This paper studies the effect of the Spanish Reconquest, a military campaign that aimed to expel the Muslims from the Iberian Peninsula, on the population of its most important cities. The almost four centuries of Reconquest offer a “quasi-natural” experiment to study the persistence of population shocks at the city level. Analyzing city growth before and after the onset of the Reconquest, we find that it had a significant negative effect on the population of the main Iberian cities. However, when we control for time effects, we conclude that in most cities this effect was transitory. In order to quantify the duration of the shock driven by the Reconquest we then estimate its average effect on the urban share of these cities considering the time dimension of the entire panel of cities simultaneously and adding city-specific time trends. Our estimates suggest that these cities regained their pre-Reconquest shares on average in less than 100 years. These results are robust to controlling for a large set of country and city-specific socioeconomic indicators and spatial effects. Our findings suggest that the locational fundamentals that determined the relative size of Iberian cities before the Reconquest were more important determinants of the fate of these cities than the direct negative impact that the Reconquest had on their population.