The Dictator's Handbook

by Bruce Bueno de Mesquita

  The Impact of Aid

  Example after example highlight the simple fact that aid is given in exchange for policy concessions far more readily and in far larger quantities than to reduce poverty and suffering. Following World War II, the rich nations seem genuinely to have thought they could free the world of poverty through their generosity. But no sooner did aid begin to flow than the politics of survival intruded on the noble goal of reducing misery. It should not be surprising that politics prevailed over benevolence. The record is unambiguous: foreign assistance has proven ineffective at alleviating poverty and promoting economic growth.

  In the aftermath of World War II, Europe faced many challenges. Even the victors had suffered enormous human and economic losses. The United States launched a widespread relief program known as the Marshall Plan. Adjusted for inflation, the United States pumped over $182 billion in economic assistance into Europe between 1946 and 1952. Britain was the largest recipient, followed by West Germany, France, and Italy.14 The United States’s goals were to build stalwart states aligned against communism. To achieve these ends, the United States needed an economically strong Europe. States that were willing to combat communism and follow US-dictated economic plans got aid; those not willing to do so, didn’t.

  Over the entire postwar period, total US economic assistance was nearly $1.3 trillion. Military aid over the same period was about $650 billion. To give some perspective, together these economic and military aid packages are roughly twice the size of the 2009 US stimulus package and Troubled Asset Relief Program (TARP) funds put together.

  The success of the Marshall Plan proved hard to replicate. Trillions of dollars have been pumped into developing economies, yet there is precious little to show for it if we measure performance by assessing improvements in the quality of life. As we’ve seen, aid has done virtually nothing to relieve poverty.

  Among policy makers, this record has prompted a fierce debate about the efficacy of aid. For critics, it is all too easy to point to many aid-dependent states in Africa that are poorer now than they were at independence. The development community likes to counter that such a direct comparison is unfair and argues that, while aid-dependent nations have performed poorly, they would have done even worse without aid. This defense, while wrong, is a sensible argument that needs to be taken seriously.

  We cannot simply condemn the aid enterprise just because nations that received aid performed so poorly. To understand why, consider the following provocative statement: hospitals kill! There is plenty of evidence to support this claim. The likelihood of dying is much higher for a person in a hospital than for a person who is not. Of course, most of us instantly see the error in the evidence. The people in hospitals are sick. Healthy people are not to be found staying in hospitals. But this kind of error from looking at statistics without thinking about where they come from is all too common.

  A colleague of ours, Peter Rosendorff, organized a petition and appealed to the Santa Monica City Council to put a crosswalk at a dangerous junction near his former house. The City Engineer said that quite to the contrary, the city was planning to take out all the crosswalks because their study showed that pedestrians were more likely to get killed in crosswalks than anywhere else. The children of Santa Monica should be grateful to Peter. He took the time to explain that the result was not because crosswalks were inherently more dangerous but rather, they are where people cross the road.

  Assessing the true impact of hospitals or a particular treatment or drug is difficult unless we understand who is being treated. The medical community uses randomized drug trials to test the efficacy of medicines. Patients are randomly split into two groups: half get the medicine being tested and half get a placebo. The effectiveness of the drugs is determined by comparing the performance of the two groups. If, alternatively, the medicine is given only to the sickest patients, then even if it is an effective treatment the group getting the medicine might do worse than the group that did not. Likewise, if aid agencies target aid at those nations facing the most serious problems, then aid could appear ineffective even if it was actually working.

  Ideally, to assess the effectiveness of aid, the international community should undertake controlled experiments, giving aid to some randomly chosen nations and withholding it from others. But since it is unlikely aid will ever be allocated in this way, economists need to use complex (and controversial) statistical procedures to adjust the results according to which nations get treated. Rather than delve into these convoluted procedures, we offer some simple evidence based on the United Nations Security Council (UNSC).

  The UNSC is composed of five permanent members (the United States, Russia, China, Britain, and France) and ten temporary members. The temporary members are elected for two-year terms on the Security Council and they are ineligible to be reelected in the two years after their term expires. Election to the UNSC is highly prestigious and, as it turns out, valuable too. Unfortunately, its value comes at a cost: bringing hardship for the people in many of the countries that get elected. On average, nations elected to the UNSC grow more slowly, become less democratic and experience more restrictions on press freedoms than eligible nations that are not elected.15 For instance, during a two-year term on the UNSC, the economy grows an average of 1.2 percent less for nations elected to the council than for nations not elected. Over a four-year period (the two years on the UNSC and the following two years of ineligibility), the difference in growth averages 3.5 percent less for elected UNSC members, that is, nearly 1 percent per year. The effects are much stronger in autocracies than democracies.

