Social policy gone bad educationally: Unintended peer effects from transferred students

What is the impact of an increase of lower-ability students in a university class? We examine a natural experiment in which students from large, low-income families had the chance to transfer to academic programs at a local university. Multiple law changes meant that there was significant, quasi-random variability in the number of transferred students over time, which was orthogonal to the quality of receiving students. We create a novel dataset for the top economics department in Greece and show that the social policy had a negative educational impact by uniformly lowering recipient students' academic performance once the proportion of transferred students exceeded a certain threshold.