Editor’s Note: The following guest blog is by Ranjit Lall, an Associate Professor of International Political Economy and a Fellow of St John’s College at the University of Oxford.
International development organizations (IDOs)—such as the World Bank, the International Monetary Fund (IMF), the World Health Organization (WHO), and more—collectively receive hundreds of billions of dollars a year in financing from donor countries. In turn, they disburse hundreds of billions to fund projects across the globe. But how effective and accountable are these organizations?
This is one of the most salient questions facing researchers and practitioners of foreign aid. Yet, comparative data on IDO performance and accountability historically has been in strikingly short supply. In a new book and article, I attempt to fill some of this empirical gap by introducing two new public data sources: the Performance of International Institutions Project (PIIP) and the Accountability in Global Governance (AGG) dataset. (In an earlier, related effort with Brad Parks, executive director of AidData, and Dan Honig, professor of international development at Johns Hopkins University, we published the Project Performance Database: the world's largest collection of development projects that includes project outcome ratings of holistic project performance.)
The PIIP, which I assembled for my recent book Making International Institutions Work: The Politics of Performance (Cambridge University Press, 2023), is the most comprehensive collection of donor performance ratings of IDOs. Since the late 2000s, several major donor countries have conducted systematic comparative assessments of IDOs they finance, drawing on diverse data sources including stakeholder surveys, project evaluations, and public consultations. These appraisals have come to be seen as the “gold standard” of organizational performance measurement at the international level.
The PIIP brings together ratings of 54 IDOs issued between 2008 and 2018 by the governments of Australia, Denmark, the Netherlands, Sweden, and the United Kingdom and by the Multilateral Organization Performance Assessment Network (MOPAN), a network of 22 large donor states. IDOs are rated on 51 separate indicators capturing numerous dimensions of organizational performance, most of which can be summarized under four headings: (1) delivery of results, (2) financial management, (3) strategic management, and (4) knowledge management.
Figure 1 below plots each IDO’s average score across all performance indicators and years on a standardized scale. Two features immediately stand out. First, there is substantial variation in ratings: some IDOs, donors believe, have been far more successful in achieving their objectives than others. The second is the conspicuous diversity of IDOs at both ends of the performance spectrum. Among highly rated IDOs, for example, what do the Asian Development Bank (ASDB), the International Fund for Agricultural Development (IFAD), and the World Food Programme (WFP) have in common?
A central finding of Making International Institutions Work is that IDOs that enjoy greater policy autonomy vis-à-vis member states tend to receive higher performance ratings. This is because powerful countries often have incentives to opportunistically interfere in IDOs to promote narrow national interests, at the expense of broader organizational objectives. Expanding staff discretion in the policy process—for instance, by strengthening the secretariat’s agenda-setting powers and preventing small groups of states from vetoing its policy proposals—is therefore essential for improving performance.
If avoiding political interference requires placing constraints on government influence in the policy process, is there an inescapable trade-off between IDO performance and accountability? This is where the AGG database—which I compiled for my recent article in the American Journal of Political Science, “Making Global Governance Accountable: Civil Society, States, and the Politics of Reform”—enters the picture.
The AGG dataset documents the emergence of modern, or what I call multi-stakeholder, mechanisms of accountability in IDOs. These are formalized routines and procedures, based broadly on principles of democratic governance, that seek to enhance the capacity of diverse public and private stakeholders to monitor, assess, and shape IDO activities. Ranging from “access-to-information” policies and evaluation offices to grievance redress systems and public-private governance arrangements, multi-stakeholder accountability (MSA) mechanisms enhance oversight without compromising state participation in the regular policy process.
Drawing on policy documents, online reporting, and communications with IDO staff, the AGG dataset provides information on five major types of multi-stakeholder accountability mechanisms: (1) transparency, (2) evaluation, (3) redress, (4) investigation, and (5) participation. The strength of each type of mechanism is measured on a five-point index for 52 of the 54 organizations in the PIIP between 1960 (or an organization’s founding date) and 2018.
Figure 2 below plots the composite value of the multi-stakeholder accountability index for all IDOs in the AGG dataset. Once again, there is considerable variation within the sample. Interestingly, IDOs with higher performance ratings in Figure 1 generally possess higher multi-stakeholder accountability index scores in Figure 2. In other words, there is a positive relationship between performance ratings and the robustness of modern accountability structures.
In “Making Global Governance Accountable,” I attempt to understand differences in the adoption of multi-stakeholder accountability mechanisms across IDOs. I find that adoption is more likely when member states, in particular the most powerful, face “bottom-up” pressures for accountability from dense transnational civil society networks—networks with the capacity to build leverage through agenda-setting, coalition-building, and advocacy strategies—and when institutions carry out governance tasks that are costly to monitor.
Taken together, these findings provide grounds for cautious optimism. They suggest that IDOs can be made more effective, and that pursuing this goal need not come at the cost of robust accountability to stakeholders, broadly conceived. Nevertheless, as the PIIP and the AGG datasets make clear, performance and accountability problems remain pervasive in multilateral development cooperation. For international development organizations, the path to improvement is not a straightforward one.
Ranjit Lall is an Associate Professor of International Political Economy and a Fellow of St John’s College at the University of Oxford. He is the author of Making International Institutions Work (Cambridge University Press, 2023) and several articles on the political economy of international cooperation.