收入冲击与家庭内部动态：来自保障性收入实验的证据

When an individual's income changes, it can affect their entire household. We study the intrahousehold effects of an unconditional cash transfer in which 1,000 individuals were randomly selected to receive a $1,000 per month transfer for three years, while a control group of 2,000 participants received $50 per month over the same time period. We find no effects on aggregate measures of household stability, although there is suggestive evidence of improved relationship quality and greater trust in partners among recipients in relationships. Transfers to friends and family increase, consistent with recipients sharing some of the additional income with others. There are also spillovers on the income and employment-related outcomes of other household members: household income excluding the recipient decreases by about $1,700 per year, and partners are less likely to be promoted or to move to a better job. Other household members are significantly more likely to participate in education or training, and recipients are also more likely to report someone in their household having to forgo healthcare, potentially reflecting greater engagement with the healthcare system and the identification of previously unmet needs. We observe limited changes in decision-making power or division of labor within the household. Some indicators of intimate partner violence decline, particularly for women, but reported incidence is low and most estimates do not remain statistically significant after adjustment for multiple hypothesis testing. Subgroup differences are generally imprecise, although point estimates suggest larger education and health spillover effects when the recipient is a woman. Household structure also appears to mediates the impacts of the transfers.