The World Bank spring meeting earlier this month introduced the new World Bank Group president, David Malpass. The uncontested election of President Trump’s nominee appears to confirm an informal understanding that an American will run the World Bank Group, with a European at the head of the International Monetary Fund.
Analysts are watching Malpass’s public statements closely.
Malpass’s plans for the World Bank are controversial. A long-standing World Bank critic, he favors capping World Bank loan amounts and restricting the number of countries that receive aid from the bank. Malpass outlined his general agenda for the World Bank in the Financial Times, emphasizing growth-promoting policies such as lower taxes and fewer regulations for low and middle-income countries. He also intends to cut funds for certain middle-income countries that currently receive aid.
Reducing inefficient spending and shifting priorities have the potential to be beneficial for the bank. Reports published by the Brookings Institution and the Council on Foreign Relations have advised the bank to focus on countries most in need (e.g., poor and war-torn countries). The bank, some critics argue, continues to try to address too many priorities at once.
Still, any reforms will need to account for two realities to be effective. First, the focus of many donors, including the World Bank, has shifted to conflict-prone regions as the “new frontier of development.” The World Bank plans to double its money allocation to fragile countries and provides about 70 percent of its aid to the African continent, where many of those states are situated. Second, China may step in to fill any vacuum resulting from a reduction in World Bank lending.
We measure the impact of aid on conflict.
Our new AidData working paper looks at both of these issues. We investigate how aid affects violence in African nations, analyzing the relationship between aid projects and conflict. Our research examines the World Bank’s role as the most important traditional, multilateral donor but also looks at China’s aid projects.
Measuring the actual impact of aid on conflict is difficult. Our statistical analysis uses modern geo-referencing techniques to match the locations of aid projects and conflicts more precisely than earlier studies. This way we advance on previous methodological, empirical and theoretical contributions to the “does aid work” debates. The figures below animate the locations of active Chinese and World Bank projects (blue dots), as well as the regions experiencing conflict (in orange) over time.
Even with this precise data, it’s easy to confuse correlation for causation. For instance, if the World Bank was more likely to start projects in regions with a higher likelihood of experiencing conflict, then simple correlations might lead us to falsely conclude that aid projects cause conflict. We address this by using several statistical methods to establish that greater aid causes greater stability. One method links the extent of available World Bank resources to aid projects across regions. Such an approach allows us to calculate aid flows that are unaffected by conflicts.
Here’s what we found.
We find that World Bank aid reduces governments’ violence against their citizens in our sample period between 1995-2012. This is probably because the threat of losing future foreign aid inflows acts as an effective deterrent.
Our results also allow us to distinguish aid projects in different sectors. In particular, projects aimed at financial development and transport infrastructure appear to be associated with a lower risk of conflict. Projects in those areas can fuel growth, and cushion economic shocks, which can otherwise often lead to conflict.
We know that other traditional donors, the African Development Bank, and the Asian Development Bank, often adopt the World Bank’s policies. Here’s why this matters. Prior research finds that overall aid budgets shrink by up to 59 percent once the bank deems countries sufficiently “developed” and no longer eligible for World Bank funding. Increasingly, China may be able and willing to step into the World Bank’s shoes. Sri Lankan officials, for example, have stated that they are more than happy to welcome China as a new donor if traditional donors hesitate or withdraw their engagement.
We also assessed the potential effects of this type of shift toward Chinese aid, by evaluating the impact of Chinese development assistance. On average, Chinese aid projects also help to reduce violent conflict and are not associated with more strikes, demonstrations or riots than World Bank-funded aid projects.
But there are more documented incidents of repression by governments against citizens in regions that receive Chinese aid — as Roudabeh Kishi and Clionadh Raleigh explore here in the Monkey Cage. We also find that Chinese aid leads to deteriorating attitudes toward democracy and greater acceptance of autocratic regimes by their citizens. Both approaches thus might provide stability in conflict areas — but with the downside of added repression.
Malpass faces crucial choices that affect the World Bank’s contributions to economic development in Africa — these funds also impact peace and stability. While Malpass’s policy of focusing on the poorest of the poor is laudable, withdrawing resources also creates a gap in countries that seem well on track. Chinese aid filling this void may undermine democratic development. If the World Bank wants to maintain its contribution to peace and democracy while restructuring its portfolio, the bank needs to consider feedback effects on stability. This may set the World Bank apart from other donors.
Kai Gehring is a senior researcher in economics at the University of Zurich. Follow @KaiGehring1.
Lennart Kaplan is a postdoctoral fellow at the German Development Institute. Follow @LennartKap.
Melvin H. L. Wong is a PhD candidate at the Leibniz University Hanover. Follow @mhl_wong.
They are the authors of “Aid and Conflict at the Sub-National Level: Evidence from World Bank and Chinese Development Projects in Africa,” a recent study published as part of the AidData Working Paper series. The views expressed are those of the authors and should not be attributed to AidData or funders of AidData’s work.