Does the Benefits Schedule of Cash Assistance Programs Affect Impulse Buying? Evidence from a Natural Experiment in Peru

A common critique of cash assistance programs is that beneficiaries may use the money in ways that adversely affect their own welfare.  One line of evidence points to payday or “first-of-the-month” effects, in which individuals are more likely to purchase “temptation goods” such as alcohol and sweets immediately after receiving cash support.  Building on the existing literature, we exploit a change in the payment schedule of a social program to identify the impact of benefit receipt frequency on the purchasing practices of member households.  We hypothesize that larger, less frequent payments lead households to make more temptation purchases.  Our study is the first natural experiment in the literature on payday effects. 

Unlike many other conditional cash transfer programs, the Juntos program in Peru provides a fixed, lump-sum payment to all eligible households.  Starting in January 2010, the government moved from a monthly to a bimonthly payment schedule.   We use annual household data from 2007 to 2012 to analyze the effect of this policy change on the share of the household budget devoted to several temptation expenditure categories: alcohol, tobacco, sweets and sugary foods, soda and sugar-sweetened beverages, prepared meals consumed at home, and food purchases at restaurants, in contrast to budget expenditures on staples.  The household data include panel and cross-sectional samples collected by Peru’s National Institute of Statistics.

We determine the impact of the payment schedule change using a triple differences estimation strategy, before and after the policy change for member and non-member households consuming temptation and staple goods.  We derive a series of demand equations using a Quadratic Almost Ideal Demand System.  Our multivariate analysis includes area-level fixed effects in the repeated cross-sectional sample and household fixed effects in the smaller panel sample.  In the latter, we track the purchasing behavior of the same households over time before and after the policy change.

We evaluate whether larger, less frequent payments induce beneficiaries to purchase a greater quantity of temptation goods.  We conduct additional analyses on the frequency and quantity of purchases to rule out that the effect is attributable to bulk purchasing and increased food storage.  To marshal further evidence in favor of the proposed causal pathway, we use area-level variation in payment dates and interview dates to examine whether beneficiaries who recently received a payment are more likely to engage in temptation purchases. 

Our study highlights the importance of benefit scheduling for the purchasing behavior of cash recipients.  Policymakers may be able to curtail the degree to which public program recipients buy temptation goods simply by distributing payments more diffusely over time.