  The effects of council membership on growth are fascinating and should cause us to question why the UN is held in such high regard. They also provide an important piece of evidence about the impact of aid. Nations elected to the UNSC get more aid. A UNSC seat gives leaders valuable favors to sell in the form of their vote on the Security Council, and the aid they receive results in worse performance for their economy. Recently there has been a profusion of studies that show that nations elected to the UNSC get financial rewards from the international community. They get more US and UN aid, better terms and more programs at the IMF, World Bank, and a host of other institutions. 16 Membership on the UNSC gives national leaders a say in formulating global policy. Many leaders, particularly those from autocratic nations, appear to prefer to sell this influence rather than exercise it on behalf of their people’s interests.

  UNSC membership comes as close to a randomized test as we are likely to get. Although who gets elected is not random, it is unrelated to the need for aid. Indeed, population size appears to be the only systematic determinant of UNSC elections. African nations, in particular, appear to have adopted a norm of rotation. Nations are elected simply because it is their turn. The key point is that prior to their election, UNSC members behave no differently from other nations. But once elected they actually underperform. To return to the medical analogy, nations elected to the UNSC are not sicker than nations not elected. They get an extra shot of medicine (aid) and it makes them sicker (poorer, less democratic, and less free press).

  UNSC membership gives leaders the opportunity to sell salient policy support. As we have seen over and over again, autocrats need to pay off their coalition. Aid provides the money to do so and that helps leaders survive. Further, aid encourages autocrats to reduce freedoms for two reasons. First, aid revenue means leaders are less dependent upon the willingness of people to work, so the leader does not need to take as many of the risks that arise from freedom, risks they must take when their revenue and worker productivity depends upon allowing people to communicate with each other. Second, the policy concessions are generally unpopular, so leaders need to suppress dissent. UNSC membership brings prominence and prestige to a nation. For an autocratic leader it also means more easy money. For the people of autocratic nations the UNSC means fewer freedoms, less democracy, less wealth, and more misery.

  The historical record shows that aid has l
argely failed to lift nations out of poverty. It is perhaps ironic that while aid affords the resources to alleviate poverty and promote economic growth, it creates the political incentives to do just the opposite. As Edward Walker, US ambassador to Egypt (1994–1998) succinctly put it, “Aid offers an easy way out for Egypt to avoid reform.”17

  An Assessment of Foreign Aid

  So what are we to think of foreign aid? Is it good for policy, or just good politics?

  It has certainly had its successes. Foreign aid, in the form of the Marshall Plan, lifted the predominantly democratic nations of Western Europe out of economic disaster. But the deck was stacked in the plan’s favor. The United States wanted to promote an economically powerful bloc as a means of combating Soviet expansion. The plan therefore promoted economic growth. Democrats need policy success and so were happy to comply with US policy goals in exchange for substantial aid. Yet as we now know, subsequent aid donations have failed to replicate the success of the Marshall Plan.

  What aid does well is help dictators cling to power and withhold freedoms. And yet, the quest to make aid work for the poor is phoenix-like in its ability to rise and rise again. Or, come to think of it, maybe, like Sisyphus, we just keep climbing the same hill only to fall down again.

  Every decade or so, donor nations launch new initiatives to “get aid working.” The most recent manifestation of this is the Millennium Development Goals. Set up by the United Nations Development Program and adopted by world leaders in 2000, this program sets poverty, health, gender equality, education, and environmental targets to be reached by 2015. For instance, the poverty eradication goals call for reducing by half the number of people living on less than a dollar per day. Commendable as such declarations are, saying you want to make poor people richer or at least less poor and actually doing so are completely different things.

  Millennium Development Goals are not the first such declaration to end poverty. They were preceded by efforts to attain “self-sustained growth” first, in the 1940s and 1950s through infrastructure development; then with the US P-4 program to make scientific and technological breakthroughs readily available to poor countries; followed in turn by John Kennedy’s declaration that the 1960s would be the “Decade for Development.” The goals, set back in the late 1940s, remain the same and scant evidence suggests that the world is closer to achieving those goals than it was in the 1950s or 1960s. William Easterly has discussed the hope and optimism that accompany these roughly once-a-decade initiatives. He laments that while each new plan says it will be different, they repeat the same errors of the past. He argues the bureaucracy involved in giving aid ensures funds are given in ways that impede rather than promote economic activity. Poverty persists.18

  Still, we don’t need to be completely pessimistic about aid. Our knowledge of how it works has greatly improved. For instance, we know that aid works much better in the presence of good governance (just as we know that more often than not it goes to places with bad governance).19 Proponents of development assistance point to the success of NGOs undertaking directed programs within nations. Some of these programs have produced wonderful successes. For instance, in 1986 the Carter Center started a plan to combat Guinea worm disease, a parasite transmitted via dirty drinking water that affected about 3.5 million people in seventeen nations across Asia and Africa. By 2009, worldwide infections had been reduced to about 3,000, mostly in southern Sudan.

