MOTIVATION: We contribute to an under-researched area of the literature on building effective bureaucracies: how much (and what type of) information to use when selecting among job applicants in the public sector. RESEARCH QUESTION: Does having more information on job applicants when hiring public health workers lead to better public health outcomes? RESEARCH DESIGN: We implement an RCT on hiring HIV/AIDS health promoters (HPs) in Mozambique. Neighborhoods (the unit of analysis) are randomly assigned to either a control group (with no HP) or to having an HP work there. Among those receiving an HP, neighborhoods are further randomly assigned to one of three treatments in which HPs are selected on the basis of different amounts of information on job applicants. The treatments involve increasing amounts of information available on job applicants as inputs to the hiring decision. The primary outcomes of interest will be the number of HIV tests and refills of antiretroviral therapy medication in study neighborhoods. Comparisons between control and treatment neighborhoods will reveal the impact of having any HP. Comparisons between treatment groups will reveal the impact of varying amounts of information in the hiring process. Secondary outcomes of interest are worker prosociality and neighborhood-level measures of HIV-related knowledge and stigmatizing attitudes. POLICY RELEVANCE: Improving the quality and effectiveness of public health personnel is a top priority for Mozambiques Ministry of Health (MISAU). This research is in partnership with a research center of MISAUs National Institute of Health (INS), Centro de Investigao Operacional da Beira (CIOB). CIOB is a regional research center with responsibility for public health research in central Mozambique. Dr. Joo Manuel, Director of CIOB, is a co-investigator on the study. This partnership ensures the relevance of our research and provides a strong basis for influencing health policy in Mozambique with our findings.
