The goal of an individual searching for a marriage partner is typically to form a
long-term relationship. Marital search is a complicated and costly activity, where opportunities
typically arrive over time at uncertain intervals, each party has to evaluate
each other’s characteristics, and expectations play an important role. Given these
features of marital search, a seminal paper by Mortensen (1988) has shown that the
matching framework can be suited for the analysis of marriage markets and also raised
the possibility of a thick market externality in these markets. We contribute to this
literature by empirically investigating whether marriage markets are characterized by
increasing, constant, or decreasing returns to scale. We focus on three societies–late
medieval and early Renaissance Tuscany, China in the 1980s, and the United States in
2000–which are different in terms of population size, economic structure, sex ratios,
marriage transfers, and the social norms governing marriage markets. Our main finding
is that in all three societies, there is no evidence of increasing returns to scale in marriage
markets, whereas the hypothesis of constant returns to scale cannot be rejected.
The remarkably similar and precise estimates suggest that the number of eligibles (and
potential contacts) in a marriage market is less important than economic factors, such
as wealth levels and income dispersion, in affecting the marriage rate across different
societies. The key message is that where individuals live, in large cities or small towns,
have a minimal effect on their marriage rates. In particular, when looking at the United
States, city size per se is not responsible for the recent decline of marriage rates.

JEL Classification: J12, J41
Keywords: marriage markets, matching, thick market externality, returns to scale,
spousal search, endogeneity