The project aims at investigating the determinants and effectiveness of sub-national aid project. This project introduces two novel databases, both of which we expect to be useful for research over and beyond this one. The first database is the Geocoded Official Development Assistance Database (GODAD), which is a geocoded, dyadic panel dataset of development aid projects assembled from various sources. We geocode data for 18 bilateral European donors and the United States from raw project data in the OECD DAC's Creditor Reporting System (CRS). The data we present in this first version of GODAD is not just instrumental to our key research question, but will hopefully also serve as important input for numerous future studies on the allocation and effects of development aid at large, which are currently limited by the lack of geocoded aid data for a large number of donor and recipient countries. We complement these data with geocoded information on Chinese aid commitments and World Bank projects. Second, we introduce what we call the Political Leaders' Affiliations Database (PLAD). The database includes the precise geolocated birthplaces of leaders and their spouses all around the world, as well as information on their education, profession, ethnicity, and their number of children, covering a sample of 177 countries worldwide over the 1989-2020 period. We thus complement existing data on political leaders that provide information on leaders' tenure and details on turnover.