Paul M. Heywood, the Sir Francis Hill Professor of European Politics at the University of Nottingman, contributes the following guest post:

In a recent post, Matthew recommended a speech by Robert Barrington, Executive Director of Transparency International UK, on the relationship between academics and advocates in the fight against corruption. I was very pleased to read the post, as Robert had given the speech at my invitation during the inaugural meeting of research projects funded under an exciting new initiative being jointly run by the British Academy and the Department for International Development (DFID). The Anti-Corruption Evidence (ACE) Programme, which I serve in the capacity of Academic Leader, is designed explicitly to address the interface between researchers and practitioners, with a fundamental focus on what actually works when it comes to fighting corruption.

Prompted in part by a highly critical report of DFID’s anticorruption approach by the Independent Commission for Aid Impact (ICAI) – itself reproached for poor use of evidence (including on this blog) – and in part by its own commissioned evidence papers into corruption (here and here), DFID has partnered with the British Academy to launch a £3.6m programme aimed at helping us understand better exactly how and why specific interventions succeed or fail in particular contexts. Some may wonder why we need yet more research on corruption; indeed, that is precisely the question I was recently asked by Jeremy Lefroy MP when giving evidence to the House of Commons Select Committee on International Development inquiry into tackling corruption overseas: “I would have thought there was plenty of evidence around. In what way do you think the evidence base needs to be strengthened, or is this just creating extra work for people who look at these things?”

The question is a reasonable one, especially given the exponential rise in the number of books and articles on corruption published over the last quarter century (as reflected in Matthew’s ever-expanding bibliography on the topic). There are many answers that could be given, but one key factor is that much of the existing research on corruption has simply been too generic to produce specific recommendations on which policymakers can act. Although it is widely recognized that corruption is not just one thing, such recognition has often not been translated into research design. Notably, many large-n studies have used an undifferentiated concept of corruption to serve as either a dependent or independent variable, seeking to explain a host of specific failings across a very wide canvas. Where there have been attempts to disaggregate corruption, these have often proposed bipartite, rather than graded, classifications (grand/petty, political/bureaucratic, need/greed, and so forth). In practice, corruption is a much more complex phenomenon than such dichotomous approaches can conceivably capture. Four observations follow from this:

• First, we should be much clearer when discussing corruption what specific type we mean. There are important differences, for instance, between kleptocracy, bribery, influence-peddling, and so forth—not just in how they operate, but also in which actors and what kinds of resource exchange are involved. Too often in corruption research, a rhetorical acknowledgement of these different types is followed by unreflective conflation of them in practice through reliance on aggregate indicators.
• Second, we should specify in what particular sector the corruption in question takes place and, therefore, which constellations of actors are involved – as well as any relevant interactions and dependencies that serve as enablers.
• Third, we need to be more aware of the level at which any specific type of corruption is operating, whether it be a complex transnational network allowing shell companies to launder huge sums of money, for instance, or a network of local law-enforcement agents taking advantage of their position to secure non-monetary favors.
• Fourth, we should look more closely at the role of culture in relation to values, attitudes, norms, roles, rituals, and framing mechanisms, and how they help shape particular manifestations of what behaviors are understood as corrupt in different contexts.

In designing the BA/DFID ACE Programme, we have tried to move beyond the kind of generic, broad brush approaches that have characterized much existing research; by the same token, there has been an explicit focus on engagement with practitioners. Not only were several members of the advocacy community involved in assessing the applications, but bidders were also required to outline exactly how they would interact with practitioners on the ground, most obviously the DFID country officers in the priority countries that are the principal focus of the Programme (Bangladesh, Ghana, Mozambique, Nigeria, Tanzania, Uganda, and Zambia).

The eight projects funded under the Programme will therefore seek to identify practical initiatives that can help developing countries tackle corruption and the negative impact it has on millions of people’s lives. The research teams’ work will range from analyzing big data from major aid agencies to better assess the risks of corruption in aid allocation, to the development of practical policy recommendations for the design of civil service systems, and a study into the role that informality plays in fueling corruption and stifling anti-corruption policies.

All the projects will have a strong focus on identifying how the impact of anti-corruption interventions will be measured, for instance in relation to scale, change over time, causality and attribution to reform. Similarly, a comparative element to understanding what works in different countries, and contexts, is an important feature of this scheme. Future blog posts will outline the projects in more detail.