For all their flaws, the major cross-country corruption indexes—Transparency International’s Corruption Perceptions Index (CPI), the World Bank Institute’s Worldwide Governance Indicators (WGI), and the like—have been quite useful, both for research (at least when used appropriately) and for advocacy.  But one important limitation of these datasets is that by focusing on corruption (or perceived corruption) at the country level, they may obscure the fact that there can be substantial within-country variation in the level of (perceived) corruption.  This variation may occur across government institutions—the same country may have quite different degrees of corruption in the health sector, the police force, the judiciary, customs, etc.  More pertinent here, there may also be significant heterogeneity across regions, particularly in large countries with substantial political decentralization.  Indeed, numerous studies have exploited within-country regional variation in corruption levels to test various hypotheses about corruption’s causes and consequences; such studies include research on Italy, Russia, China, the Philippines, and the United States, among others.  But these studies typically make use of particular data sets that are not reproduced year-to-year.

As we’re starting to see rapidly diminishing returns from the major cross-country corruption datasets, it is high time for those organizations with the resources and capacity to compile information on corruption perceptions on an ongoing basis to turn their focus to within-country regional variation in corruption.  I propose the creation of a sub-national corruption perceptions index (snCPI), starting with the so-called BRICS countries (Brazil, Russia, India, China, and South Africa), which would gather and compile data (primarily perception-based data, perhaps supplemented with more objective data when available) on perceived corruption levels within the major sub-national units (states/provinces, autonomous regions, and municipalities) within each of those countries.

The creation of such indexes seems feasible (though admittedly resource-intensive, at least at first), and once created, these snCPIs would be useful for both research and advocacy.

From a research perspective, the advantage of studying the causes of within-country variation in corruption, as opposed to cross-country variation, is that a within-country study controls for many (though of course not all) of the cultural, historical, and political factors that may hamper cross-country comparisons.  And corruption is more likely to take similar forms within a single country.  Additionally, though concerns about the accuracy of perceived corruption measures would persist, comparisons across regions within the same country might be more reliable, at least when many of the same survey respondents might have experience in multiple regions.  In contrast, cross-country perception studies often require respondents to compare corruption levels in very different countries, and individual respondents often have only limited experience with some of the countries in question.

From an advocacy perspective, just as the international cross-country indexes helped raise the profile of corruption as a problem and may have stimulated some valuable international competition, an intra-national cross-region index may prompt a healthy form of competition among jurisdictions, particularly in countries that are trying to attract foreign investment and mobile businesses.  For example, think about the incentives of a Chinese provincial governor or party secretary: These officials —and their bosses in the central government—may care a great deal about how their province ranked relative to its neighbors in some public index of corruption, and this could stimulate some productive “race-to-the-top” style competition.  Furthermore, in a democracy like India or Brazil, one could easily imagine that a province or city’s snCPI score could become a significant issue in local elections.

In terms of practicality, it’s true that creating snCPIs would require more resources than the creation of composite indexes like the CPI and WGI.  Those indexes, after all, aggregate existing data sources, most of which are available only at the country level.  So, what I am proposing would most likely require original data gathering efforts, which would require resources and institutional support.  At least initially, the most promising sources of data would be surveys of foreign investors and, where possible, domestic businesses.  The ideal survey respondents would be those who have had dealings in several jurisdictions within the same country, who could therefore compare corruption levels in different places.

The BRICS seem like the ideal targets for snCPI indexes.  Of course, there’s no reason one would need to limit snCPI indexes to BRICS; other large federal countries like Mexico or Indonesia are natural candidates as well, though again, one needs to start somewhere, and the extent of foreign investor experience in the BRICS and their larger economic significance seems to make those countries a good place to start. If creating an snCPI for each BRICS country seems too daunting, one could imagine starting with a pilot project in one or two of the BRICS countries, and if it proved successful, expanding from there.

Of course, it’s easy for me to sit here and propose this as an academic idea.  The question is whether anyone out there with the necessary resources and institutional backing would be interested in pursuing this.  TI?  World Bank?  Looking at you…