Some aspects of a comprehensive anticorruption strategy are specifically targeted at corruption itself. But sometimes it makes sense to consider how broader political or economic reforms might ameliorate or exacerbate the corruption problem. Indeed, fighting corruption is often invoked – perhaps sincerely, perhaps strategically – as a justification for more general political and economic liberalization. But the relationship between political and economic liberalization, on the one hand, and corruption, on the other, is complicated, and beset by two seeming paradoxes:

Here’s the first paradox: On the one hand, longstanding democracies seem to have lower levels of corruption than do non-democracies. However, the process of democratization – the introduction of democratic reforms, along with a general liberalization of the political system – often seems associated with a significant increase in corruption. Indeed, countries going through democratic transitions, or those that have been democratic for a shorter time, seem not only to have more corruption than established democracies, but also to have worse corruption problems, on average, than non-democracies.

The second paradox, on the economic side, is similar: Some research suggests that more open economies – with more market competition, fewer state-owned enterprises, and less central economic planning – have lower levels of corruption than more statist economies. (This should not be overstated: it’s not that more regulation always increases corruption, nor is there convincing evidence that “bigger” governments, measured by the size of the public sector relative to GNP, have more corruption. Still, the balance of the evidence suggests that more open, liberal economies have less severe corruption than more statist economies.) However, the process of economic liberalization—including privatization, deregulation, etc.—often appears associated with an increase, often a dramatic increase, in corruption.

So, we seem to have two parallel paradoxes: Liberal political systems have lower corruption, but liberalizing the political system increases corruption; liberal economic systems have lower corruption, but liberalizing the economic system increases corruption. What’s going on? Let me suggest three (or maybe four?) possible interpretations.The first possibility is that these seeming “paradoxes” are illusions, the result of inaccurate data or conceptual confusion. For example, maybe the process of democratization (or political liberalization more generally) doesn’t actually increase corruption, but merely causes corruption to take more visible forms, or enables citizens and journalists to discuss corruption more openly. This might create the perception that democratization led to increased corruption, when in reality actual corruption was steady or decreasing. Or maybe longstanding democracies do not really have significantly lower corruption levels, but rather benefit (in the various perception surveys often used to measure corruption) from “halo effects” (countries perceived as good on some observable dimension, like political freedom, are assumed to be good on some other, unobservable dimension, like control of corruption); perhaps newly democratizing countries don’t benefit from this “halo”, but are instead perceived as chaotic and unstable. Similar data problems could explain the relationship between economic liberalization and increased (perceived) corruption. For example, if economic liberalization creates more wealth, or more opportunities for government officials to expropriate wealth, then the increase in perceived corruption may be due simply to the fact that there is more to steal, even if the rate of theft is actually declining.

But let’s assume for the moment that it is true – and not merely a data illusion – that political liberalization increases corruption (even though established democracies have lower levels of corruption than non-democracies), and that economic liberalization increases corruption (even though open economies have lower levels of corruption than statist economies). What might explain that? This brings us to two other possibilities, one of which is more optimistic, the other of which is more pessimistic.

The more optimistic possibility is that the spike in corruption during periods of liberalization is a temporary phenomenon, associated with rapid transition, which will dissipate after the transition is complete. During periods of rapid change, when norms are fluid and institutions are unstable, there are fewer constraints, more opportunities, and perhaps greater incentive to engage in illicit activities. (And on the economic side, the acts of privatization and deregulation are themselves government-directed activities that may be susceptible to precisely the same sorts of corruption that pervade statist economies.) If the corruption spike associated with liberalization is primarily a transitional phenomenon, then we can stay confident that a combination of political and economic reform will, in the medium to long term, help get corruption under control. On this view, the big challenge is to make sure that citizens don’t lose faith in the reforms during the transitional period, when things seem to be getting worse rather than better.

A more pessimistic possibility is that our assumptions that political democracy and open markets tend to decrease corruption are not correct, at least not as a general matter. Perhaps the set of complex historical processes that produced certain polities with stable democratic political system, relatively open economies, and lower corruption levels cannot be straightforwardly replicated elsewhere. Maybe the fact that certain countries that have, over the course of decades or centuries, established stable democracies and market economies have lower levels of corruption tells us virtually nothing about the impact of (rapid) political or economic liberalization on corruption levels in other countries, with very different histories.  If that’s right, then although we may favor political and economic liberalization on other grounds, it would be a mistake to see these reforms as anticorruption measures.

Finally, let me suggest one more possibility, somewhere between the optimistic and pessimistic accounts. This may be closest to what I actually believe, though I don’t really have any solid evidence at the moment. It may be that rapid liberalizations create very high corruption risks that need to be managed, explicitly and carefully, throughout the transitional period (and beyond). That is, maybe it’s not just a matter of seeing the transitions through, and expecting positive anticorruption results if we give it enough time. Maybe political and economic liberalizations have the potential to reduce corruption, but also have the potential to produce endemic corruption in the “liberalized” systems, and it all depends on the specifics of what happens during the transitional period. If that’s right, then moments of rapid political or economic transition are particularly important moments for effective anticorruption interventions.