Sovereign governments, multilateral institutions, and non-governmental organizations have created a wide array of tools to assess the policy performance of developing countries, including conditional aid and debt relief programs, blacklists, watch lists, performance-based organizational accession procedures, cross-country benchmarking exercises, and country-specific diagnostics. Scholars and policymakers generally agree that these types of performance assessments can influence the policy priorities and actions of public sector decision-makers in low-income and middle-income countries. However, there is little systematic evidence about the specific conditions under which performance assessments instigate changes in policy behavior. There is also a lack of understanding about the causal mechanisms through which different types of performance assessments facilitate policy changes. We seek to close this evidence gap by leveraging a survey of 1784 government officials that provides comparative data on the agenda-setting influence and reform design influence of more than 100 performance assessments in 123 low-income and middle-income countries. Using a multilevel linear model that accounts for the hierarchical structure of the survey data, we find robust evidence that performance assessments yield greater policy influence when they make an explicit comparison of government performance across countries and allow assessed governments to participate in the assessment process. This pattern is consistent with the theoretical argument that policymakers carefully weigh the credibility signaling benefits and policy autonomy costs of addressing performance assessment requirements.
Funding: This study was made possible through generous financial support from the John Templeton Foundation and the Smith Richardson Foundation.