Score contribution per author:
α: calibrated so average coauthorship-adjusted count equals average raw count
SummaryThe ‘BCG hypothesis' suggests that the Bacillus Calmette-Guérin (BCG) vaccine against tuberculosis limits the severity of COVID-19. We exploit the differential vaccination practices of East Germany and West Germany prior to reunification to test this hypothesis. Using a difference in regression discontinuities (RD-DD) design centred on the end of universal vaccination in the West, we find that differences in COVID-19 severity across cohorts in the East and West are insignificant or have the wrong sign. We document a sharp cross-sectional discontinuity in the severity of the disease, which we attribute to limited mobility across the long-gone border and which disappears when we control for social connectedness. Case and death data after the end of the first lockdown on 26 April does not display a discontinuity at the former border, suggesting that mobility (as opposed to BCG vaccination) played a major role during the initial outbreak.