<p>This study analyzes the relationship between shadow education expenditures and fertility rates at the provincial level in South Korea. Shadow education expenditures are defined as per-child household spending on private tutoring and other supplementary educational services outside of regular schooling. To estimate the causal relationship, I exploit changes in the provincial composition of parents with school-aged children, as distinct from prospective parents without school-aged children, as a source of exogenous variation in shadow education expenditures. Fixed-effects instrumental variable estimation results indicate that a 1% increase in average shadow education expenditures leads to a 0.18–0.26% decrease in the total fertility rate. In particular, increases in shadow education expenditures have a more pronounced negative effect on fertility rates for higher-order births. These findings suggest that rising shadow education expenditures raise the expected education costs of children among prospective parents, thereby reducing the number of children. They also imply that the ultra-low fertility rates in East Asian countries may, to some extent, be attributable to high levels of shadow education spending.</p>