City size and the quality of marriages: Evidence from 3 societies

The paper investigates how marriage rates and the gains to marriage are affected by city size in three societies, medieval Tuscany, China in 1980 and the United States in 2000. Internal migration was severely limited in China until the late seventies. Population supplies in each US city were instrumented with their own population supplies twenty years earlier. The scale effects parameters are estimated precisely for all three societies. To a first order, there is no return to scale in marriage rates in all three societies. There are small decreasing returns to scale effects in China and the US. The paper rejects the presence of a thick market externality in marriage markets for all three societies