Corruption in Transition Economies: Effects of Government Size, Country Size and Economic Reforms

Article excerpt


This paper uses annual pooled data over 1998-2002 for transition countries to examine whether government size or country size matters more in its impact on corruption; and whether piecemeal reforms or comprehensive transition reforms are desirable for corruption reduction. Our results show that greater economic prosperity leads to lower corruption, and contrary to findings for other nations, a bigger government size seems to reduce corruption in transition nations. The geographic size of a country is positive and significant, suggesting that more spread out countries would have a harder time controlling corruption. Comprehensive transition reforms might work best at corruption reduction.

(JEL H1, K4, P2)


The determinants and effects of illegal activities have intrigued researchers for quite some time. In this broad class of illegal activities, studies on corruption have garnered their share of attention. Empirical investigations of corruption have proliferated in the last decade due to the availability of consistent indices of corruption (or indices of the perception of corruption, to be precise) across countries (see Méon and Sekkat (2005) for a recent example; Jain (2001) for a survey).

Attention to the prevalence of corruption in transition nations is relatively more recent, especially due to two main reasons. First, since transition nations have been in existence for less than two decades, adequate and reliable information on corrupt practices was not forthcoming in the initial years. Consequently, many existing corrupt activities escaped scrutiny and those that did come to the surface were probably treated "kindly" during the "honeymoon period". However, from the second half of the nineties, the spread of corruption was recognized by international financial organizations as "the single largest obstacle to development" (The World Bank), and financial and technical assistance they were to provide was linked to a country's anti-corruption efforts. Secondly, due to technical reasons, researchers had to wait a few years before data for enough years were available to conduct meaningful studies of corruption in these countries. Finally, a number of transition nations have been eager to join the European Union and corruption control has been an important criterion for selection.

Transition countries are in somewhat of a unique situation. On the one hand, they have the benefit of learning from the best practices (corruption control policies and contractual mechanisms) of other countries; on the other hand, the massive structural adjustments needed in the transition phase (e.g., privatization on a large scale) likely created unique opportunities for rent-seeking activities, both legal and illegal.

There is now a small, but growing literature on the determinants and effects of corruption in transition countries [see Abed and Davoodi (2000), Budak and Goel (2004a,b), Eilat and Zinnes (2002). Kaufmann and Siegelbaum (1996), and May et al. (2002)]. These studies have focused on numerous determinants of corruption including the effects of economic and political freedom, and governance, etc. (Budak and Goel. 2004a.b; May et al., 2002). This paper adds to this literature by examining new dimensions of determinants of corruption in these countries - namely, the effects of government size and country (geographic) size on the prevalence of corrupt activities. Does a bigger government or a bigger country size contribute more to corrupt activities? The effects of government size have been studied earlier [Goel and Nelson (1998, 2005)]. but we are not aware of the country size being included as a determinant of corruption. To that extent this study also makes a contribution to the literature on the economics of corruption at large (surveys of the literature can be found in Aidt. 2003; Bardhan, 1997: Jain. 2001; and Rose-Ackerman, 1999). From the policymaking perspective, country size is largely exogenous but government size is endogenous, especially over time. …