Since I started working in the anticorruption field a few years back, I’ve noticed that a substantial amount of the discussion in this field—at conferences, in journals, on blogs like this one, etc.—is given over to debates about definition and measurement. This is something I’ve discussed, and complained about, before (see here, here, and here)—though I concede that every time I bring this up, I’m contributing to the very problem I’m complaining about.

Now, one of the reasons there’s so much debate about definition and measurement in this field is because corruption is, relative to other concepts, particularly difficult to define and measure. Another reason—in my mind the main one—is that while “corruption” is sometimes used as a purely descriptive term (that is, to describe certain conduct, which we can try to measure empirically), it is also an evaluative/normative term—one that connotes “bad” behavior of a certain sort. So any attempt to define corruption (for purposes of positive analysis or empirical research) will often, perhaps inevitably, suggest a normative position on the sorts of conduct, people, or institutions that ought to be condemned.

That’s not an original point, nor even a terribly interesting one. But the more of these “what is corruption” conversations I’ve been a part of, the more I get the sense that there’s a more specific political/ideological subtext to some of the arguments about how corruption should be defined. Nobody ever articulates these ideas in so many words, and so I may be way off base, but I’m going to offer up some conjectures, in this post and in the next one, about what I sense is the ideological subtext of some of these definitional debates.

Here I’ll focus on a fairly narrow issue: Should those organizations that focus on (and sometimes try to measure) “corruption” emphasize forms of corruption that involve the public sector (government, or entities with a sufficiently close connection with government to be considered essentially public instrumentalities), or should the “anticorruption agenda”—as well as the definition and measurement of corruption—also include purely private sector corruption?Of course, there’s no abstract right answer to this question. From an analytical or research perspective, there are good arguments for considering public and private sector corruption as manifestations of the same basic problem (essentially a principal-agent problem). There are also good arguments for treating public sector corruption as distinctive (mainly because governments have different powers, objectives, and constraints compared to private firms, making the causes and consequences of corruption sufficiently different to warrant separate treatment). Again, neither of those positions is correct in the abstract; whether it makes sense to group public and private corruption into a single category, for research purposes, depends on the nature of the research question.

Yet somewhat surprisingly (at least to me), this seemingly mundane methodological question gets quite a bit of airtime, and declarations that corruption simply must be defined to include private sector corruption can get surprisingly passionate. Why is this? Again, I can totally see someone saying something like: “The basic analytical structure for private sector corruption and public sector corruption are sufficiently similar that differentiating them is unhelpful: A public procurement officer who takes a kickback for allocating a public contract to Firm X rather than Firm Y is doing basically the same thing as a buyer for a private firm who purchases supplies from Firm X rather than Firm Y in exchange for a bribe. The causal factors, such as cultural norms and institutional incentives, in both cases are likely to be essentially the same. So we should group these phenomena together when defining and measuring ‘corruption.’” But I’ve never heard anyone actually make the case in those terms. Instead—and here I admit I’m speculating—the case for including private sector corruption in the definition of corruption seems more ideological. The subtext seems to go something like this: “The focus on public sector corruption as opposed to private sector corruption is driven by a neoliberal [Boo! Hiss!] assumption that governments are bad. Those who define ‘corruption’ as ‘government corruption’ are suggesting that corruption only exists in the public sector, and further implying that the solutions to corruption are deregulation [Boo!] and privatization [Hiss!].”

OK, by interspersing a bit of sarcasm, I’ve already signaled I’m not so enamored of this argument, and I’ll explain why in just a moment, but before I do, let me say that there’s certainly a version of the argument that I actually agree with quite strongly: If our measures of corruption only pick up government corruption, but equivalent corruption occurs in the private sector, then privatization would indeed create the illusion of decreased corruption, even if nothing on the ground had actually changed for the victims. Suppose that in Country A, electricity is supplied by a public utility, while in Country B, electricity is supplied by a private firm. And suppose in both countries, 10% of citizens can’t get an electricity connection without paying a bribe to the employee responsible for connecting their houses to the grid. If only public sector corruption “counts” in our measure, then Country A will seem more corrupt than Country B. Moreover, if Country A privatizes its electric utility, it will appear—according to measures that only capture public sector corruption—to have become less corrupt, even if bribery rates stay the same or increase.

So that form of the argument is clearly correct. But my sense is that this isn’t actually the main concern of those who I’ve heard making the most forceful arguments for including private sector corruption in the definition of corruption, and (perhaps more significantly) on the agenda of major anticorruption activist organizations. The ideological subtext seems to be more the idea that those who choose to focus on public corruption are implicitly conveying and endorsing the idea that governments and public employees are somehow morally worse or more dysfunctional than those in the private sector. Insofar as that latter instinct really is part of the argument for considering private sector corruption together with public sector corruption, I respectfully disagree, for two reasons:

• First, as noted above, in some contexts there are very good analytical reasons for considering public sector corruption to be distinct from private sector corruption. Though there are important similarities, there are also critical differences. The absence of market competition and associated market discipline, the fact that in many contexts the government has a monopoly on providing crucial functions, and the fact that a government agent’s “principals” (the citizenry) are more diffuse, and the available control mechanisms are more limited, all mean that public corruption will often be a sufficiently distinctive phenomenon to merit separate consideration. This won’t always be true. But when it is, it would be a shame for an ideological agenda to get in the way of conceptual clarity and good analysis.
• Second, and equally importantly, the ideological critique of the focus on public corruption—the idea that this focus abets a neoliberal deregulation agenda—gets things not only wrong, but backwards. The focus on specifically government corruption derives, I would suggest, not from an implicit sense that the public sector is deficient relative to the private sector, but from a sense that governments have a special and distinctive role to play in promoting social welfare, one that cannot be adequately replaced (at least in many contexts) by the private market. The traditional focus of organizations like Transparency International and Global Witness—hardly crusaders for neoliberalism—is on public sector corruption precisely because of a sense that the public sector owes special obligations to its citizens, such that corruption in the public sector is destructive to social welfare in a way that other forms of intra-organizational dishonesty may not be. I don’t want to push this point too far: Some forms of purely private sector corruption can do massive damage to society. But I do want to push back against the idea that a focus on public sector corruption implies some kind of libertarian skepticism of the capacity for activist governments to make a positive difference in people’s lives. In fact, this focus may imply precisely the opposite view.

My larger point here is that I think we would all be well-served to bring to the surface what I sense are the political-ideological subtexts of what appear to be methodological questions. I’ll continue to pursue that line of thought in my next post.