New DfID Report: Few Donor-Supported Anticorruption Policies Effective

The United Kingdom’s Department for International Development released a new report February 25 summarizing the learning on corruption in developing nations and how to combat it.  Why Corruption Matters: Understanding Causes, Effects and How to Address Them was commissioned to help donor agency staff who advise on anticorruption policies and to assist in the design of programs to control corruption.  As its title advertises, the report examines three issues: the causes of corruption; its costs, both financial and non-financial; and what measures reduce it.  Those searching for what developing nations can do to fight corruption will turn immediately to chapter 5, “Anticorruption Measures,” which evaluates a variety of different efforts to control corruption from ratifying UNCAC to reforming customs and tax agencies to conducting public expenditure tracking surveys.

Readers looking for new steps developing countries can take to control corruption or confirmation that the standard approaches are working will be disappointed.  Few interventions have had any effect, and with one exception, the evidence showing these have had an impact is thin.

The report is the seventh in DfID’s evidence paper series, works intended to present “an authoritative and extensive assessment of the evidence in a given development field.”  Jointly authored by staff from the Overseas Development Institute and the U4 Anticorruption Resource Centre, it evaluates a dozen different anticorruption policies along two dimensions: 1) evidence of their effectiveness, taken from case studies and quantitative analyses and 2) the amount of evidence supporting the finding of effectiveness.  Public expenditure tracking surveys is the only one for which there is any more than a “small” amount of evidence showing it reduced corruption.  For four others a “small” amount shows effectiveness: i) central budget planning and management, ii) media, iii) procurement, and iv) supreme audit agencies.  For the remaining seven, the literature reviews shows two are ineffective: national anticorruption strategies and anticorruption agencies.  For the remaining five — revenue and customs, organized civil society, transparency and access to information, decentralization, and social accountability mechanisms, the picture is mixed.  For each there are studies showing it is effective and others showing it is not.

The authors examined three other reforms briefly: civil service reform, the adherence to UNCAC and the Extractive Industries Transparency Initiative, and police reform.  Given the sparse, and mixed, evidence of effectiveness, they were unable to draw any conclusions about their value.

The chapter summarizing the confusing welter of work on the causes of corruption concludes that “corruption dynamics are shaped by the interaction between political, social, and economic processes.”  Among the factors noted are “weak governance,” with corruption likely to be especially prevalent in neo-patrimonial systems characterized by blurred lines between the public and private sector, patron-client and identity-based politics or the “big-man” syndrome.

The research on the impact of corruption is reviewed in chapter 4.  With the exception of its effect on economic growth, where difficulties of measurement and the confounding influence of so many other factors make it impossible to reach a definitive, unqualified conclusion, the deleterious effects of corruption are clear.  The more corruption, the slower firms grow and the less foreign firms invest.  Corruption exacerbates inequality, reduces the quality and level of public services and the trust and legitimacy of government.  Corruption is bad for the environment, leading to higher levels of pollution, more deforestation, and other harms.  The relationship between the fragility of a state and corruption is more complex.  As Sarah Chayes documents in her new book on Afghanistan, corruption can increase the chances a state will fail; on the other hand, the report cites case studies of Nigeria and Sierra Leone showing that corruption has shored up these states.

The report also includes a short chapter reviewing claims that increasing women’s voice and their numbers in government leads to a reduction in the level of corruption. It finds that “existing evidence is insufficient to conclude that raising the percentage of women in government per se is likely to lower corruption levels.”  The recent literature, the authors say, does not suggest women are less prone to corruption than men; the relationship between gender and corruption is more a function of the social conditions in which corruption opportunities arise.

For time-pressed development staff, the most useful chapter is the one documenting the impact of corruption.  The results are clear and easily digestible.  The least useful is the one discussing the causes of corruption.  As the brief précis above shows, the synthesis of the learning is a series of obvious, banal statements unlikely to help in any dialogue with policymakers.  This is less the fault of the authors than of the terms of reference.  Asking that a complex area of research with conflicting findings about a topic as large as the causes of corruption be boiled down into a series of “headlines” was unlikely to yield anything more.

The authors caution as the outset that their findings should be read “in the context of the full report to gain complete understanding of why and how the conclusions have been drawn.”  This warning is particularly important for those trying to find effective interventions to curb corruption.  The details left aside in summarizing the findings on effectiveness, such as explanations of how context affected the impact or where the effect was produced by the interaction of two or more interventions introduced simultaneously, are critical for program designers.  Anyone designing an intervention or contesting the design of one on the basis of the summaries alone would be committing development malpractice. Those working backwards from the literature summaries to the original sources will find a gold mine of useful information.

2 thoughts on “New DfID Report: Few Donor-Supported Anticorruption Policies Effective

  1. Pingback: (In)effectiveness of anti-corruption measures | World Policing

  2. In terms of anticorruption agencies, their effectiveness may well be adversely affected by lack of independence, executive powers, and ability to mount their own prosecutions.

Leave a Reply to Simon Baker Cancel reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.