This study examines gender differences in financial literacy in Peru using nationally representative ENCF data for 2013–2022. It assesses whether observed gaps persist after accounting for measurement choices, socioeconomic composition and distributional heterogeneity.
The analysis compares alternative financial literacy measures: the OECD/INFE core index, an extended six-item index, a wave-specific PCA-based score and an equal-weighted extended index. Mean gender gaps are estimated using pooled regressions with survey-year fixed effects and socioeconomic controls. Distributional heterogeneity is examined through conditional quantile regressions and unconditional quantile regressions based on Recentered Influence Functions (RIF).
Gender differences are sensitive to measurement design and model specification. Women display lower average financial literacy in unadjusted estimates, but these gaps attenuate substantially once education, employment, informality and household characteristics are controlled for. Evidence of a robust mean gender gap is limited. Conditional quantile regressions show no significant differences among comparable individuals, whereas unconditional RIF regressions reveal a female disadvantage only at the lower tail of the population distribution. This suggests that observed gaps are mainly compositional, reflecting women's overrepresentation among socioeconomically vulnerable groups.
The paper shows that conclusions about gender gaps in financial literacy depend on both measurement design and distributional perspective. By distinguishing conditional from unconditional differences, it offers a more nuanced interpretation of gender disparities and supports targeted financial education policies in developing-country contexts.
