, the politicians have incentives to raise funds via corruption because these funds are not shared with the upper government levels.[ Barberis et al. (1996).] Therefore, in the absence of fiscal independence, the local authorities have no incentives to foster the development of their own tax base.

There are also some possible disadvantages to fiscal independence, because it may be associated with fiscal decentralization, resulting in non-uniform regulations and tax rules.[ See, for example, Tanzi (2001).] Notice, however, that fiscal independence does not need to be accompanied by decentralization of regulations and revenue collection. Tax decentralization may be necessary to tailor tax rates and regulations to the local circumstances, but as was stressed above, fiscal independence could be achieved even under a uniform tax system, provided that the efforts of local officials can affect the size of the tax base.

Given its importance for the politicians' incentives, it is not surprising that several recent papers attempted to assess the degree of fiscal independence in different economies in transition. [Zhuravskaya (2000), Jin et al. (1999), and Alexeev and Kurlyandskaya (2001).] This assessment can be done in at least two different ways. One approach is to evaluate the formal and informal rules according to which tax revenues and transfers are assigned to different budget levels. For example, in China, lower government levels have to contribute certain more or less fixed revenues to the upper level budgets, but after that they largely control the use of the remaining revenues. In Russia, however, the regional governments have much greater discretion in determining the final revenues of local governments.

Perhaps an even a better way would be to look at what happens to the transfers from the upper level of government when local own revenues either increase or fall. The most careful work on this was done at the level of regional budgets. Thus, Zhuravskaya (2000) analyzed the budgets of 35 large Russian cities in the mid-1990's and found that whenever own revenues of these municipalities changed in either direction, the transfers to them from the regional governments changed in the opposite direction and by about the same amounts. In other words, when a local government managed to increase its own revenues by 1 ruble, the subsidies from the region declined by 1 ruble. Alexeev and Kurlyandskaya (2001) detected the same trend in the budgets of smaller municipalities in the Rostov oblast'.
It appears that in China this type of offsetting is almost absent. Jin et al. (1999) reported that an increase in own revenues of the Chinese provinces by 1 yuan resulted only in 0.16 yuan decrease in the subsidies from the center. A similar arrangement presumably exists at the lower government levels. I have not seen any work of this type on Poland, but Shleifer (1997) claims that the main source of local government revenues in Poland are local taxes and fees and that subsidies from the center do not significantly offset the changes in these local revenues.

While fiscal independence enhances the incentives of local politicians to develop their own tax base, it also generates certain important costs. To the extent that revenue shortfalls happen independently of the local authorities' efforts and are not compensated by the transfers from the upper budget level, the local populations may suffer reductions in the provision of public goods. Therefore, fiscal independence may result in significant inequities in public good provision across localities. The optimal degree of fiscal independence is strongly affected by the preferences over this equity-efficiency tradeoff.[ The equity-efficiency tradeoff in this context is analogous to the tradeoff between the strength of the agent's incentives and misallocation of risk in a standard principal-agent model with risk-averse agent.
] Alexeev and Kurlyandskaya demonstrated, however, that as long as the efforts of local authorities significantly affect tax collections, it is optimal for the upper level politicians who value their own funds to give these local authorities at least some measure of fiscal independence. This happens largely because when local officials have no incentives to expand their tax base, the upper budget level on average has to spend more resources on transfers to the local budget.

The lack of fiscal independence of local governments in Russia and its presence in Poland and China may explain the differences in the incentives and, therefore, performance of local authorities in these countries. All this does not explain, however, WHY the upper levels of government in Russia have not been willing to give a measure of fiscal independence to the lower levels. In particular, what distinguishes the circumstances of the regional authorities in Russia from those in Poland and China that lead to their different attitudes toward fiscal independence of municipal budgets?

In my opinion, the explanation has to do with the pre-reform conditions and with the political situation at the early stages of reform in these countries. First, from the region's point of view, fiscal independence is useful only if the municipality's behavior can significantly affect its own revenues. (Otherwise, there is no reason to give it incentives to try to increase these revenues.) In most countries, the conventional own tax base that is more or less responsive to the efforts of local authorities in the short run is the small business sector. Therefore, the region would be interested in giving its municipalities fiscal independence only if these municipalities could significantly affect tax collections from the small business sector. In the short run, the municipalities' efforts in this regard work is not likely to work via the creation of new enterprises, because that presumably takes a significant degree of confidence in the permanence of the business-friendly policies of the local authorities. Besides, start-up firms usually do not generate positive profits or cash-flow and, therefore, do not contribute to tax revenues for some time after their creation. Therefore, the municipal revenue enhancing efforts work mostly through the increase of output of the existing firms. This effect, however, very much depends on the initial size of the small business sector. If this sector is negligible, the best efforts of municipal authorities would not change local revenues appreciably and most of the short-run fluctuations of their own revenues could be safely assumed to be independent of their efforts. If the small business sector is substantial, the efforts of municipal authorities could result in relatively large increases in own revenues even in the short term.

On the eve of reforms in Poland, the private sector (presumably small enterprises) accounted for 29% of industrial employment and 16% of industrial output. For the entire economy the respective percentages were 47% and 30% (Ernst et al, 1996). The above argument implies then that in Poland, fiscal independence of localities would have been important for the upper levels of government early on, which in turn would create incentives for the local governments to create the conditions that foster further development of the small business sector.

