Background: Since the mid-1990s, South Korea has undergone two remarkable social changes: a large-scale expansion in higher education and a transition to lowest-low fertility. These changes offer an appropriate quasi-experimental setting for the causal inferences of the impacts of college education on transitions into marriage and parenthood.

Objective: I examine the effects of the large-scale college expansion on first marriage and first childbirth, using data from South Korea.

Methods: I define two cohorts of women depending on their exposure to the expansion (pre-expansion versus post-expansion), and from this I identify a marginal group affected by the college expansion. Using a difference-in-difference approach, I examine how marriage and childbirth changes in this group (the New College Class) differed in comparison with the changes in other groups (the High School Class and the Traditional College Class).

Results: I found a considerable impact of college expansion on the falling rates of first marriage and first childbirth among the New College Class women. The growing divide in family formation between college graduates and non-college graduates explains a large part of the total college expansion effects, while the effect of increased education among New College Class women was minimal.

Conclusions: The college expansion in South Korea did have an impact, but the impact was mostly indirect from interactions with other social structural changes.

Contribution: I provide causal evidence on the impact of the large-scale expansion in higher education on family formation, in particular fertility, utilizing a novel analytical approach and a rare empirical case in South Korea.