Portal de Periódicos da CAPES

Corruption, Culture, and Communism

2005; Taylor & Francis; Volume: 15; Issue: 1 Linguagem: Inglês

10.1080/03906700500038678

1469-9273

Wayne Sandholtz, Rein Taagepera,

Culture, Economy, and Development Studies

Abstract Cultural factors, as measured by the two dimensions of values identified by Inglehart, explain 75% of the variation in the Perceived Corruption Index across non-communist countries. A strong 'survival' orientation contributes twice as much as a strong 'traditional' orientation to higher levels of corruption. When controlling for these cultural variables, communism and post-communism increase the levels of corruption even further, both directly and by contributing to heavier emphasis on survival values. Communism created structural incentives for engaging in corrupt behaviors, which became such a widespread fact of life that they became rooted in the culture in these societies – that is, the social norms and practices prevailing in communist societies. The transitions toward democracy and market economies have not yet erased this culture of corruption. In addition, the process of privatization itself has opened myriad opportunities for corruption. The effects are manifest in comparisons of corruption in non-communist and (post-)communist countries in five cultural zones. Acknowledgments A grant from the Center for Global Peace and Conflict Studies at the University of California, Irvine supported a portion of this research. We are extremely grateful to Ronald Inglehart for his generosity in providing us his data on dimensions of culture, derived from the World Values Surveys, and for providing us with insightful and constructive comments. We also appreciate the capable research assistance of Erik Faleski. Notes Wayne Sandholtz and Rein Taagepera, Department of Political Science, University of California, Irvine, 3151 Social Science Plaza, Irvine, CA 92697-5100, USA. Tel: +1-949-824-5726. Fax: +1-949-824-8762. See Rose-Ackerman (Citation1999). See, for example, Mauro (Citation1995, Citation1997), Knack and Keefer (Citation1996), Tanzi and Davoodi (Citation1997), Rose-Ackerman (1998), Gupta et al. (Citation2002). Potential reverse effects also must be kept in mind. Do higher incomes reduce the pressure to supplement meager incomes by corrupt means, so as to feed one's family? Does low initial corruption, however induced, enable democracies to develop in the first place? Do corrupt actors push for presidentialism and federalism so as to broaden their opportunities? Correlation analysis can hope to indicate the main direction of causation only when using lagged data, but past comparable data for corruption are limited. While recognizing that corruption itself may well be a causal factor for many other features, it is at least as much a result as a cause. Could it be that inherently lower corruption levels encourage people to shift to Protestantism? The Protestant idea of personal responsibility would sound adverse to people who accept corruption as a way of life. Selling indulgence letters (a form of heavenly corruption?) seemed to shock northern Europeans more than the Mediterraneans. The results of this ongoing inquiry have been presented in numerous articles and books, most prominently in Inglehart (Citation1990, Citation1997). We rely here on the idea of the cultural map, as presented in Inglehart (Citation2000), and on factor scores generously provided by Ronald Inglehart. The boundaries of the cultural areas in Figure 1 are our own. These clusters by no means imply immutability of values. World Values Surveys themselves document generational shifts. But the ability to delineate such clusters suggests that countries in the same broad cultural areas do tend to move in parallel. Question V213 in 1995–1998 read: 'How widespread do you think bribe taking and corruption is in your country? (1) Almost no public officials are engaged in it. (2) A few public officials are engaged in it. (3) Most public officials are engaged in it. (4) Almost all public officials are engaged in it. (5) Don't know'. It may be argued that the Transparency International norms for what constitutes corruption are not universal, being based on the so-called Protestant work ethic. This ethic is now heavily embraced by Catholic Europe as well, but only marginally by Orthodox Christianity and Latin America. In other world cultures different criteria of ethic behavior may prevail. This may well be so, but then they might have to accept poverty as inherent part of their culture. Respect for Protestant criteria of elite integrity tends to coincide with greater material wealth and well being, possibly because these criteria create trust and hence synergy. There is some correlation between Self-expression and Secular-rational in the set of non-communist countries; they have a Pearson's correlation of 0.61. However, in the multiple regression, the variance inflation factors (VIF) for the two variables are low (1.8), indicating little danger of collinearity problems in the model. The regression was also run to include the product of Secular-rational and Self-expression. The results: TableDownload CSVDisplay Table Consider these areas one hundred years ago. Given that Latin America differs from the West European Catholic area, the same could have been the case for East European Catholic countries; still, it's debatable whether Poland and the Czech lands were markedly less traditional or survival-oriented than Spain and Austria. Marked differences between them developed only during the last 50 years. The claim of pre-communist differences becomes even more tenuous when comparing Western Muslim countries ranging from French-controlled North Africa to the Ottoman and Russian tsarist realms. The same applies to the Confucian area. The Orthodox and Protestant areas enter our analysis only marginally, since the former has only one non-communist case (Greece) and the latter only one communist-affected case (Estonia). The gap is around 2 units for all areas except European Protestant, for which the communist subgroup has only one case. If the error on mean Self-expression and Secular-rational is 0.1 units, the error on the gap between two subgroups is magnified by a factor of 20.5, leading to 0.15. Error on the predicted gap is 20.5(1.73+0.80)0.15/2≈0.3. If the error on mean EI is 0.2 units, the error on the actual gap between two subgroups is 0.3 units. The difference between expected and actual gaps then involves an error larger by 20.5, meaning 0.45 units. The weighted mean for five cultural areas reduces the error again to 0.3. Additional informationNotes on contributorsWayne Sandholtz Wayne Sandholtz and Rein Taagepera, Department of Political Science, University of California, Irvine, 3151 Social Science Plaza, Irvine, CA 92697-5100, USA. Tel: +1-949-824-5726. Fax: +1-949-824-8762.