Fiscal decentralization, corruption, and the shadow economy.

Author:Alexeev, Michael

    This paper analyzes the implications of a particular form of fiscal decentralization for the incentives of local governments to provide productivity enhancing local public goods and extort bribes from local entrepreneurs, and for the size of unofficial or shadow economy. More specifically, we show that an increase in the share of locally raised tax revenue left with the local government raises its incentives to provide productivity-enhancing public goods and brings more entrepreneurs into the official economy. Interestingly, corruption, measured by the size of the bribe that the local government charges entrepreneurs for letting them operate officially and use public goods fully, may increase or decrease, depending on the extent to which public goods enhance productivity of the entrepreneurs. The implications of the model with respect to the size of shadow economy are tested using cross-sectional country-level data.

    We define fiscal decentralization as the share of locally collected tax revenue left at the disposal of local government. We obtain three results from our model, at least two of which are novel. First, the provision of local productivity-enhancing public goods by the local government increases in the degree of fiscal decentralization. This rather straightforward result is quite intuitive given that in our model local public goods are produced only by local government. (1) Second, the amount of corruption measured by the bribes extorted from entrepreneurs operating in the official economy may either increase or decrease depending on the elasticity of entrepreneurs' production functions with respect to the supply of public goods by the local government. This result is surprising, because the straightforward intuition is that if the local government receives a greater share of locally collected tax revenue, it would have incentives to reduce bribes that limit the entry of tax-paying entrepreneurs into official economy. If production is highly elastic with respect to public goods, however, the local government may to be able to charge higher bribes for the access to a greater amount of public goods it produces. In the most empirically-relevant case, however, greater fiscal decentralization reduces corruption. Third, and most importantly, we find that greater fiscal decentralization always reduces the share of entrepreneurs operating in the informal economy. This result is particularly interesting in light of the ambiguous result with respect to the link between decentralization and corruption in our model.

    Given that the link between fiscal decentralization and corruption has been tested elsewhere (e.g., see the survey by Fjeldstad 2004), we focus on testing the third implication of our model and obtain results that strongly support it. The relationship between the degree of fiscal decentralization and the estimated size of the unofficial economy is negative and highly statistically significant in both in OLS and in instrumental variable regressions, and is robust to the inclusion of various controls.


    In our model, the entrepreneurs operating in the official sector are taxed by the central government and also may have to pay bribes to the local government that provides local public goods such as contract enforcement and property rights protection. Bribes may also be paid for getting licenses that entitle entrepreneurs to operate officially and use public goods. The entrepreneurs have an option to escape both taxation and bribes by shifting their activities underground, but in this case they would not be able to access public goods fully. (2)

    We assume that taxes imposed by the central government and shared with localities depend on the entrepreneur productivity, while bribes charged by local officials are lump-sum. (3) We argue that to a large extent the difference between taxation and bribes stems from the more limited administrative and technical resources of bribe-takers. Due to limited administrative capabilities of corrupt local officials, bribes are imposed in a relatively unsophisticated manner. In particular, bribe-takers cannot make bribe amounts strongly contingent on the individual characteristics of the entrepreneurs such as their productivity. For example, it is hard to imagine that officials issuing permits to open a small business would somehow be able to make the amount of the bribe contingent on future productivity of the entrepreneur. Of course, these officials do not use their bribe revenue to produce public goods, but their official wages are presumably lower than in the no corruption environment, leaving more tax revenue for public good production. Therefore, other things equal, even bribes collected by an individual official can be viewed as potentially useful for the production of public goods.

    We assume that the local government maximizes its rents, i.e., total revenue net of the expenditure on public goods. (Alternatively, we can think of the local government maximizing the funds that can be used for producing public goods that do not enhance productivity of the entrepreneurs such as support for the elderly and the poor.) The main tradeoff faced by the local government is between the desire to increase rents by increasing bribes and reducing the provision of public goods, and the fact that these actions shrink the size of the official economy, reducing the government's revenue base. The resources for the provision of public goods by the local government come from its share of the tax revenue collected on its territory. The local government may also spend part of its bribe revenue on public goods if this would enhance its overall revenue (from both taxes and bribes) net of all expenditures. (4)

    The role of the central government in our model is passive. Both the tax rate and tax revenue share transferred to local government are treated as exogenous. The behavior of the central government is assumed to be influenced by a number of different groups in the economy, including the local governments and the entrepreneurs who impose potentially complicated constraints on the behavior of the central government. Therefore, instead of modeling these constraints, we examine the comparative statics of the local government behavior with respect to the parameters set by the central government. (5)

    Our analysis is limited to the situation when the entrepreneurs' ability to move to a different locality is restricted. While this assumption is difficult to justify for developed economies, it reasonably reflects the environment in many developing countries and the economies in transition where regional disparities are often large and persistent. In fact, the differences in the degree of mobility of entrepreneurs may explain why corruption is greater in some countries than in others. (6)


    While the link between decentralization and corruption is a subject of vast literature, (7) the relationship between fiscal decentralization and shadow economy has been examined in only a handful of papers. A recent paper by Teobaldelli (2010) is particularly relevant to our study. This paper presents a simple model that compares a country with competing local jurisdictions to a completely centralized country. Similarly to the standard results on horizontal tax competition, the overall tax rate is lower in a federal country. Moreover, competition among jurisdictions disciplines the behavior of local officials. For these reasons, the federal country in the model has lower tax rates and spends all of its tax revenue on the provision of public goods. This reduces the size of the shadow economy relative to the centralized country. Teobaldelli then tests the model's predictions using cross-country data for (more or less) democratic countries. As is explained above, our model is quite different from Teobaldelli's. The main difference is that while Teobaldelli's results are driven by the inter-jurisdictional mobility of the (identical) agents, such movement is assumed away in our model with heterogeneous agents. (8) In addition, Teobaldelli's model does not include the ability of the government officials to receive bribes from entrepreneurs. In terms of estimation approach, the main contribution of our paper relative to Teobaldelli's is our use of instrumental variable estimation that alleviates potentially serious endogeneity and measurement problems.

    Torgler et al. (2010) also shows a negative relationship between decentralization and the size of the shadow economy. This purely empirical paper, however, relies exclusively on the data for Switzerland and similarly to Teobaldelli (2010) does not employ instrumental variables estimation.

    While our model is quite simple, it combines taxation administered by the central government, locally produced productivity-enhancing public goods, bribes charged by the local government from heterogeneous entrepreneurs, and the possibility for the latter to escape both bribes and taxation in the unofficial economy. We are not aware of any other model in the decentralization literature that has this combination of features. For example, in a model somewhat similar to ours, Choi and Thum (2005) examine the effect of the presence of the shadow economy option on the amount of corruption, but they do not consider any taxes and do not focus on decentralization. Treisman (2006) models taxes and bribes, but he has no unofficial economy option in his model. Brueckner (2000) considers decentralization in the presence of corruption, heterogeneous consumers, and tax evasion, but his focus is on the implications of these factors for the Tiebout model rather than the impact of the degree of decentralization on corruption and unofficial economy size in a given locality. Moreover, in Brueckner's model the amount of corruption, its welfare costs, and tax evasion are exogenous. Ahlin's (2005) model does...

To continue reading