Political corruption is a massive barrier to economic development and good governance. International institutions have become leaders in the effort to combat the problem. A growing number of such institutions have crafted official anticorruption rules, procedures and policies designed to deter the abuse of power within their membership and within institutional practice. Despite these regulatory developments, little is known about the role these institutions play in influencing corruption or whether the growing set of governance rules now in place have any effect.
Drawing upon our previous work in the social sciences, which is based on the statistical analysis of historical (1986–2013) data for over 190 countries participating in hundreds of international institutions, our contribution to this symposium explores whether international efforts to codify anticorruption norms into formal rules make a difference.1 Our central message is that the composition of the membership often matters more than the content of the rules. Depending on the makeup of the institutions, some clearly tolerate the abuse of power by their membership and, there, the rules designed on paper to deter corruption amount largely to cheap talk. By contrast, for those international institutions composed of less-corrupt members, such mandates may indeed produce results because the instituitons’ members divert loans away from corrupt recipients.
Anticorruption Standards and the Good Governance Movement
When thinking about how to solve an entrenched political problem like corruption, we must start with the causes. Three ingredients fuel corruption: discretionary power over resources, the ability to control and disperse “capturable” rents, and perceptions of impunity.2 In short, corruption thrives on the ability of politicians to engage in illicit behavior and their chance of getting away with it. The good governance movement, and the international anticorruption regimes it has inspired, operates to affect these causes by raising awareness of the problem and attempting to generate tangible consequences for those embroiled.3
Since the 1990s, a number of international institutions have adopted formal anticorruption mandates. These regimes are both pervasive and also quite diverse in who they target and how they operate. One model is the 1997 Organisation for Economic Co-operation and Development Anti-Bribery Convention, which requires states parties to criminalize the bribery of foreign public officials involved in international business transactions. Another model is the 2006 African Union Convention on Preventing and Combating Corruption, which addresses corruption in both the public and private sectors.
While international institutions of many stripes and functions have taken up these standards, they have become almost ubiquitous in the field of international development. In 1996, then-World Bank president James Wolfensohn declared that, by giving money to corrupt places, international development organizations (IDOs) enable bad practices.4 In response, multilateral aid-giving organizations took up the mission, crafting various standards designed to identify and deter the abuse of power, both within the donor organizations and among their recipients. In 2002, for example, the Monterrey Consensus became the first UN framework to embrace the fight against corruption specifically in the field of international development. Today, many IDOs have put related policies into place that should, in principle, generate more selective aid allocation, moving aid toward recipients with better governance.
At least on paper, the picture looks promising. A wide swath of international norms, rules, and procedures are broadly in place to combat corruption by creating expectations and definitions for appropriate and inappropriate behavior and the means to detect, and sometimes adjudicate, punishable offenses. Despite this uptake, however, not all rules are equally effective. Their effectiveness depends on the composition of the membership.
Effectiveness Depends on Membership Composition
Our central point is that the extent to which international anticorruption rules are likely to have a deterrent effect on the spread of corruption often depends more on the membership composition of an international institution than on its formally stated commitment to anticorruption norms.
Let's start with an important fact: while it may at first glance seem that only international institutions with less corrupt membership would adopt good governance standards for member participation, such as the common acquis of the European Union, this is not the case in practice. To the contrary, many international institutions composed of corrupt states parties readily take on anticorruption rhetoric. The Southern African Development Community (SADC) is an organization whose membership is characterized by pervasive levels of corruption,5 yet it adopted a formal protocol in 2001 that explicitly claims to seek regulatory cooperation in matters of corruption among member states. One of the stated purposes of the Protocol is “the prevention, detection and prosecution of corruption in the public and private sector.”6 On paper, the Protocol is extremely impressive and includes mechanisms to promote public education and awareness of the problem as well as to enforce good governance standards of conduct within some of the world's most corrupt countries.
