Matching, Management and Employment Outcomes: A Field Experiment with Firm Internships

Young professionals in developing economies face significant barriers to high-quality professional employment. We evaluate a novel intervention in Ethiopia that places recent graduates into established firms to shadow middle managers. Using random assignment into program participation, we find that the one-month placement significantly increases the probability of wage employment in the short run. Six years later, treated individuals see a 12\% increase in average earnings, driven by a shift among the top decile of the wage distribution. We rule out signalling and social networks as primary drivers, finding instead that the effects are rooted in the accumulation of practical managerial skills and familiarity with professional organisational practices. We run the experiment using a firm-proposing deferred-acceptance algorithm to match professionals with firms. We then develop a generative model of preferences and use this for counterfactual mechanism design. Our results show that the choice of matching algorithm is pivotal: under random matching, the program would likely have yielded no significant short-run impacts. Our results demonstrate that incorporating participant preferences via matching algorithms can improve the design and efficacy of field experiments.