In my post last week, which reacted to Dani Rodrick’s discussion of the political situation in Turkey (and also to some of the commentary on recent anticorruption enforcement patterns in China), I noted the ambivalence that many people (myself included) feel about anticorruption enforcement that is simultaneously (1) legitimate (in the sense that there is evidence that the targets have indeed violated the law) and (2) politically-motivated (in the sense that the targets may have been selected not only, or not primarily, because of the alleged corruption, but also because of partisan or factional political conflict).
One thing that I should have made clearer in the post, but didn’t, is that concerns about politically-motivated anticorruption enforcement are not limited to developing countries. Indeed, there’s some fairly strong evidence of partisan political bias in anticorruption enforcement in other countries, like the United States. The strongest evidence that I know of is a terrific paper by the political scientist Sandy Gordon, which finds very strong evidence that the U.S. Department of Justice is more likely to prosecute state and local officials for various corruption offenses if those officials belong to a different political party than the one that controls the White House. (This effect was particularly strong during George W. Bush’s presidency.)Now, to be clear, I’m not arguing that President Bush, or any other President, used anticorruption laws to engage in a purge of political opponents. There’s a big difference between (1) the possibility of some bias at the margins (a willingness to pursue legitimate but minor cases against political opponents, while tending to give co-partisans a pass in similar circumstances), and (2) the deliberate asymmetric use/abuse of anticorruption laws to damage the political opposition. Likewise, I’m not suggesting bad faith on the part of U.S. prosecutors; there’s plenty of evidence that bias can be subconscious. But all that said, the work of Gordon and others is a useful reminder that anticorruption enforcement in wealthy Western democracies is not an entirely pristine exercise, untainted by political bias.
Perhaps this is also a usual occasion to note the difficulty in figuring out how much anticorruption enforcers should be politically insulated. On the one hand, a politically independent enforcement agency can resist pressure by powerful incumbent politicians either to lay off the incumbents’ allies or to go after their adversaries. On the other hand, independent enforcement agents may have their own agendas, and their own political biases, and we might reasonably worry that without adequate accountability, these entities could abuse their authority (intentionally or unintentionally). I’m not sure we’ve quite worked out how to manage the trade-off between independence and accountability in the anticorruption context.
GAB | The Global Anticorruption Blog 
I wonder if targets of politically-insulated, unbiased anti-corruption probes would still dismiss any charges against them as being politically motivated. Regional leaders in India cry “political conspiracy” when they face anti-corruption probes, even when they are in power. If this is the case, how much should neutrality matter to an anti-corruption probe?
I’m sure targets would (almost) always object in this way. The question is how credible these objections are. If the anticorruption enforcement authorities have a reputation for political neutrality, targets will have a harder time undermining support for the prosecution by claiming that it’s political. On the other hand, if the enforcement authorities are widely perceived (whether correctly or not) as biased, then an official who is targeted, and who is willing to take on the anticorruption authorities (refusing to cooperate, denying their legitimacy, etc.), may be more likely to get public support, or at least neutrality.
I wonder if it is plausible to figure out to what extent anti-corruption enforcement is politically motivated by analyzing its consequences/effectiveness. It seems to me that politically motivated enforcement will fade away when other factions are weakened or even eliminated through such enforcement, and as a result corrupt practices might increase since there is less check-and-balance among different factions. I was thinking of different waves of anti-corruption enforcement that took place in China during the past three decades but were all fruitless in general.
Independence is critical in pursuing anti-corruption efforts. One of the problems in Turkey is precisely that they did not have an indepedent anti-corruption body. In my view accountability is best ensured by (1) internal accountability: codes of conduct, disciplinary procedures, objective criteria for case take-up, conflict of interest regulations, etc. and by (2) external accountability: public reports, reports to parliament, parliamentary oversight, auditing and, if you like, oversight by other independent public bodies (as in Hong Kong). What indepedence entails is reflected in the Jakarta Principles. See: http://www.iaca.int/images/sub/activities/EPAC/Jakarta_Statement.pdf. These have recently been recognised by the International Association of Anti-Corruption Authorities in its Panama Declaration (Nov. 2013) and by the Conference of States Parties to the UNCAC in its resolution on prevention of corruption. At the moment, anti-corruption bodies around the world are rarely fully independent. Only if they are truly independent, will the allegations of politicization subside.
