Source: Authors’ calculation.

Note: All specifications account for the stratification variable, caste indicators, and a baseline indicator depicting whether an individual has used sanitary pads in the past. Column (2) adds household and personal controls such as father’s education, mother’s education, family income, relationship dummies, age, and age squared, as well as variables from the baseline survey pertaining to the household’s attitude and one’s knowledge regarding menstruation. Column (3) adds indicators representing whether the household’s distance to the nearest college and market is less than 30 minutes on foot, and column (4) includes area fixed effects. Additionally, columns (5), (6), and (7) account for the compliance probability, rate of turnout specific to a neighborhood (⁠|${\frac{\text{number attended}}{\text{number invited}}}$|⁠), and inverse Mills ratio (IMR), respectively. White standard errors robust to heteroskedasticity are presented in parentheses. The q-values adjusted for the false discovery rate based on Storey (2002) and Storey, Taylor, and Siegmund (2004) are presented in brackets. The F-statistics and |${{Pr}}(\gt F)$| are from the joint hypothesis testing under the null that the coefficients on the interaction terms are jointly equal to zero. The proportion of redemption for the 10 percent discount in the subsidy-only group (omitted category) is 2.9 percent. *p < 0.1, **p < 0.05, ***p < 0.01. The marker X denotes inclusion of the designated controls in the model specification.