India reached below-replacement fertility (TFR = 2.0) despite low income, persistent traditional gender norms, and declining female labour force participation. We study how the expansion of a decentralized, fee-based education market reshaped fertility decisions by increasing households’ exposure to the cost of child quality. Because sons are more likely than daughters to be enrolled in private schools, this cost channel interacts with son preference to generate gender-differentiated fertility stopping behavior. We test predicitons from a dynamic model of parity progression using survey data from NFHS (2015–21) merged with administrative district-level UDISE data on 37,251 women in 525 districts at the parity-2-to-3 margin. We find that households with only daughters are less responsive to local private-school prevalence than otherwise similar households with sons. This result is stable across subsamples, the exclusion of recent migrants, controls for modernization proxies, and controls for differential baseline-fertility trends. At the district level, private-school expansion is associated with lower parity progression, but this effect is sensitive to district-level confounding.