Are Democracies Least Corrupt?

May 8, 2009

[First published October 27, 2005] One of the extraordinary characteristics of dictatorships, especially absolutists ones, is their government corruption. This comes out in biographies of those who, for example, have lived in North Korea or in South Vietnam when it was defeated and occupied by the North. And under authoritarian regimes, this corruption seems only marginally less, as under the Chinese Nationalists before their defeat by Mao. My impression, consistent with that of others, has been the democracy is among the least corrupt types of government.

Now, this has been tested. Transparency International has provided for 2005 a perception of corruption index for 146 nations (here). Kenneth Sikorski added to this index the freedom house ranking of nations on their civil liberties and civil rights (from here), which measures their freedom, and found that the index included 67 free, 45 partly free, and 34 unfree nations (excluding North Korea). He then averaged these three political groups on their perceived corruption, as shown below (total scores for all nations in the group/number of nations in the group — personal communication):

Free (2901/67) = 43.3
Partly Free (4076/45) = 96.6
Not Free (3470/34 = 102.05

So, partly free and not free nations are perceived to be over twice as corrupt as democracies. This is another plus for democracies, of course. They don’t war on each other, have the least domestic violence, virtually never kill their own people, experience no famines, and also are least corrupt.

This gets almost embarrassing after awhile in relating this to people who ignorant of research on the democratic peace, as I did in a talk today. It seems that one is obsessed with a one-factor theory of humanity’s major problems. This runs counter to general intuition, and to common sense in the social sciences, which is that the socio-political world is complex with multiple causes and conditions interacting to produce events. No one factor is sufficient, so it is felt. Well, there is one major factor, and that is democratic freedom. The evidence, such as the above, is always available to doubters, if only they will look at it. All I can say is what Galileo Galilei said when his astronomical observations were doubted and he was persecuted for them. “Look through the telescope,” he responded.

Global Corruption And Democracy

January 12, 2009

[First published February, 2006] Kenneth Sikorski has tested whether democracies are the least corrupt compared to other forms of government. He showed they are, using Transparency International’s 2005 Corruption Perceptions Index (CPI). Transparency International has just released its Global Corruption Report 2006, which includes Chapter 10 on “Ten years of the CPI: determining trends,” by Johann Graf Lambsdorff. Their global index for previous years is here.

Lambsdorff found that:

Overall, our findings indicate that significant improvements between 1995 and 2004 occurred (in descending order of significance) in Estonia, Spain, Italy, Bulgaria, Mexico, Hong Kong, Colombia, Costa Rica, Taiwan, Australia, Iceland and Russia. Deterioration, on the other hand, was significant in Argentina, Ireland, Poland, Czech Republic, Zimbabwe, United Kingdom, Ecuador, Indonesia, Turkey, Canada, and the Philippines.

In a following chapter, “Governance matters IV: new data, new challenges,” Lambsdorff discusses governance indicators covering 209 countries for 2004. These indicators are based on 352 different underlying variables measuring perceptions of a wide range of governance issues. The variables are drawn from 32 separate data sources constructed by 30 different organizations worldwide. For those of you interested in global performance and the effect of freedom, this report and those discussed and linked below are a bonanza.

One of the findings is:

that there is a strong causal impact of institutional quality on per capita incomes worldwide. Figure 12.1 [shown below]shows a representative set of estimates of this “development dividend” of good governance. These estimates suggest that a realistic one-standard deviation improvement in governance would raise incomes in the long run by about two- to-threefold. Of course, there is variation around these relationships, since governance is not the only thing that matters for development – but it certainly is a very important factor deserving policy-makers’ attention.

The rule of law, as measured on the X axis, is a major indicator of democracy, and as shown is closely related to a countrie’s wealth — its GDP per capita. Note that this is logged, which means that the wealth of countries curves sharply upward with the presence of the rule of law.

Then, there is chapter 13 on “Corruption in the United States of America” by Edward Glaeser and Raven Saks. This is measured by the number of public officials convicted for corruption in each of the 50 US states. They find that “states with higher incomes and a larger share of college-educated population are less corrupt.” States that are most corrupt during 1976-2002 are Alaska, Mississippi, Louisiana, South Dakota, Tennessee, Illinois, New York, Oklahoma, Montana, and North Dakota. States least corrupt are Colorado, Wisconsin, Nebraska, Utah, Iowa, New Hampshire, Minnesota, Vermont, Washington, and Oregon. The authors conclude:

In general, the patterns documented in the data for US states reveal the same basic relationships that have been found using international evidence. This similarity is particularly interesting given that, here, corruption is measured using federal conviction data rather than the type of opinion survey that is the norm in the cross-country literature.


“Wolfowitz’s Corruption Agenda”:

In sum, Wolfowitz’s World Bank presidency, which had seemed to lack an organizing theme, has acquired one. The new boss is going to be tough on corruption, and he’s going to push this campaign beyond the confines of the World Bank; [and he has] persuaded the heads of several regional development banks to join his anti-corruption effort.

RJR: The empirical results mentioned above and in the links below provide strong support for Wolfowitz’s campaign.

“Legal Corruption”:

We undertake to identify general determinants of the pattern of legal and illegal corruption worldwide . . .

RJR: One of the things that the study found is “that fundamental accountability may [play] a clear role in development. This may be a key variable in the determination of corruption in richer societies — policies oriented to its reinforcement may be very fruitful.” What is most important about this is the way accountability was measured — the freedom of the press. This is also a cental indicator of democratic freedom.

“Myths and Realities of Governance and Corruption”

A number of popular notions and outright myths on governance and corruption are addressed in this chapter. We distinguish clearly between governance and anti-corruption, while probing the links between both notions. In so doing we challenge the conventional definition of corruption as being too narrow, legalistic and unduly focused on the public sector, while underplaying the role of the private sector.

“Corruption, Governance and Security: Challenges for the Rich Countries and the World”:

We suggest that the undue emphasis on narrow legalism has obscured more subtle yet costly manifestations of misgovernance, which afflict rich countries as well….Further, we find that governance constraints, and corruption in particular, is a key determinant of a country’s global competitiveness. These findings challenge traditional notions of what constitutes the country’s ‘investment climate’, and who shapes it. It is also found that illegal forms of corruption continue to be prevalent in the interaction between transnationals of the rich world and the public sectors in many emerging countries. Finally, we suggest an empirical link between governance and security issues.

My latest democide painting