On the other hand, the legal private sector in Russia outside of agriculture was miniscule prior to the beginning of reforms. In the fall of 1991, only about 2.5% of Russia's industrial output was produced at above ground non-state-owned enterprises (Ivanov and Kolbasova, 1992). Employment in cooperatives, which were essentially small private enterprises, comprised about 3% of all non-agricultural sector employment in 1991 (Rossiia-1993, 1993). Even during 1992, the first year of reforms, small businesses employed slightly more than 10% of non-agricultural labor force. Moreover, a quarter of this employment concentrated in Moscow and St. Petersburg (OECD, 1997). Given the small size of the small business sector in Russia, the efforts of municipalities would not have made much difference in terms of fostering the tax base in the short run. Therefore, the regions were quite rational not to value fiscal independence of municipalities much, at least in the short run. And, of course, if regions did not give fiscal independence to the municipalities, the latter had no incentives to foster the small business sector, and this in turn continued to justify the regions' unwillingness to give fiscal independence to the municipalities, creating a vicious circle that impeded economic growth. This logic is supported by the fact that small business employment in Russia grew only moderately during 1992-96 period (OECD, 1997).

While the above argument may explain the difference between the behavior of upper level governments in Poland and in Russia, it cannot be used to distinguish between Russia and China, wh ere there was no legal private business sector prior to reforms either. Notice, however, that the initial size of the small business sector matters a great deal only if the regions have a short planning horizon. Over the long term, the extent of creation of new firms arguably becomes more important than the expansion of the original businesses. The length of the planning horizon of the upper level authorities is determined mainly by the degree of institutional stability of the country's political system and the overall degree of uncertainty in the country. In both respects, China appears to have significant advantage over Russia, implying that Chinese authorities at all levels must have had longer planning horizons than the corresponding authorities in Russia. This, in turn, would make upper level governments more interested in providing some fiscal independence to the lower level authorities. Moreover, the longer term perspective would presumably also improve the incentives of entrepreneurs to expand the existing businesses and start new ones for any given investment climate.

To summarize, both the initial size of the small business sector and the degree of political and economic uncertainty at the early stages of reforms might have resulted in rational reluctance by the upper levels of government in Russia to give a significant degree of fiscal independence to the lower levels, thus impeding the incentives of the lower government levels to foster local economic growth.

Russia's reformers have for some time understood the importance of fiscal independence of the lower budget levels, particularly the municipalities. They have also understood that the regional authorities did not have the incentives to provide this fiscal independence. To a large extent, many of the attempted and proposed fiscal federalism reforms in Russia since 1997 have been aimed at forcing the regions to provide greater degree of fiscal independence to their constituent municipalities. Without trying to diminish the importance of these reforms, the above discussion suggests that until the regions and the central government themselves come to appreciate the advantages of fiscal independence of lower levels of government, these reforms will at best be incomplete.

References

Alexeev, M. and G. Kurlyandskaya. 2001. "Fiscal Federalism and Incentives in a Russian Region," mimeo.

Barberis, N., M. Boycko, A. Shleifer, and N. Tsukanova. 1996. "How Does Privatization Work? Evidence From the Russian Shops," Journal of Political Economy, 104,4:764-90.

Ernst, M., M. Alexeev, and P. Marer. 1996. Transforming the Core: Restructuring Industrial Enterprises in Russia and Central Europe, Westview, Boulder, Colorado.

Frye, T, and A. Shleifer. 1997. "The Invisible Hand and the Grabbing Hand," American Economic Review, 87, 2:354-8, May.

Ivanov, V., and A. Kolbasova, Eds. 1992. Rossiiskaia ekonomika v 1991 gody, Moscow, Russia: Institute for the Economics of Transition.

Jin, H., Y. Qian and B. Weingast. 1999. "Regional Decentralization and Fiscal Incentives: Federalism, Chinese Style," Stanford University, Working Paper, SWP-99-013.

OECD. 1997. Russian Federation 1997, OECD Economic Surveys, Organization for Economic Cooperation and Development, Paris, France.

Polishchuk, L. and A. Savvateev. 1997. "Spontaneous Emergence Of Property Rights: A Critical Analysis," mimeo.

Roland, G. 2000. Transition and Economics: Politics, Markets, and Firms, The MIT Press, Cambridge, Mass.

Rossiia-1993: ekonomicheskaia kon'iuktura. 1993. No. 1, Moscow, Center for Market Conditions and Forecasting, February.

Shleifer, A. 1997. "Government in Transition," European Economic Review, 41:385-410.

_______, and R. Vishny. 1993. "Corruption," Quarterly Journal of Economics, 108, 3:599-617.

Tanzi, V. 2001. "Pitfalls on the Road to Fiscal Decetralization," Carnegie Endowment Working Papers, Paper no. 19, April.

Treisman, D. 2001. "Corruption, Fiscal Incentives, and Output in Federal States:
On the Neutrality of Fiscal Decentralization," mimeo, March.

Wei, S. 1997. "Why is Corruption So Much More Taxing than Taxes? Arbitrariness Kills," NBER Working Paper Series, Working Paper 6255, November.

Zhuravskaya, E. 2000. "Incentives to Provide Local Public Goods: Fiscal Federalism, Russian Style," Journal of Public Economics, 76:337-68.