Today’s guest post is from Robert Barrington, currently a professor of practice at the University of Sussex’s Centre for the Study of Corruption, who previously served as the executive director of Transparency International UK, where he worked for over a decade.

I have read with great interest the recent exchange of views between Professor Bo Rothstein and Professor Matthew Stephenson on the academic study of corruption and anticorruption. As an anticorruption practitioner who now works within an academic research center, I was particularly struck by how their exchange (Professor Rothstein’s initial post, Professor Stephenson’s critique, and Professor Rothstein’s reply) surfaced some extremely important issues for anticorruption scholarship, its purposes, and its relationship to anticorruption practice.

I find it hard to agree with Professor Rothstein’s analysis, but this is before even looking at his points of difference with Professor Stephenson. My main beef with Professor Rothstein’s analysis is with his starting assumption of widespread failure. Like so many prominent scholars who study corruption, he proceeds from the premise that pretty much all of the anticorruption reform activity over the last generation has failed. He asserts that “[d]espite huge efforts from international development organizations, we have seen precious little success combating corruption,” that anticorruption reform efforts have been a “huge policy failure,” and sets out to explain “[w]hy …  so many anti-corruption programs [have] not delivered[.]” Professor Rothstein then offers three main answers, which Professor Stephenson criticizes.

In taking this downbeat view, Professor Rothstein is not alone. The scholarship of failure on this subject lists among its adherents many of the most prominent academic voices in the field. Professor Alina Mungiu-Pippidi has framed as a central question in corruption scholarship, “[W]hy do so many anticorruption reform initiatives fail?” Professor Michael Johnston asserts that “the results of anticorruption reform initiatives, with very few exceptions, have been unimpressive, or even downright counter-productive.” Professor Paul Heywood, notable for the nuance he generally brings to anticorruption analysis, asserts that there has been a “broad failure of anticorruption policies” in developing and developed countries alike. And many scholars proceed to reason backwards from that starting point of failure: If anticorruption reform efforts have been an across-the-board failure, it must be because anticorruption practitioners are doing things in the wrong way, which is because they are proceeding from an entirely wrongheaded set of premises. The principal problems identified by these scholars, perhaps not coincidentally, are those where academics might have a comparative advantage over practitioners: use of the wrong definition of corruption, use of the wrong social science framework to understand corruption, and (as Professor Rothstein puts it) locating corruption in the “wrong social spaces.”

That so many distinguished scholars have advanced something like this assessment makes me wary, as a practitioner, of offering a different view. But I do see things differently. In my view, both the initial assessment (that anticorruption reform efforts have been an across-the-board failure) and the diagnosis (that this failure is due to practitioners not embracing the right definitions and theories) are incorrect; they are more than a little unfair, and potentially harmful. I want to emphasize that different take should not be considered as an attack on eminent scholars, but a genuine effort to tease out why, when presented with the same evidence, some academics see failure, while many practitioners see success. Here goes:

Just to be clear, I do not believe, and I am not arguing, that the anticorruption movement has been an unmitigated success. And part of the perception of failure may be because some of the aspirational rhetoric deployed by anticorruption reformers sets out lofty goals—a “vision of a corruption-free world” or “zero tolerance of corruption.” Of course anticorruption reform efforts have fallen short of that vision. They have not, as Professor Rothstein puts it (quoting Francis Fukuyama) succeeded in transforming Somalia into Denmark. However, that’s not the right way to think about, or measure, success. To use an analogy, advocates for environmental conservation often speak of achieving “zero net loss of biodiversity” and “net zero greenhouse gas emissions,” but if meeting these aspirational targets was the sole measure of success, we would need to conclude that the environmental protection movement was an utter failure. In reality, that movement has achieved many important, if incremental, changes. So too with the anticorruption movement.

Now, even measured against the lesser standard of incremental change, nobody—not even the most optimistic anticorruption practitioner—would assert that the anticorruption movement has been a rousing success, or even that it has entirely lived up to expectations. But neither has this movement been the failure that so many scholars assert. On this point, perhaps I might gently suggest that academics may be observing events at a considerable distance of geography, time, and culture, whereas practitioners are monitoring change on the ground. Both perspectives can be valuable, and the academic perspective particularly so when it is based on years of extensive fieldwork on particular countries and/or rigorous program evaluations, and not simply broad-brush assertions. But practitioners have a valuable perspective to offer as well. In their on-the-ground work, they not only consider data and information that may be discounted by, or unavailable to, academics; they also often evaluate success or failure based on judgments, by experienced and well-informed people on the ground, about whether the lives of ordinary people are improving—sometimes, just the life of one person or family or village, not a whole region or country.