Despite these rules, the SADC membership—and its headquarters—are riddled with allegations of corruption. Many charges of misconduct are levelled, but few lead to prosecutions or tangible consequences for the accused.7 Most states parties have shown little to no improvement in their alleged corruption levels since the adoption of the Protocol; some even have shown an increase in corruption in their global rankings.8 The SADC Lawyer's Association, which strives to promote good governance in the region, describes the Protocol's failures as driven not only a lack of real independence among anticorruption bodies but also by a “lack of true political will among the leadership in all SADC countries to address the problems that exist.”9 To date, there appears to be little cost to the membership for having adopted an anticorruption regime that allows corrupt politicians to remain in power with near impunity.
It is one of the central tenets of international relations that international institutions, when they work, can provide a powerful source for monitoring member behavior in accordance with the rules and expectations of the institution. That source, in turn, should increase the likelihood of detecting rule violators and create incentives for compliance.10 But that is “in principle.” International institutions with a highly corrupt membership are not genuinely motivated to detect or monitor corruption problems. Corrupt states and leaders have no reason to actually scrutinize—or allow independent citizens or organizations to scrutinize—corrupt behavior in the membership, which would only draw attention to their own misdeeds. Even when they do allow some degree of scrutiny, they have little reason to react to allegations made, and every reason to turn a blind eye. Current politics inside SADC countries are a good illustration: even though the organization has agreed on common anticorruption rules, there is scant enforcement of the rules and persistent corruption. No one is punishing these states for breaking their own rules, and they certainly are not punishing themselves.
Another illustration is the African Union's Convention on Preventing and Combating Corruption described above. There are clear rules in place, and yet recent estimates suggest that almost seventy-five million people in Sub-Saharan Africa alone paid bribes in 2015 to buy off police or judges or buy access to basic services.11 The organization has not demonstrated much willingness to contribute to a change in this trend. To the contrary, it has routinely ignored corruption scandals among its prominent membership, such as the many ongoing accusations against Jacob Zuma, former President of South Africa.12 And it has gone so far as to formally refuse to enforce the International Criminal Court's arrest warrants for war crimes against the highly corrupted government of Sudan.
One should not be tempted to wave this aside as an African problem. This tendency to turn a blind eye to corruption despite the rules extends far beyond Africa. It exists everywhere corruption exists. For example, the charter of the Organization of American States (OAS) formally advocates for a broad range of good governance principles, including the effective exercise of representative democracy, the elimination of extreme poverty, and the promotion of social justice. Their first legal rules specifically against corruption emerged in 1996, followed by a formal mechanism in 2001 to evaluate their fulfillment. Yet many members of the OAS ignore the Charter's core purpose, limiting its enforcement by and large to extreme and rare situations like political coups.13 And while more Latin American states are adopting laws or joining initiatives to reduce corruption on paper, massive corruption schemes involving powerful elites remain prevalent and punishment scarce.14
These examples are not exceptions to the rule. According to our statistical analysis of over 190 countries, and their membership in 317 active regional organizations between 1985 and 2013, they represent a general trend. By evaluating the relationship between a country's exposure to corrupted international institutions—with and without anticorruption mandates—and their future levels of corruption (measured on a scale by the International Country Risk Guide), we come to a stark finding. Persuading corrupt states to take up the rhetoric of the good governance movement has proven relatively easy, because there is little cost in doing so and little punishment for enforcement failures. Getting them to internalize and actualize those commitments against themselves or others is another story. In the same way that policing the police only works if the body overseeing the process is independent and judicious, cycles of corruption propagate despite the rules. In fact, estimating time-series cross-section models with interaction effects, we find that anticorruption mandates matter only in international institutions that are composed of relatively noncorrupt membership. They do not have an independent effect in international institutions composed of corrupt membership.
The Case of Foreign Aid
A great example of where these dynamics play out is within IDOs. It is here where the call for anticorruption measures remains the loudest. Of the IDOs for which we have information, over one-third have implemented formal anticorruption mandates since Wolfensohn's famous 1996 speech. But despite the many promises made by IDOs like the World Bank or the Asian Development Bank to redirect aid away from corrupt places, not all IDOs are equally likely to keep their promises.
