Errors in the measurement of corruption indices help rich countries avoid accountability

"In a goal-oriented society, metrics matter. What we measure affects what we do. If our measurements are wrong, we will set the wrong goals." These are the arguments of the authors of the Economic Measurements Report of 2008. The Commission disputed the role of the GDP indicator as the main measure of development, but its conclusions can also be applied to corruption, which is traditionally (and wrongly) measured only as a one-dimensional problem. .

Global corruption indices, including Transparency International's Corruption Perceptions Index (CPI) and the World Bank's Corruption Perceptions Index, give each country a separate score. The result of this approach is that rich countries are constantly found to be "very clean" and poor countries "very corrupt". For example, in the 2023 CPI, the UK (with a score of 71) ranks 20th on the list of least corrupt countries in the world, much cleaner than China (42) and Brazil (36). Most of those who use the CPI, including the media, firms and analysts, treat these figures as fact.

But are rich countries really less corrupt than poor ones? One-dimensional indicators, such as the CPI, conceal the fact that qualitatively different forms of corruption cannot be reduced to a single result. This approach systematically undervalues ​​what I call 'corruption of the rich', which tends to be legalised, institutionalized and ambiguously unethical, in contrast to 'corruption of the poor'.

In poor countries, corruption takes explicitly illegal and scandalous forms such as theft of public funds or giving/receiving bribes. Many believe that this problem has disappeared in rich countries. Moreover, in The Quest for Good Governance, Alina Mungiu-Pippidi concludes that developed countries have reached a final state of "ethical universalism" in which "everyone is treated equally." Great Britain, followed by "parts of the British Empire, populated mostly by inhabitants of European descent," in this sense it is a “classic historical example.” In other words, the rich West is clean.

Yet given the current rise of populism in high-income democracies, which is largely a reaction to the vast advantages enjoyed by the wealthy and politically connected, "ethical universalism" seems more illusory than real. As revealed in a 2020 New York Times op-ed, during the covid-19 pandemic half of the British government's medical supply contracts went through a special VIP channel to "companies run by friends or acquaintances of politicians".

So how did the CPI rank the UK as the 20th least corrupt country in the world? The country rating is not based on surveys conducted internally by Transparency International, but on a combination of data from various surveys conducted by third parties. Almost all of them come from Western organizations (eg the Economist Intelligence Unit), and are usually heavily influenced by the answers given by Western managers.

In addition, the wording in these surveys is often very vague. For example, one source of the CPI is the World Competitiveness Index (VCI). This survey presents business managers with a rough binary choice: "Bribery and corruption - do they exist or not?" It is no wonder that rich countries are "very clean" year after year in the CPI index, although ordinary citizens do not agree with this opinion.

Since there is still no alternative to these traditional indices, although they are often criticized (even by the creators of the CPI index), I have compiled an experimental "United Corruption Index" (UCI). Like the CPI, the UCI measures perceptions of corruption and is based on expert surveys. But he divides corruption into four distinct variations: petty theft (eg extortion coming from police officers), grand theft (embezzlement of funds by politicians), money for speed (small bribes to overcome bureaucratic obstacles or trouble), money for access (large payments in exchange for exclusive, lucrative privileges like contracts or financial aid).

While the first three types of corruption - endemic in poor countries - are blatantly illegal and directly harmful, access to money can be illegal (as in bribery) or permitted (as in campaign financing). Sophisticated methods of buying privileges can involve entire institutions where no individual is corrupt. For example, the process of money laundering (and the city of London is one of the famous centers for this) implies the smooth movement of funds across borders - with the help of reputable financial institutions. In the US, banks collectively spent billions of dollars lobbying for lax regulation that led to the 2008 financial crisis. However, only one banker was charged.

The UCI index uses its own expert survey to rank all four types of corruption. In order to improve the quality of the measurement, I use the vignette method: respondents are asked to assess the reality of specific scenarios, rather than the general level of corruption. My prototype, which covers 15 countries, shows each country's overall UCI score, but breaks it down further into four categories of corruption, with the dominant view highlighted (in color). We now have the opportunity to compare not only the aggregate level of perceived corruption, but also the types and configuration of corruption in different countries.

An enlightening comparison between the US and China. In general, America is less corrupt than China, but in the "cash for access" category (which is the dominant form of corruption in both countries) the gap between them is the smallest. Notably, America scores higher in the money-for-access category than some lower-income countries, such as Thailand and Ghana. If we used only the total score, then we would conclude that America is clean, without corruption. But once we break down the overall result, then we can begin to explain what the populist call to "drain the swamp" means.

Even more interesting is that different forms of payment for access to money prevail in the US and China. When assessing the likelihood of bribery scenarios through the personal networks of individual politicians, China clearly dominates. But when we consider the "revolving door" (the transfer of employees from the private to the public sector and vice versa) or the takeover of regulation by private interests through lobbying, then America comes out on top.

In other words, in the US corruption in the "cash for access" category is mostly institutional, while in China it is still on a personal level, including bribes and piles of hidden money. It cannot be said with certainty that China is a more corrupt country than the US, but corruption in China certainly has different qualities.

Errors in measuring the corruption index are not only technical. Basically, they confirm the illusory, hypocritical and generally Eurocentric notion that high-income countries have achieved a permanent state of ethical purity. In reality, corruption may not disappear as countries become richer. Instead, it evolves, becoming more complex and less noticeable.

We must continue to fight the "corruption of the poor". But by detailing corruption, capitalist democracies can draw much-needed attention to some of their most pressing problems. These include growing inequality, declining public trust in government, and "modern corruption," as USAID Administrator Samantha Power calls it (an example of such corruption is transnational networks of illegal financing). To solve these problems, you must first measure them accurately, instead of pretending they don't exist.

The author is a professor of political economy at Johns Hopkins University

Copyright: Project Syndicate, 2024. (translation: NR)