Press Releases   |    In the News   |    Op-Eds   |    Transcripts   |    Multimedia


Rethinking Foreign Aid

Paul Farmer on Five Ways to Improve Development Assistance

Foreign Affairs
December 12, 2013

(Click here to view the original article.)

So much is written and said about foreign aid that it has become difficult to contribute meaningfully to the debate about whether it is effective. But if we are charting our fates as citizens of a crowded, fragile planet, then any honest assessment must conclude that progress has been made, whether in terms of child survival or literacy or access to basic sanitation. Still, profound social disparities exist; so too does extreme poverty. And the prospects of those living on less than two dollars a day remain grim.

With the stakes as high as they are, the need to challenge the assumptions we make about aid is paramount. Myths and mystifications about aid persist. Whether we speak of feedback loops or best practices -- or, perhaps, simply better practices -- we have a long way to go.

Despite agreements on aid effectiveness reached in Rome, Paris, Accra, and Busan over the last decade, 80 percent of aid from major bilateral and multilateral donors to fragile countries still bypasses the systems of local public institutions. But the aspiration to improve the lives of those living in extreme poverty through better public health, public education, and public works by definition requires public-sector capacity.

By way of example, consider some data from Haiti after the 2010 earthquake. Based on United Nations estimates, bilateral and multilateral donors channeled $6.04 billion in humanitarian and recovery funding to Haiti from 2010 to 2012, but disbursed less than 10 percent of it directly to the Haitian government. Just 0.9 percent of immediate relief aid right after the earthquake (totaling $2.41 billion) made it directly to the Haitian government. Even the local NGOs and businesses were excluded: less than 0.6 percent of that $6.04 billion was invested in Haitian organizations and businesses. One of the top bilateral donors in Haiti awarded only 1.4 percent of its contracts to local companies.

What are the consequences of bypassing national and local institutions? Haiti is the setting for about half of the world’s recently reported cases of cholera, which is a striking example of what happens to a country lacking a functional public sector. Can we really imagine eliminating the disease without building municipal water systems? Given how little aid is delivered in a manner that could durably build local capacity in fragile settings and extreme poverty -- the raison d’être of official development assistance -- we see the trap in which we’re stuck. In order to free ourselves from it, we must begin by confronting the myths surrounding foreign aid.


The first myth is that foreign aid doesn’t work. But the aid enterprise has contributed to a number of achievements around the world, not the least of which is the distribution of public goods that have led to higher rates of child survival and reduced HIV infection, malaria, and other infectious diseases. These improvements have been registered across the globe, but certain specific examples are worth citing.

Nineteen years ago, Rwanda lay in ruins. Today, it is the only country in sub-Saharan Africa on track to meet all health-related UN Millennium Development Goals, among them reducing under-five mortality by two-thirds, between 1990 and 2015. Rwanda, in fact, has seen one of the most dramatic declines in premature mortality ever recorded. Although many economic measures of well-being miss important disparities, the trends in Rwanda are encouraging: since 1994, per capita income has almost tripled and GDP has quadrupled, growing at a rate of 8.1 percent annually from 2000. Research shows a link between progress towards key targets of the Millennium Development Goals and direct investment in government institutions. So it is no surprise that almost 56 percent of Rwanda’s official development assistance is disbursed through its own country systems and institutions.

The U.S. President's Emergency Plan for AIDS Relief (PEPFAR) and the Global Fund to Fight AIDS, Tuberculosis, and Malaria are the largest and loudest counterfactuals to the claim that foreign aid does not work, especially when adequate resources are brought to bear on pressing social problems. The fight against AIDS is in many ways a rebuke to pessimism and unambitious aspirations. Remarkable developments have occurred on three fronts: the scientific discovery and understanding of HIV; the development of tools to diagnose, stage, and treat HIV disease; and, astoundingly, the delivery of these advances to millions of the poorest and sickest people in the world.

This is not to say that PEPFAR and the Global Fund haven’t made missteps. But together they have changed the global conversation about AIDS. Some used to claim, for example, that it was impossible to treat a chronic infection with a multi-drug regimen in settings of poverty. But that was not any more true than the claim that treating infections alone would suffice. More funding for AIDS therapy and the rollout of ambitious national plans produced a great rush to enroll patients and integrate prevention with care. PEPFAR and the Global Fund promoted equity by providing the highest standard of care, which included the prompt diagnosis and treatment for both diseases.