  Nongovernmental organizations (NGOs) have proven that they can effectively deliver basic health care and primary education. Yet harking back to our discussion of public goods provided by small-coalition regimes, we can’t help but notice that these benefits are precisely the kinds of public policy programs that even the most autocratic leaders want to initiate. NGOs are less successful at providing advanced education. Autocratic leaders in recipient states don’t want people to be taught how to think independently enough that they could organize opposition to the government.

  The successes of NGOs in promoting basic education, basic health care and sanitation, and other basic necessities—digging wells, electrifying villages, making very small business loans (at what we would describe in the United States as usurious interest rates)—all point to a fundamental failing of aid programs and to the harm being done unwittingly by many NGOs and their supporters. It is a simple fact that aid money is fungible. This means recipient governments have nearly complete discretion about moving funds from one project to another. With direct government-to-government transfers it is easy to see why this is so. Autocrats want to provide private rewards for their supporters. NGOs don’t typically want to help the rich get richer and so they provide funding for specific projects or do the work themselves. However, in practice, recipients are very skilled at converting aid into the kinds of rewards they want rather than the kind of rewards donors want them to provide.

  The most sensible criterion for assessing aid’s effectiveness asks not how much money is spent or even how many wells are dug, schools built, or villages electrified, but rather how many people are helped. NGOs count how much money they spend to evaluate their efficacy, but this is a flawed criterion. It encourages charities to help the easiest to reach and the more visible cases while ignoring the difficult and harder to reach people who might well be those in greatest need. Counting the number of people helped also encourages agencies to undertake work that the government would have otherwise done on its own. Remember that NGOs are most successful at providing basic public goods like primary education and basic health care—services even autocrats want. When aid funds are used to substitute for government spending, then few, maybe even no one, has actually been helped unless the government uses the freed-up money for other projects of benefit to the general population. Of course, they don’t. They use the money to shore up their political position and the loyalty of their essential backers.

  Cambodia is a case in point. Half of the Cambodian government’s budget is made up of foreign aid. Rather than supplementing government programs, these donor funds are largely directed toward the bank accounts of government officials. Indeed, Cambodia ranks among the world’s most corrupt nations. As USAID reports, “Donor funds have flowed into education and health, and some of these are passed on to ordinary citizens. But, there can be little doubt a significant portion of funds earmarked for schools, teachers and textbooks, and for clinics, health workers, and medications are diverted.”20

  That is, the funds intended for the people are diverted to rewards for Cambodia’s rich. Often when NGOs provide aid, the amount of assistance is substantially less than the numbers reflect. Suppose an NGO provides basic education to 100 children in a village at a cost of $100 per child per year, for a total expenditure of $10,000. It sounds like 100 people are helped by the NGO, pleasing their donors and bringing in more money. The reality of how many are helped, however, is less clear. The government might well have paid to educate half of those children (or even all of them) itself, even if there were no expectation of aid. Nominally the agency helps 100 children. But in reality they help fifty children at twice the nominal cost and let the leaders abscond with $5,000. Is this good? Well, yes, for the fifty extra children. Is it bad? Well, yes, for all of the people since the NGO is facilitating the government’s opportunity to steal more money and the NGO is helping to further entrench a bad government in power to plague the people for many more years to come.

  Even some of the simplest acts of charity have bad consequences that enhance government control and irresponsibility. To take a personal example, Alastair took his children on a tour of Kenya in 2009. One of the stops was at a primary school where they were encouraged to help paint classrooms. It seems like a nice idea to help out and many people enthusiastically grabbed paintbrushes, eager to brighten the classroom. Alastair objected on principle and went outside and taught some of the kids how to use a digital camera. Was he being a Grinch or was he encouraging better economic policy? From the economic perspective, having highly skilled tourists an
d their families paint classrooms is at best ineffective and at worst downright harmful.

  Comparative advantage lies at the heart of economics. Everyone should specialize in what they are relatively good at and then trade their goods and services. This way everyone ends up with more than if everyone tried to do a little of everything by themselves. Consider the comparative advantage of Kenya relative to Britain, where most of the people on Alastair’s trip were from. Education levels are low in Kenya and there are lots of unemployed manual laborers. Kenya’s comparative advantage is therefore in industries requiring lots of relatively unskilled labor. Indeed this is where it flourishes: Kenya is a huge exporter of flowers. It has a great climate for growing and lots of people to tend to the labor-intensive processes of growing, picking, and packaging flowers. The flowers are then flown to Western Europe for sale. In exchange, Europe exports goods that require human and physical capital to produce—pharmaceuticals, machinery, and computer software. Europe has a relative abundance of human and physical capital. It trades its capital-intensive products for Kenya’s labor-intensive agricultural products and both nations are better off.