Abstract
Using a statistical methodology guided only by data and based on a genetic algorithm, we select the best econometric model for explaining the determinants of the size of the shadow economy, its main determinants being: taxes on capital gains of individuals, corporate taxes on income, profits and capital gains, domestic credit, bank secrecy, ethnic fractionalization, urban population, globalization, corruption and the socialist legal origin of country.
Acknowledgements
The authors are very grateful to Ulrich Thiessen for kindly providing the data set.
Funding
Financial support from the Spanish Institute for Fiscal Studies (project IEF 13/2013) is gratefully acknowledged.
Notes
1 We have data on 30 OECD countries (Australia, Austria, Belgium, Canada, Cyprus, Denmark, Finland, France, Germany, Greece, Iceland, Ireland, Italy, Japan, Latvia, Lithuania, Luxembourg, Malta, Mexico, Netherlands, New Zealand, Norway, Portugal, South Korea, Spain, Sweden, Switzerland, Turkey, United Kingdom and United States) and 8 Eastern European countries (Bulgaria, Czech Republic, Estonia, Hungary, Poland, Romania, Slovak Republic, and Slovenia).
2 See Thiessen (Citation2010, Appendix 1) for a detailed account of the variables included in each of these seven major groups of potential determinants.
3 See Friedman et al. (Citation2000) for a simple model of an entrepreneur's decision to operate officially or unofficially depending on corporate taxes.
4 To facilitate the interpretation, we have multiplied the original variable in Thiessen's database by -1, so a higher value of this variable indicates greater corruption.