Should aid fight corruption? New book questions logic behind this week’s anti-corruption summit

Over at the Center for Global Development, Charles Kenny wants comments on the draft of his book on Aid and Charles Kenny portraitCorruption (deadline end of May). Let’s hope this becomes standard practice – it worked brilliantly for me on How Change Happens – more varied voices can chip in good new ideas, spot mistakes or contradictions, and it all helps get a buzz going ahead of publication.

But let me take it one step further. As a contribution to Thursday’s corruption summit, hosted by David Cameron, I thought I would summarize/review the book. Charles gave the green light, provided I stress the ‘preliminary, drafty, subject-to-revisiony nature of the text’. Done.

The summit is about a lot more than aid – for example the rich countries putting their houses in order on tax havens. Which is just as well, because the book poses some real challenges to the whole ‘anti-corruption’ narrative on aid. What’s more, it is erudite, engagingly written and upbeat – as you’d expect given Charles’ optimistic previous takes like Getting Better. He’s got a great eye for telling research and ‘man bites dog’ surprise findings. Example: ‘Taking a cross section of countries and comparing current income (2010) to corruption perceptions in 2002 and income in 2002, results suggests more corrupt countries in 2002 have higher incomes in 2010.’

His core argument is pretty striking – when it comes to aid and corruption, corruption does indeed matter, but the cure is often worse than the disease: ‘an important and justified focus on corruption as a barrier to development progress has led to policy and institutional change in donor agencies that is damaging the potential for aid to deliver development.’ Ouch.

anti corruption summit logoNot only that, but ‘there is a second ‘problem of corruption’ in development – and that is one of perception and response. A narrative that suggests weak governance is the major problem of development and an intractable one justifies aid fatigue (it’s broke, and we can’t fix it). Concern with weak governance and the risk of malfeasance is the primary justification for donor programs that involve heavy oversight or even direct selection, design and management of projects from distant donor capitals. Perhaps more importantly, countries perceived as corrupt also attract less foreign investment and trade.

Were corruption an insurmountable stumbling block to delivering development and if we knew where was particularly corrupt, this second problem of corruption would not be a bug, it would be a feature. Tight control (if it worked) and aid fatigue (if it didn’t) would be a logical response. But the evidence that weak governance is a barrier to all development progress or effective aid programs just doesn’t stack up. And we know considerably less than is usually thought about which countries (or sectors, or activities) are weakly governed or particularly corrupt.’

Kenny argues that the focus on corruption feeds a generalized distrust of the state, but ‘distrust is a big problem because no country has become wealthy without a large government – one involved in a huge range of regulatory, investment and spending roles. By pandering to those who distrust government in the West, donors are helping to hobble governments in the developing world, with dire consequences for aid levels and effectiveness as well as broader development progress.’

And the double standards don’t end there, a focus on bribery (generally more prevalent in poor countries – see chart)

What if it was tax evasion or spending on lobbying?
What wd this look like if it was tax evasion or spending on lobbying?

ignores the other kinds of corruption preferred in rich countries, from tax evasion to lobbying and political capture and ‘the general corrupting influence of money in the system’.

In the aid business, the obsessive focus on corruption has generated a ‘web of oversight and control institutions that ensure staff spend most of their time ensuring oversight requirements are met’. That means centralized approaches that ‘contradict the internationally agreed Paris Principles on aid effectiveness that aid which is ‘owned’ by recipient countries and runs through government budgets is likely to be more effective.’

Kenny’s suggestion for what to do?

‘It is time for a fundamental rethink of anti-corruption approaches in donor agencies. Rather than trying to measure the dimensions of the black box of corruption and change its internal dynamics, this book suggests shrinking the box through approaches that minimize the impact of corruption on aid delivering results. It also suggests using nonaid approaches to support systemic institutional reform.

There is a role for donor agencies in governance and anticorruption efforts, but, first, corruption is only one of many barriers to development, and, second, the role of outsiders in the process is limited, context specific and based on many unknowns.’

PbR theory of change
PbR theory of change

My main comment on reading the draft was that he dismisses too lightly the political economy arguments for donors obsessing about corruption (he just discounts them as ‘not convincing’). I think he needs to come up with a serious and plausible theory of change for how donors might change their ways. Just saying ‘build the evidence and they will come’ is a classic researcher’s cop-out that seldom works.

I was also a bit skeptical about the proposed solution, which is switching from corruption to payment by results. Is every CGD author now contractually obliged to say that, whatever the problem, the answer is PbR?………

Anyone else read it?

Subscribe to our Newsletter

You can unsubscribe at any time by clicking the link in the footer of our emails. For information about our privacy practices, please see our .

We use MailChimp as our marketing platform. By subscribing, you acknowledge that your information will be transferred to MailChimp for processing. Learn more about MailChimp's privacy practices here.


6 Responses to “Should aid fight corruption? New book questions logic behind this week’s anti-corruption summit”
  1. Alice Evans

    Nice summary, Duncan. I likewise really enjoyed the manuscript. So engaging and well-written.

    I share your concern about an exclusive focus on financial incentives for results. The manuscript seems to assume that financial incentives are donors’ only tool. Maybe there are other ways by which they might try to motivate improved outcomes?

    One possibility (for developing country governments and donors alike) is to shift norm perceptions. That is, individuals’ beliefs about what others think and do. The belief that corruption is widespread can reinforce the practice. How might this be addressed? Perhaps donors/ countries could amplify exposure to positive deviants/ islands of effectiveness – be this ministries/ sections/ geographical regions/ of Kenya, for example. Such exposure might shift norm perceptions about what is possible and catalyse confidence in the possibility of social change.