So what do anticorruption practitioners, particularly those who work in civil society, think about the anticorruption endeavors of the past three decades? Nobody thinks we’ve achieved anything close to a “corruption-free world,” even in parts of the world that perform well on standard indices. But two of the giants of the anticorruption movement, Transparency International (TI) and Global Witness, have copious information about successes and impact. Indeed, many of TI’s hundred-plus national chapters have documented important, if sometimes modest, incremental successes—even in countries that academics regularly point to as examples of consistent failure. Those with experience in government and international organizations have offered similarly nuanced but optimistic assessments. Ngozi Okonjo-Iweala, a former Nigerian finance minister who now heads the World Trade Organization, has written that although “success stories about how to systematically fight this corruption” are relatively uncommon in the broader discourse, in fact “I can tell you, from my experience, that it is indeed possible to fight corruption successfully with the right knowledge, patience, and commitment to transparency.” James Anderson, a Lead Governance Specialist at the World Bank, offers a similar overall assessment: ‘‘Corruption continues to be a serious impediment to progress and fairness in many countries but there are things that can be done about it and, yes, in many countries, this is making a difference.”

Self-evidently, assertions are not the same as proof—whether they are offered by academics, activists, or international development practitioners. And cynics might insist that practitioners have a self-interest in touting alleged anticorruption successes so as to convince more donors to pour money into the “anticorruption industry.” (Though in fairness, cynics might similarly accuse academics of over-hyping the importance of their contributions in order to justify getting more research funding.) Ultimately, working out which interventions have helped make progress, and why, requires close attention to, and rigorous assessment of, particular reforms, policies, actions, and activities. While there is not the time or space to delve into that evidence here, let me offer a few examples where I think that many academics have made overly broad assertions of failure, which has consequently led, in my view, to an inapposite diagnosis of the challenges and difficulties.

• First, a number of scholars have asserted that anticorruption reforms have failed because activists and reformers have been using the wrong definition (or, more grandiosely, the wrong “conceptualization”) of corruption, and are therefore measuring the wrong thing and aiming at the wrong target. With respect to definitions, Professor Rothstein, for example, writes off TI’s working definition of corruption (“the abuse of entrusted power for private gain”) as an “empty signifier.” As for measurement, there’s a whole cottage industry of academic scholarship devoted to criticizing TI’s Corruption Perceptions Index (CPI) and debunking its methodology. So why does TI continue to use this definition and publish this index? Are the people at TI just stupid or ignorant? Well, not really. For one thing, on the subject of methodology, a number of the critics seem to be unaware of post-2011 improvements in the CPI. But even if we put that aside, the people at TI are well aware of the imperfections of their corruption definition and the weaknesses and limitations of the CPI. Indeed, the TI experts know the methodological flaws as well as anyone else after two decades of critiques. But TI—both the Secretariat and the national chapters in countries around the word—also know that the annual appearance of the CPI is an opportunity for engaging governments to improve their anticorruption performance, with enough grounding in reality to be useful despite its well-known imperfections. The expectations of (some) academics about what anticorruption measurement should deliver are apparently very different from the expectations of practitioners about how an imperfect tool can be used most effectively. As for TI’s working definition of corruption, here I agree with Professor Stephenson that, while there are some genuine ambiguities and questions about exactly what’s in and what’s out, overall it provides a workable general definition. It is certainly sufficient for practitioners both to carry out their daily tasks and to build long-term strategies. Is it a precise standard that would allow one to sort all activities neatly and mechanically into “corrupt” and “non-corrupt”? No. But is it an “empty signifier”? I’m not entirely sure what that even means, but I don’t think so. The TI definition, when used in practice, has provided enough of a common understanding to make progress in addressing an important cluster of related issues. Devoting a lot of time and energy to reworking the definition would risk misallocating resources for little benefit. It’s frankly a bit puzzling to me how consistently academic commentators on corruption seem to imply that we can’t proceed to do meaningful anticorruption work unless we get the conceptualization right or come up with a near-perfect measurement system. But that is a recipe for paralysis.
• Next, consider the debate over anticorruption agencies (ACAs) and other specialized anticorruption units. Reading the academic literature, you would come away with the impression that, apart from Hong Kong and Singapore, they are basically useless, particularly in developing countries. But this is quite different from the perspective of anticorruption activists, who continue to advocate for the creation and strengthening of such agencies, in countries as diverse as Lebanon, Indonesia, Guatemala, Australia, and France, among many others. All these campaigners could be deluded. But their support for ACAs is, in many cases, borne of experience and practical observation. Have these agencies, where they have been created, eliminated corruption? No. Have some of them proved entirely ineffective? Certainly. But a blanket assertion that implies bodies like Indonesia’s KPK or South Africa’s Office of the Public Protector have made no difference whatsoever seems at odds with reality. Even specialized bodies that were eventually reined in or shut down, like CICIG in Guatemala or the Car Wash Task Force in Brazil, made enormous strides in holding the powerful accountable for their misdeeds, and inspired the public to see that the world could be different. Such agencies are not a panacea for corruption. But no serious anticorruption activist or reformer thinks that they are or could be. Academics regularly caution against “silver bullets” and “one-size-fits-all” solutions, but in pointing out that ACAs (or other specialized anticorruption bodies) are not a magic solution, academics aren’t telling practitioners anything they don’t already know. In the course of campaigning for a given institution or institutional reform, of course advocates talk up the importance of that initiative—after all, you need to sell it to the public and to those with the power to make it happen. But few if any sophisticated anticorruption practitioners are laboring under the misapprehension that this or that reform is a “silver bullet,” or that an approach that worked in one place will automatically work elsewhere.
• Finally, let me say a bit about a reform effort that’s particularly close to my heart: TI’s campaign for the UK Bribery Act. Professor Rothstein writes that corruption is “not a legal problem,” so presumably improving laws shouldn’t make much difference. He does not discuss the Bribery Act explicitly, but his position seems to imply this kind of landmark law is an irrelevance—an example of a reform based on the “failed” principal-agent theory. I confess that during the whole time I was working on the campaign for the Bribery Act, I never heard, let alone used, the term “principal-agent theory.” We thought about these issues in simpler terms: bribery is a bad thing, and the UK needed an effective law to prevent it. That said, we may well have been drawing on an intuitive understanding of principal-agent problems, both in how we organized our campaign and in how we tried to design the law. However, if “corruption is not a legal problem” and a principal-agent framework (or a focus on changing incentives more generally) is a dead end, then the effort to secure the Bribery Act was presumably wasted. But the Bribery Act passed. And while it did not eliminate bribery by UK firms or citizens, nobody—not even the most starry-eyed campaigner involved in the process—expected that it would. We don’t have a good way of measuring how much bribery has been prevented by the Bribery Act, but all available proxies—ranging from the prosecutions of high-profile companies to the substantial investment that many firms are making to improve their compliance systems and manage risk—point to a more positive picture than Profesor Rothstein’s analysis suggests should have been possible. The bottom line here is that the world is better off with the Bribery Act than without it, and that, in my view, looks more like success than failure.

