The following is a guest post from Fredrik Eriksson, an independent anti-corruption expert. From 2009-2011, he served as Senior Advisor in Anticorruption and Illicit Financial Flows at the Norwegian Agency for Development Cooperation (NORAD).


Fighting aid-related corruption is no simple task. Donor agency officials often perceive measures to detect and prevent corruption in aid projects as costly and distracting. However, left unchecked, corruption can undermine the political legitimacy of overseas aid activities. In this post, I summarize several of the key lessons I have learned as a staff member of (and later as anti-corruption consultant to) bilateral and multilateral aid agencies.

Lesson #1: Educate Senior Management and Change the Way Success is Measured

Much has been written about the shortcomings of detection approaches, such as financial audits and forensic accounting methodologies. The same can be said about prevention approaches that seek to enhance financial management practices and strengthen the capacities of counterpart country institutions.

A relatively new trend among development aid agencies is the adoption of corruption risk management approaches. For example, Norwegian law requires that those responsible for managing public funds also effectively manage the associated risks. Corruption is a major risk to sound financial management. However, new attempts to introduce corruption risk management approaches and  'mainstreaming' of anti-corruption considerations are surprisingly difficult to implement due to institutional interests (e.g. limiting administrative costs, domestic political reporting requirements), organizational incentives to push money out the door, and pre-existing donor agency staff beliefs and values.

When the success of a donor agency's senior management team is measured according to the speed at which it can spend approved budgets, corruption risk management is often perceived as an obstacle to achieving results. However, when senior management is assessed vis-a-vis independent indicators of how corruption risk management practices are applied, this obstacle can disappear very quickly.

Lesson #2: Tailor Your Approach to the Local Context

Given the many steps required for efficient aid delivery, making the prevention of corruption a priority is bound to cause frustration. This is not to say that donor agency staff are ignorant of the risks posed by corruption in general. Rather, there is little understanding of why corruption must be understood in the particular historical and cultural context of a given recipient country.

To date, there is relatively little empirical evidence identifying the specific anti-corruption interventions that most effectively reduce or control corruption. It is equally difficult to empirically compare the effectiveness of anti-corruption interventions across cultural and political contexts. Consequently, anti-corruption professionals are not able to offer donor agency staff many 'hard and fast' solutions. This need to understand local context often generates resistance to anti-corruption measures among donor agency staff who must juggle a variety of competing demands.

Consider an example from my own attempt to interest a particular donor agency department in the WHO’s program for good governance of health systems and medicines. Without understanding the broader governance context and the potential for long-term negative consequences, one staff member indicated that petty corruption in the form of doctors seeking a ‘top up’ of their salaries was justified given low salaries. If this type of petty corruption were eliminated, he reasoned, the agency would risk losing the cooperation of pivotal actors (doctors) who are needed to bring about urgent improvements to national health systems. Another staff member indicated that focusing on corruption risk management or mainstreaming approaches would detract from the more pressing issues of strengthening health systems and providing access to medicines.

Lesson #3: Accepting Corruption in the Short Term Reduces Aid Benefits in the Long Term

What these staff expressed was the problem of opportunity costs. In order to address urgent issues in the short- and medium-term, certain unwanted costs—like corruption—may be unavoidable. In their view, while it may be possible to reduce corruption in a certain project or program, corruption will still be the societal norm rather than the exception. Thus, corruption ought to be addressed as a long-term issue to achieve a sustainable solution. It should not be adding costs to aid while yielding no tangible results, when there are more pressing and immediately measurable issues such as vaccinations and surgical operations.

The problem with this reasoning is that it ignores two key problems. First, research and experience demonstrate that corruption disproportionately impacts the poorest and most vulnerable members of society. Therefore, when corruption is left unaddressed in a particular aid project or program, serious questions about long-term impact and sustainability often arise. One can use aid to pay for critical medicines and health supplies, but if poor patients cannot afford to visit the doctor because of the cost of "informal payments" (tacitly endorsed by donor agencies), serious questions must be asked about the real impact of aid.

Second, aid-related corruption directly impinges upon domestic political support for overseas aid expenditure, especially in donor countries where aid is already perceived to be ineffective or over-budgeted.

Unfortunately, there is not much donor governments can do to “punish” flagrant corruption and fraud in recipient countries. In 2009, officials within the Zambian government embezzled nearly $2 million of donor support to the Ministry of Health. But since bilateral donors requested separate accounts for their earmarked funds, the fraud was concealed by shifting the funds between various accounts. The lack of coordination between donors, mixed with the inherent fungibility of budget support, enabled the conditions for such a scheme, and diminished the impact of collective anti-corruption sanctions.

To maximize aid effectiveness, corruption prevention and sanctioning strategies must be conceived for the particular context of each country; international donors must also coordinate their anti-corruption strategies more effectively. But this will not happen unless rooting out corruption is a priority for all donors.

This post first appeared on The Monkey Cage and is re-published here.


Is foreign aid effective? Much research on this question has focused on quantitative measures of large-scale outcomes like a country’s economic growth or level of democracy. These studies have come to mixed conclusions, and debate still rages over whether aid can effectively promote economic development, human rights, democracy, or countless other outcomes.

Prominent critics of aid, such as William Easterly and Dambisa Moyo, recite many of these studies’ findings about aid ineffectiveness. Easterly and others attribute a good share of aid’s failings to the lack of feedback and accountability. As Easterly (2006:17) says, “The needs of the poor don’t get met because the poor have little political power with which to make their needs known and they cannot hold anyone accountable to meet those needs.”