    This would retain the focus on results, but expand it beyond financial incentives.

    People’s reasons for acting go beyond cash.

  2. Paul

    Sitting in a hospital setting in Malawi it looks a bit different. There are daily stories of corruption in the newspapers – many around the health sector and hospitals in particular (eg There can be little doubt that this corruption leads directly to the drug shortages that are ever present in the hospitals here. In Malawi it’s the Malawians who are concerned about corruption.

    ‘An important and justified focus on corruption as a barrier to development progress has led to policy and institutional change in donor agencies that is damaging the potential for aid to deliver development.’ In Malawi I think it’s possible to turn this logic on its head – in that donor agencies historically giving budget support in a country with inadequate systems for control exacerbated the corruption which has hindered development progress and democratic accountability.

  3. Thanks, Duncan, for the kind words, nice summary and helpful critiques.

    On the political economy of obsessing about corruption… point taken. To some extent I’m actually less worried about the obsession over corruption than the toxic form that obsession has taken (towards a narrative of failure and a justification of ring-fencing). And in that regard I think this week’s corruption conference is a really helpful thing. It is *not* primarily about telling developing countries what to do as if the problem is a problem of poor countries for which we have the solution. Instead it looks to be largely about what we should all be doing together –publishing contracts and budget data, beneficial ownership registries, and so on. I hope that framing spreads to the aid discussion, too: here’s a shared challenge to returns on aid investments. What can we do to reduce corruption that will increase returns on aid investments…

    On the results focus, I admit I’m a convert to PbR, but I do think (and should make clearer in the text) the issue is broader than aid payment timing and conditions. It is just realizing and acting on the fact that if you only track receipts and ignore outcomes you’ll do nothing to reduce corruption, indeed you’ll probably move it into a more damaging kind. Despite that, lots of donors seem to be sacking engineers and public service delivery professionals to replace them economists and accountants. And economists aren’t usually best placed to judge investment project outcomes (something I have learned from personal experience…).

  4. Thanks for summarizing this work, it is on my to-read list. I have a few thoughts to contribute to the discussion. For the past 2 years we have been piloting new means to analyze corruption in justice systems using systems approaches with the intent to develop an analytic means that would generate more effective anti-corruption programs. We totally agree that the current donor approach which either focuses so much on preventing fraud of their resources or is entirely void of any political reality using a principle-agent model ignores so many of the key enablers of corruption in a context. We have also struggled with applying political economy analysis in this regard, however do firmly believe that analyzing power is critical to thinking through anti-corruption programming. We recently posted a blog on this exact issue:

    Further, if I think of the realm of corruption and access to justice in fragile states and the idea of PbR, it makes me wince. I am a professional evaluator, so I am firmly on the results matter train, however if one was to set the standard that better access to justice was required in order to be paid, then this would be grossly flawed. Our research in DRC and Uganda indicate that there is a complicated web of incompetence (some caused by nepotism/cronyism), lack of resources (some caused by fraud), waste and inefficiencies that all drive an ineffective justice system. So in theory one could be quite effective at minimizing corrupt transactions, but not change the overarching access to justice. Conversely focusing on diminished corruption as the result is also difficult due to the measurement challenges. If one tried to use the common default of perceptions of corruption, this too is very problematic as our research shows local populations conflate many of the inefficiencies under one label ‘corruption’. This in addition to the standard challenges of measurement by perception. I am not fluent enough in the PbR debate to make a firm statement, but it seems that the approach would need to be very nuanced to be accurate. And yet, without results whats the point?

  5. Anders Östman

    The fish rots from head down. If the highest echelons of society are allowing corrupt practices in general to go on and there is no sincere actions taken against corrupt officials, donor oversight and all supposedly elaborate control mechanism do not change anything. So called petty corruption will only eliminated or significantly reduced when grand corruption is tackled. Good governance and the rule of law are not the the most essential ingrediences for rapid and poverty reduction driven development. More important is the policies for development that are being implemented. If a country do not show that it has pro -poor, inclusive and pro-rural strategies for development large amounts of aid, whether budget support or more donor controlled project aid should be discontinued. Donors should in such cases provide technical assistance and crucial institutional support, preferably at local level.

  6. Allan Moolman

    It would be great if for once someone turned the narrative on its head, named the corruptors and quantified the benefits they have accrued . It is not the much maligned governments that are offering the kickback deals and paying ‘brokerage’ fees to intermediaries – they are engaged with counterparts.Corruption is a two way process and we very rarely hear of investigations, let alone prosecutions against the companies undertaking the ‘bribing’ (especially if they are large multinationals).
    And before I am misunderstood – I am not asking for ‘corruption’ to be understood and accepted because ‘we are only continuing the practices of the West’ – I am simply pointing out that people in developing countries see the impacts of corruption and want to change it, but are also very conscious of the inequities in prosecution and the strengthening of the narrative that ‘that is the way things are done there’.
    Anti corruption initiatives still have to overcome the age old problem that development has not been able to address – the perceived double standard of one set of rules for the West and another for developing countries. Take for example the practice of brokerage/introduction fees – in many cases deemed a corrupt activity in developing countries, but standard business practice in the West.
    It would be interesting to understand which Western governments have failed to act against companies registered in their countries who have engaged in transactions in developing countries that have been deemed illegal or corrupt. It would also be interesting to see how much of the tax revenue collected in developed countries, comes from transactions or relationships that could be deemed corrupt. Revenue from extractives deals in the DRC or maybe all of the multinationals involved the South Africa arms deal might be a good place to start?