You get the point by now. Too many academic commentators on corruption have overstated the case for failure. I don’t contest the fact that in this field, as in so many others (such as human rights and environmental protection), there have been too many failures and too little success. But the relentlessly negative tone of some of the academic discourse is unjustified when one looks more closely, and with reasonable expectations, at the actual empirical record. And this relentless negativity can have bad consequences. Tackling corruption on the front lines can be a dispiriting endeavor, and having the leading experts in the field constantly declaring that your efforts are in vain—and implying that this is because you’re ignorant or foolish—can be demoralizing. Worse, it can strengthen the hand of governments and other actors who do want the anticorruption movement to fail. Activists need allies, not stone-throwers.

Of course, academics have a job to do: they are researchers, analysts, iconoclasts, thought-provokers, and innovators. That is the privilege and success of academic freedom. And sometimes, the latest academic research can be important for practitioners to hear even though (or perhaps because) it may make them uncomfortable, calling their assumptions and preferred approaches into question. But with academic privilege comes responsibility—a responsibility to be careful, nuanced, and well-informed. To return to one of my earlier examples, what may for academics be an interesting intellectual exercise about the CPI’s methodology can be for anticorruption activists undermining one of the few successful tools they have in a dirty fight where the odds are heavily stacked against them. A careful critique of an index’s methodology, which leads to improvements in how it is constructed or how it is employed, is undoubtedly a good thing. Yet balance, based on full understanding and information, and an awareness of the possible consequences, should also be a part of such critiques. By all means trash anticorruption agencies or the CPI if they are all useless. But they aren’t. And using eye-catching words like “failure” seems like a careless use of a poorly-defined term—exactly the thing that many academics tend to criticize in others. Moreover, broad-brush assertions that the last generation’s anticorruption efforts have been a total failure, and the whole enterprise needs to be rethought from the ground up, may discourage useful, nuanced research into what has worked and why.

My point here is not to denigrate academics. Many scholars—including those who have asserted that the anticorruption movement has been a failure—have produced extremely useful research. There are some superb academic books, articles, and monographs on corruption, many of which are based on years of painstaking inquiry, that challenge established orthodoxies in helpful ways. And while some academics contribute exclusively through their writing, some have taken the time to join civil society boards or advisory panels. Civil society organizations and other practitioners should certainly do more to help bridge the gap by between scholarship and practice by educating themselves about the latest research and thinking. And there are number of very good programs, such as the UK’s Anti-Corruption Evidence Programme, the Natural Resource Governance Institute, and the U4 Anti-Corruption Resource Centre, that are helping to bridge the research-practice divide.

But still, a certain strain in academic writing about anticorruption efforts disturbs me. This strain overemphasizes and exaggerates failure. At times it also exhibits a certain doctrinaire conceptualism, and needless contrarianism, that I find unhelpful. I use Professor Rothstein’s blog to illustrate this, but his underlying assumptions are widely shared by other notable academics. My challenge to the world’s finest scholars on corruption is this: spend as much time as you have on analyzing failure on finding out what has worked, and why it worked. Practitioners need good, high-quality, challenging academic research to inform their efforts. The next generation of scholars and practitioners alike needs to make well-informed choices about their objectives and how to pursue them. Given the size of the challenge of corruption and the comparatively slender resources available to the anticorruption movement, having those resources more aligned might help to create the successes that practitioners yearn for and academics might want to write about.