We strongly agree with the point that feedback is a major problem for foreign aid. But criticisms of aid seem to assume that, in the absence of foreign funds, domestic governments would do a good—or even better—job helping the poor. However, we know that even in rich democratic countries the poor have a very hard time getting their voice heard by their own governments (Bartels 2008; Gilens 2012). The broken feedback loop motivated us to ask people in recipient countries what they thought of aid. In particular, we think the implicit assumption that governments in recipient countries are more attentive to the needs of the poor may be questionable.

We wanted to look at the question of aid effectiveness from a different angle.  What do the citizens in recipient countries think of aid? Do they perceive it to be useful and desirable? Few, if any, systematic studies of citizens’ views of aid in recipient countries have been conducted. Foreign donors sometimes collect such information, but do not release it for public consumption or scrutiny. To explore citizens’ reactions to aid, this summer we conducted a survey and field experiment on a nationally representative sample of Ugandans, as well as on a sample of local village council leaders, provincial governors, and members of parliament. We now have the data from roughly 3,600 Ugandan citizens, and our first report on these data is now available: http://ssrn.com/abstract=2134409.

We learned that Ugandans really like aid, and they want more of it. Uganda is a very poor country and it is heavily aid dependent. But even so, Ugandans by large majorities support additional foreign aid. More than 80% of respondents told us they wanted to see aid increased a lot as opposed to the one percent preferring that it be decreased substantially; 93% preferred at least some increase in aid versus 4% that preferred some decrease.

How do these results relate to aid effectiveness? Aid’s popularity seems to indicate that citizens in recipient countries see aid as beneficial.  However, nearly 80% of respondents also reported that they themselves have not directly benefited from aid, and nearly two-thirds of participants believed that more than half of aid dollars were not spent as intended.  How do we reconcile these beliefs with their strong desire for more aid?


It is critical to ask the question about aid effectiveness in a comparative way.  Is aid effective relative to other policies that could improve the quality of life for the public? This question of how aid compares to relevant alternatives is rarely asked. But it is key. We do not live in a perfect world; every policy has downsides and few may work at all. (Look at the long and intractable debates on macroeconomic policy where we have much evidence but no consensus.) So how does aid compare to other possible policies for serving the needs of Ugandans?

Our field experiment allowed us to address this question. We randomly assigned our nationally representative survey participants to receive information about identical projects sponsored by different foreign aid donors and compared them to a control group that was told about the projects without a donor specified, implying a domestic government initiative. This allowed us to compare foreign aid to government programs. We could also compare different donors to each other.

Data from our experiment clearly show that Ugandans significantly prefer foreign aid over government programs. Citizens are much more willing to pay personal costs by signing a petition or sending an SMS message to support aid projects than they are for government programs.  Our survey showed that citizens see aid as superior on many different dimensions. They view aid as less politicized, less corrupt, and more transparent than government programs. They also trust international organizations and foreign aid agencies more than all domestic government levels, including the popular president. And a large number of citizens support aid conditionality, where strict requirements must be met to receive future funds.

The study produced much evidence that Ugandans – when comparing the likely alternative of domestic programs – prefer foreign aid. Aid is not perfect, far from it, but it may be the best alternative for poor countries with weak domestic institutions and limited state capacity. And this may help explain why people strongly support aid, even if they do not benefit personally.

Ugandans had less intense preferences over which foreign donors were the best.  In general though, they preferred multilateral aid donors, such as the World Bank, to bilateral agencies, such as USAID. And this is despite the fact that bilateral donors bypass the government more in their aid projects relative to multilateral organizations. Again, citizens gave many of the same reasons as for their aid preferences in general: they saw multilateral assistance as more transparent and less politicized than bilateral aid.

A vigorous debate has raged for some time now over the effectiveness of foreign aid. Our study suggests that the views of citizens in recipient nations ought to be considered in the discussions.  Citizens in poor countries can provide vital insights about how well foreign aid works, especially compared to other feasible alternatives. Their views can inform our policy debates.  For instance, if large majorities in recipient countries want aid, as we find, should the donor countries reduce or end it? And if sizable numbers of citizens in recipient countries support aid conditionality, as our survey also reveals, should we end it, as some such as Easterly (2006) have proposed?


Helen Milner is the B. C. Forbes Professor of Politics and International Affairs at Princeton University and the director of the Niehaus Center for Globalization and Governance at Princeton's Woodrow Wilson School. Dr. Daniel Nielson is the Director of the Political Economy and Development Lab and Associate Professor of Political Science at Brigham Young University. Dr. Michael Findley is an Assistant Professor in the Department of Government at the University of Texas at Austin. 



Bartels, L. M. (2008). Unequal democracy : the political economy of the new gilded age. New York Princeton, Russell Sage Foundation; Princeton University Press.

Easterly, William. 2006. The White Man’s Burden. NY: Penguin Press.

Gilens, Martin. 2012. Affluence and Influence: Economic Inequality and Political Power in America. Princeton University Press and Russell Sage.

Moyo, Dambisa. (2009). Dead aid : why aid is not working and how there is a better way for Africa. New York, Farrar, Straus and Giroux.

Bartels, L. M. (2008). Unequal democracy : the political economy of the new gilded age. New York Princeton, Russell Sage Foundation; Princeton University Press.

Easterly, William. 2006. The White Man’s Burden. NY: Penguin Press.

Gilens, Martin. 2012. Affluence and Influence: Economic Inequality and Political Power in America. Princeton University Press and Russell Sage.

Moyo, Dambisa. (2009). Dead aid : why aid is not working and how there is a better way for Africa. New York, Farrar, Straus and Giroux.