The argument is often made that it is not cost-effective to provide tertiary health care services such as cancer treatment to those now struggling to leave poverty behind. How many times have we heard in places like Haiti and Rwanda that we shouldn’t think about, much less focus on, specialized health care when so few people have access to basic public health interventions such as vaccines? Thus, we are socialized for scarcity; thus, we set our sights and our aspirations lower.

April 15, 2013, was a tranquil day, at least for a massive sporting event like the Boston Marathon, until suddenly there were hundreds of injuries, many severe. And they all happened at once, as is always the case with bomb blasts and disasters less subject to human control, including earthquakes. The city’s teaching hospitals did a splendid job that day and in the days that followed: not a single patient who made it to a hospital died. In addition to heroism and compassion, there was a system in place in Boston’s hospitals, and it worked.

Shortly after the bombing, I was consulting with George Dyer, an orthopedist at the Brigham and Women’s Hospital in Boston. Knowing I was headed back to central Haiti, Dr. Dyer told me: “This was a small disaster, in the scheme of things. It made me think again how important it is to have clear plans in place for Haiti’s next big disaster and the role HUM will play in it.” HUM is the acronym for the recently built public teaching hospital in central Haiti, Mirebalais University Hospital. It was built to be a teaching hospital for Haitian doctors and nurses and administrators based on the hypothesis that the quality of medical care will improve whenever training and research occur in tandem with compassionate care.

In recent decades, such aspirations have sometimes been derided in public health, especially by those tasked with serving the poor. Some in the development world have dismissed teaching hospitals, such as those that provide care for AIDS or cancer, as not cost-effective, nor sustainable, nor a wise use of scarce resources. But the Mirebalais University Hospital is now up and running because a lot of builders and supporters in Haiti, Boston, and across the United States and other countries rejected these low-ball aspirations.


Others argue that governments in poor countries are too corrupt to work with. But the numbers regarding aid to some fragile states tell a reassuring story. The UN recently analyzed data from six donors on how much of their foreign aid is lost to corruption and fraud: Australia, Belgium, the European Commission, Denmark, the United Kingdom, and the United States. Together, they gave an estimated $47.8 billion in aid in 2011, which amounted to approximately 31.7 percent of all aid reported to the Organization for Economic Cooperation and Development. The quantity of aid that these donors detected either as being lost to fraud or as being an improper payment ranged from only 0.006 to 0.16 percent. The United Kingdom’s Department for International Development spent approximately $10.9 billion in aid in 2010–2011 and reported only 0.015 percent ($1.9 million) as lost due to fraud. The U.S. Agency for International Development actually has the highest rate of detected fraud among donors who report this data publicly. It also has the lowest percentage of funding going through government systems, at 0.8 percent.

Now, even before we dispute how such losses are detected and then reported and how reliable that data might be, it’s surprising how low these numbers are. These figures don’t include anything beyond “development assistance,” a slippery enough concept. But even if these numbers were off by a factor of ten, and included many projects of dubious value, moving tens of billions of dollars across such steep gradients of inequality and losing between 0.15 and 8 percent of such aid would probably compare favorably to most commercial transactions. And it doesn’t seem to make much difference if we move this money through so-called country systems, thereby strengthening them, or run all of these public funds through private NGOs and contractors. There’s still not the industrial-strength corruption we’ve seen in many companies, from Enron to several others peddling strange financial derivatives right up until the financial crisis of 2008 and beyond.

If we are able to strengthen in-country capacity so recipients can manage their own affairs, one day we will eliminate the need for anything other than partnerships. We can do this by partnering with discernment and building systems of transparency and accountability.


There is another myth in foreign aid: NGOs are the solution. But no health or educational intervention can be brought to scale without an effort to strengthen the public sector. Indeed, for those who believe in the aspirational notion of a right to health care or education, it’s best to note that such rights, like civil and political ones, are conferred by the state.