Our measurement choices approximate the decision-making process within most IDOs. Formally, votes are weighted by donor contributions, which are highly correlated with their national incomes. Informally, donors usually operate by consensus, where the outcomes reflect compromise, though governments with greater structural power carry greater weight in the decision-making process. Since aid allocation decisions are shaped by powerful donors and a norm of consensus, we use the ten wealthiest countries in each organization as a baseline. We weight the measure by differences in structural power within the group of ten.
The good news is that IDOs composed of donor states with lower levels of corruption are visibly more willing to channel resources away from corrupt recipients toward recipient states with a better track record. This is in part because these donors value the good governance movement's objectives to dampen corruption among recipients and in part because they face real threats of voter backlash for sending scarce resources to countries where aid is siphoned off. In the United States, for example, the majority of Americans strongly oppose funneling aid to corrupt places.15 Similar trends define the European mood on the issue.16 However, it is especially noteworthy that IDOs composed of a less corrupt membership act this way whether or not anticorruption rules are in place. For them, these rules are more a reflection of what they already consider to be best practice, and less a genuine constraint on their behavior.
Unfortunately, our empirical results demonstrate that the flip side is also true. IDOs composed of corrupt donor states are more willing to turn a blind eye and provide loans or other financial assistance to corrupt borrower states, for a couple of reasons. First, since the donor states that make up these IDOs are themselves corrupt, they are unlikely to deter aid to corrupt recipients because good governance is not a concept they value, embrace, or seek to spread. They may very well adopt anticorruption rules and polices—these are especially easy to adopt within development organizations because they are unenforceable outside the donor state base. Yet the member states of these institutions are prone to turning a blind eye to the issue when it comes to aid allocation decisions because they too undertake unscrupulous political behavior.
Second, corrupt donor states are not as likely to face pressure from their voters or citizens over how or where they allocate foreign aid. In many cases, a corrupt donor state is unlikely to represent or reflect the general will of its domestic population when deciding how to allocate foreign aid resources.17 In these places, even if the average citizen had an opinion about her government's foreign aid policy, her opinion would be unlikely to generate real consequences on the government for funneling aid to corrupt places.
IDOs thus are a good example of the more general trend. Like other international institutions, IDOs can—and many do—generate rules and procedures on paper to deter donor states from fueling the “cancer of corruption.”18 The problem is that most corrupt donors simply ignore those rules when they care to since they do not embrace norms of good governance. In fact, our statistical research has shown that corrupted IDOs are as likely to adopt anticorruption rules as more honest donor organizations and at the same time more likely to give more money to more corrupt places.19
Implications and Recommendations
The composition of membership often matters more than the rules for determining how well international efforts to combat corruption actually work. In effect, donors rule: the effects of the rules are not independent from the membership.
To be clear, we are not suggesting that anticorruption rules are harmful or always ineffective. As political scientists working from global historical data, we analyze for big trends. There are certainly instances where an anticorruption regime has had a positive effect on—or at least relationship to—a reduction in corruption. While we cannot prove a causal effect of the law, our research shows quite clearly that states participating in less corrupt international institutions with anticorruption mandates experience a reduction in national corruption over time. And there are even a few examples of this phenomenon in our data from the most corrupt of places. The membership of the Islamic Development Bank, for example, is highly corrupt, but it has consistently avoided providing financial aid to corrupt countries. But the historical trends are quite clear that these are the exceptions, and not the rule.
The implication is not to give up on the global governance movement. Far from it. What our research does imply is that generating additional anticorruption regimes that are regulated mainly by corrupt state parties is unlikely to generate real progress. They simply will not follow the rules. In this instance, the best way to police the police is to ensure agreements that warrant participation by more independent, honest contributors. It follows that creating more “top down” global regulatory efforts to combat the problem may not be the most fruitful approach. This does not mean that the West tells the rest what to do—there is a cost to preaching global norms that fall on deaf local ears. But the future of anticorruption efforts looks grim if they continue to empower the corrupt as self-arbitrators.