Very interesting points. I’m generally sympathetic to your argument, and of course independence from political interference is critical to the success of anticorruption enforcement efforts, especially when high-level targets are involved. That said, I wonder whether the tension between the interests in independence and accountability might not be somewhat harder to resolve than your post (or the Statement principles) imply. Take “parliamentary oversight,” one of the forms of external accountability you note. I agree that too little, or the wrong kind of, parliamentary oversight can lead to serious accountability problems — and the danger of a “runaway” or “rogue” agency with its own agenda. But too much, or the wrong kind of, parliamentary oversight can lead to precisely the lack of independence that you rightly criticize. And some of the other “soft” forms of external accountability (like public reports, etc.), while surely valuable, may be quite limited in their effectiveness.
More generally, I’m glad you noted the Jakarta Principles, and their recent endorsement by the IAACA. I think that may be worth a separate post, after I’ve had a chance to reflect a bit further.
I’m very sympathetic to your acknowledgment that corruption of all forms (including corrupt anticorruption enforcement) is present in the U.S. (and likely other industrialized nations), and not only the developing nations where much of the attention of the anticorruption movement exists.
I found particularly interesting the caveat you included about how you’re not arguing that the discrepancy means that a political purge was ongoing. I think that’s right, but it’s curious how much more difficult it might be to identify with confidence certain forms of corruption or abuse of power in advanced industrialized nations. I’m not as familiar with the politically motivated anticorruption prosecution literature in developing countries, but I wonder whether it’s been your experience that legal and political scientists are willing to extend presumptions of good faith more readily when studying industrialized or Western countries than when studying developing countries?
(Of course, political outrage over the IRS-Tea Party scandal notwithstanding.)
Interesting question. It’s hard to generalize, but I actually don’t think political scientists are generally more likely to presume good faith when studying industrialized or Western governments. I actually think social scientists are sometimes too quick to infer deliberate discrimination, or other forms of bad faith, from statistical evidence of bias, even when dealing with wealthy democracies.
That said, I think the question you raise is a good one, and I can definitely think of examples where observationally similar behaviors are treated as evidence of bias or bad faith when certain countries are involved, but not when other countries are involved (possibly fairly, possibly unfairly). For example, the Chinese prosecution of Western pharmaceutical companies has led to speculation that China is using its anticorruption law as a protectionist tool. The fact that most of the biggest U.S. FCPA enforcement actions have been against non-U.S. companies has led to some grumbling in Europe and elsewhere, but not very much complaint (especially within the U.S.) that the U.S. uses its anticorruption laws for protectionist purposes. Now, there might be good reasons for this. But there does seem to be an undercurrent, in some of the discussion, of the idea that Chinese enforcement authorities couldn’t actually take these anticorruption laws seriously. Maybe, maybe not. But merits aside, I do think you see a bit of what you’re describing, where some countries get more of the benefit of the doubt with respect to their motives than do others when it comes to anticorruption enforcement.
I don’t know whether political scientists and economists are more willing to extend the presumption of good faith to Western or industrialized economies, but there are studies showing that “corruption experts,” those whose opinions form the basis for perception measures of corruption, are biased in favor of wealthier, democratic governments. See UCLA Professor Daniel Treisman’s “What Have We Learned About the Causes of Corruption from Ten Years of Crossnational Empirical Research?” and Dilyan Donchev and Gergely Ujheli, “What Do Corruption Indices Measure,” both of which can be easily found on the web.
I will have more to say about this bias and its implications for the validity and reliability of perception measures of corruption in a forthcoming post.
I think this is a very interesting point, and is extremely relevant whenever there is a global ranking of corruption or “cronyism.” Perception in a complicated and value laden thing, which other posts have explored. However, I do think this is a slightly different issue when it is played out in a domestic context.
The source of bias noted by Rick is less relevant when it comes to bias against Democrats and Republicans. In the American context I think an independent anti-corruption agency really comes down to the classic Administrative law arguments of accountability and independence (as Matthew noted in his reply). Given the observed inclination for politically-motivated anti-corruption enforcement, I think independence wins to the extent we want to penalise corruption fairly regardless of party affiliation. Maybe I place too much trust in the expertise/peer-effect argument, but I think the normal mechanisms used to control independent agencies (funding, judicial review, removal for cause etc.) should be similarly sufficient to control rogue agency behavior in the corruption context.