It’s not an accident that in Haiti, which has the most privatized educational system in the Western Hemisphere, less than half of all school-age children were in school on the eve of the earthquake. A proliferation of private citizens doing public good does not necessarily mean that NGOs and civil society groups should bypass national and local institutions in their effort to promote the public sector’s capacity to deliver basic goods such as primary health care and clean water.

Just as many people outside of development work are quick to say that development assistance never works, some enthusiasts are prone to romanticize or exaggerate the capabilities of those living in extreme poverty. I’ve seen a lot of this pathology, which assumes that our goal as citizens is to know everything under the sun, from which vaccines to give our children to how best to build bridges and roads. One of the most commonly encountered liberal pieties of development work comes in a million flavors: “Local solutions for local problems.” Well, it depends. Many problems originate outside of people’s own communities: most trade regimes, all epidemics, and just about anything to do with climate change. Should every community be manufacturing its own vaccines or pedagogic materials or shoes? Of course not. But school backpacks and fortified peanut butter should, whenever possible, be manufactured locally.

Mindful of the journalist Evgeny Morozov’s critique of the idea that we must bring every citizen and consumer up to speed on arcane and complex topics in order to solve them, all of us need to learn a lot more about how and when our systems work -- as health systems did in the Boston Marathon bombings -- and how and when they don’t, which is often. In that vein, we need a better understanding of how and when aid is effective, and when it’s not. We need a cultural shift that comes with changes in aid systems and in the rules that govern them.


As is the case with so many plans to fix the world, the major challenges lie in implementation. In Rwanda, ideas have escaped from the realm of abstract discussion. Plans are being carried out right now, transforming a so-called failed state into one where people are living increasingly longer lives, where institutions are stronger, and where human capabilities might flourish. To help other nations make similar progress to reduce extreme poverty, here are five recommendations.

First, reward aid institutions and staff who localize aid dollars. Staff members of bilateral and multilateral institutions are not usually promoted for increasing in-country procurement or investment. Advancement and retention could be made contingent on increasing the proportion of aid invested locally. To this end, program staff would report on such factors as the amount of goods and services procured locally, the proportion of funding channeled through national and local institutions, and resources allocated for local job creation, including payment of civil servant salaries. When aid institutions award grants or contracts, they should be evaluated using the same criteria.

Second, prioritize implementation with national counterparts at every step of the process. It requires marrying our policy ideals to our commitment to meet the challenges of implementation on the ground.

Third, reassess how we evaluate risk. The issue of corruption and money lost to fraud is an area that merits more study and better data. But many people still -- sometimes without much analysis -- write off developing countries as hopelessly corrupt and too dysfunctional to work with. But let’s not conflate weak systems with corruption. All too often we worry about the risks to ourselves and our institutions when we make decisions about how to invest in fragile settings. But those notions of risk are inverted. The risk of people living in poverty and dying needlessly should come first.

Fourth, challenge common assumptions about what is considered sustainable and cost-effective in fragile settings. These terms, although created with the best of intentions, have sometimes become a reason for policymakers to discount complex intervention. Rather than serving the poor, these development standards have at times become a blunt instrument used against them. I have seen it firsthand in Haiti and beyond. Why, for example, is investment in job training -- an exercise sometimes more akin to social fiction than actual professional development -- considered sustainable when it is not linked to jobs with living wages? Training is, of course, critical and needed, but it should be linked to real jobs if it is to have long-term impact.

Fifth, prioritize the transfer of aid functions to local authorities. Aid implementers should consider every decision they make within the context of a long-term plan to transfer their tasks and functions to local institutions. It is our hope that development staff will ask, before even beginning a new project, “How can we hand this over to the public sector when the time is right?” “How do we ensure the local authorities have the support they need to lead the program?” This is not meant to be an excuse for aid agencies to leave before their local counterparts are strengthened, but rather a challenge for them to stay as long as it takes until these systems can stand on their own, and possibly even beyond, if needed.

Only the power of public systems can provide health care, clean drinking water, education, and the multitude of other services all societies require to reap the benefits of modernity and escape the shackles of entrenched poverty. None of these proposals is easy. But the rewards awaiting us at the end are well worth the arduous journey of getting there.