Are Development Statistics Manipulable? (SWP 37)

Coordinating foreign aid distribution to the poorest countries requires classifying them into developmental cohorts. In principle these designations are objective and immune from manipulation by aid-seeking countries. The objectivity and reliability of these data are important for aid distribution as well as for the use of these data in social scientific applications. We ask whether there are indications that these data are being influenced by aid-seeking manipulation. To do so we examine the distribution of GNIs per capita around the eligibility threshold for World Bank’s International Development Association (IDA). We examine the data as whole and separately for countries that are plausibly more motivated to aid-seek by virtue of their aid-dependence or more capable of doing so by virtue of being perceived as trustworthy. We show that the distribution of GNIs per capita from aid-dependent countries displays indications of aid-seeking data manipulation. This finding is robust to a variety of model specifications, but somewhat sensitive to the exclusion of individual countries from the sample. As such, these findings are more suggestive than definitive, but they do lend credence to the idea of data generation as a strategic process and suggest the need for more research in this area.