Business regulation and taxation: effects on cross-country corruption

Abstract

This paper adds to the literature on the government-corruption nexus by examining the effects of taxation and various business regulations on cross-country corruption. Regulations considered include the number of procedures and related costs for business startup, licensing and property registrations. Results show that regulation, not taxation, generally positively impacts corruption and the effects of non-monetary regulatory costs are more significant than monetary costs. Findings are generally robust to an alternate corruption measure and to simultaneity between corruption and regulation. Results for ‘standard’ determinants of corruption largely support the literature. Policy implications are discussed.

Keywords:

• corruption
• regulation
• taxation
• business startup
• licenses
• property
• transition
• democracy

JEL Classifications:

• H26
• H87
• K42

Acknowledgments

Comments by Iikka Korhonen, two referees, Public Choice meetings participants, and research assistance of Mike Naretta and Robert Stanford are appreciated. An earlier version of this paper was circulated as a BOFIT Discussion Paper #10/2008.

Notes

1. A widely accepted definition of corruption is the abuse of public office for private gain.

2. The other religions are the reference group.

3. Our hypotheses posit linear relationships between these regulatory determinants and corruption. In practice, some of the relations could very well be nonlinear.

4. Note that the number of observations varies across the different models estimated due to missing values for some variables.

5. However, appropriate caution needs to be exercised in interpreting results based on perceived, rather than actual, corruption (see Treisman 2007, also Williams and Siddique 2008).

6. We also employed the corruption perceptions index from the World Bank (WB) as an alternate corruption measure (http://www.govindicators.org). The correlation between the CPI and WB was 0.98 and the two indices yielded very similar results. Hence, corresponding findings are not reported.

7. In the case of taxation, the corresponding measures are the number of tax payments and the time taken to pay taxes (in hours).

8. A cross-sectional approach is employed because the available cross-national corruption indices, especially the CPI, are not well-suited for time series interpretation.

9. The relatively larger R ²s in Panel A compared to Panel B in Table 2 signify that the chosen explanatory variables are collectively more effective at explaining corruption perceptions (captured by the CPI index) than they are at explaining the risk of corruption (captured by the ICRG index).

10. According to Djankov et al. (2002, p. 3) “The creation of rents for the bureaucrats and politicians through regulation is often inefficient, in part because the regulators are disorganized, and in part because the policies they pursue to increase the rents from corruption are distortionary. The analogy to tollbooths on a highway is useful. ..sIn a political equilibrium, however, each town through which the road passes might be able to erect its own tollbooth. Toll collectors may also block alternative routes so as to force the traffic onto the toll road. ..sIn the tollbooth theory the regulation of entry enables the regulators to collect bribes from the potential entrants and serves no social purpose.”

11. Some governments have recognized this aspect and have even tried to make the duration exogenous. For instance, the Indian government has placed maximum limits on the time taken to issue passports and other identification documents.

12. See Table 1 for details about these instruments.