The failure of development aid.

• Author: Ovaska, Tomi
• Position: Research on aid for 1975-98 in 86 developing countries

In the last half century, developed countries have paid increasing attention to the problems of developing countries. Not only has the disastrously low level of economic development in large parts of the world become apparent over time, but also all the undesired side effects of enduring poverty: poor health, widespread diseases, low life expectancy, and the general lack of means of entire nations to deal with basic needs. Wealthy countries do not necessarily have purely unselfish motives when helping poorer countries through financial aid. Part of any aid constantly flows back to donors through highly stipulated procurement contracts. Aid has also increased the potential for donors to buy preferential future treatment for the business firms of their own nationality. Politically, aid can be seen as serving to buy increased international and regional clout through new political allies. Moreover, aid may bring more stability to world affairs, assuming it increases economic growth in recipient countries. Development aid is also seen as a way to advance some core values of the donors.

Despite the steady flow of development aid to poor countries in the last 50 years, the results have been somewhat disappointing, as noted by Bovard (1986), Burnside and Dollar (2000), the World Bank (1998), Vasquez (1998), Easterly (2001), and Easterly and Levine (2001). Even though some countries, notably in East Asia, have managed to break out of poverty, many of the poorest countries have actually seen their real per capita incomes decline since the 1970s. More than one billion people still live on less than $1 a day. Many of the advances in basic health care and education in the last few decades have been negated by the rapid spread of HIV/AIDS, particularly in the world's poorest countries. Given the ineffectiveness of past aid and fiscal constraints the donor countries have faced in the 1990s, a type of aid fatigue among the donors has become apparent. The long-term trend of increasing aid disbursements was broken in the mid-1990s, and has given way to a systematic reevaluation of donor-initiated development strategies.

Even if development aid has not been as effective as the international community has wished, world leaders have been consistent in signaling their determination to improve the lot of the poorest countries. The United Nations 2001 conference of 140 world leaders in Monterrey reiterated further support for the Millennium Development Goals--cutting in half the proportion of people hang on less than $1 a day, eliminating gender disparity in education, reducing child mortality by two-thirds, and developing a global partnership for development--that are to be achieved by 2015. How are these goals to be achieved? The World Bank offers a two-pronged solution: first, double the current flow of aid to developing countries, and second, make a new commitment to good governance on the part of the recipient countries. In this study, I look at whether an increase in the level of development aid is likely to result in increased growth rates for developing countries and whether the quality of governance affects the results of aid.

Previous Literature on the Effectiveness of Development Aid

Previous literature on development aid has looked at how aggregate flows of aid have affected economic growth of individual countries. Dalgaard, Hansen, and Tarp (2000) summarize the findings from the last 30 years and a total of 131 cross-country studies by noting that aid has increased aggregate savings and investments, though by less than the aid flow itself, and has led to increased economic growth in cases where the lack of capital base was the most important factor in holding back growth. This conclusion, however, sheds little light on how to improve the growth rates in the larger group of aid recipient countries where growth lags below potential.

The most recent econometric studies have concentrated on suggesting alternative ways to reach better results from development aid. For instance, contrary to many previous findings, Boone (1996) found that in a sample of 96 recipient countries, foreign aid did not significantly contribute to investment or economic growth rates, or to an improvement in human development indicators. However, giving aid to politically liberal regimes did seem to lead to lower infant mortality.

In another recent study, Burnside and Dollar (2000) examined the effectiveness of aid in a sample of 56 developing countries. The authors found that aid given to countries with good institutions and policies has a far higher likelihood to affect growth positively than aid given to countries of poor institutions and policies. For this the authors developed their own quality of governance measures--namely, budget surplus, inflation, trade openness, and institutional quality. However, Alesina and Weder (2002) note that there is little evidence that well-governed countries actually receive more foreign aid.

Using the same sample as Burnside and Dollar, Hansen and Tarp (2001) did find a positive relationship between foreign aid and real per capita growth, but also noted, as have Levine and Renelt (1992), that the results are highly sensitive to the choice of regressors and to the econometric models. The governance measures turned out to be significant, but only under conditions the authors deemed highly objectionable.

Measuring Aid and Governance

Compared with previous literature on the effectiveness of development aid, this study uses a more comprehensive governance measure in aid regressions, two alternative data sets on measuring aid disbursements, and country-specific econometric modeling. Gwartney and Lawson (2002) have constructed an Economic Freedom of the World (EFW) index that goes back to the year 1970, and is far more comprehensive than that used by previous studies. Instead of concentrating on only a few policy variables (budget surplus, inflation, trade openness, institutional quality), the EFW comprises 37 variables that fall under five general headings: size of government; legal structure and security of property fights; access to sound money; freedom to exchange with foreigners; and regulations of credit, labor, and business (Gwartney and Lawson 2002: 8-9).

Even though the index by its nature ultimately relies on individual researchers' assessments of the importance of various variables to economic growth, several recent studies--such as Gwartney, Lawson, and Holcombe (1999); Haan and Sturm (1999); Wu and Davis (1999); Heckelman and Stroup (2000); and Ali...