Recent economics literature has taken interest in the role of (mis-) perceptions regarding one's relative financial well-being as a determinant of preferences for redistribution policy. Through a randomized field experiment, in a natural setting, I test whether misperceptions regarding relative financial need directly affect the propensity to request a subsidy and one's reported willingness to pay (WTP) for private consumption when it is salient that drawing on subsidy resources may decrease subsidies available to peers.

I survey applicants of co-curricular educational travel programs at U.S. universities on their perceived percentile of need for financial assistance relative to university peers. To a randomized subset I provide accurate information on the distribution of Expected Family Contribution from the Federal Application for Student Assistance (a nationally standardized measure of the family financial well-being of college students). I then observe whether students apply for program financial assistance and, if so, what they report in their aid application as their maximum willingness to pay out of pocket.

Preliminary results suggest a null or slightly positive average treatment effect on reported WTP, driven by heterogeneity in treatment response across the distributions of bias in prior beliefs and of level of financial need. For students who initially overestimate their position of relative need (i.e., who are actually more well off relative to peers than believed), the treatment effect on reported WTP is positive and increasing in the magnitude of bias of their prior. There appears to be no symmetric affect for those who initially underestimate their relative need. Moreover, students in the highest quintile of need for support tend to decrease their reported WTP in response to information, while students across lower quintiles of need increase reported WTP.

Abstract: We conduct an RCT in Mexico City to study the effects of provision of air quality information and provision of a reusable N95 mask on air pollution knowledge and avoidance behavior. Households are assigned to four cross-cutting treatments: (1) a free reusable N95 mask, (2) 50% higher compensation for baseline participation, (3) a year subscription to pollutant-specific SMS air quality alerts, and (4) a year subscription to monthly SMS pollution trend and avoidance behavior reminders. At baseline, we elicit willingness to pay for the alerts service after revealing participant compensation and whether the household will receive an N95 mask, but before revealing whether they will receive alert or reminder services. We observe no significant impact of mask provision on WTP, however, higher compensation increases WTP, suggesting a possible cash-on-hand constraint. Follow-up survey data demonstrates that alerts treatment households are more likely to report receiving air pollution information via SMS, to correctly identify whether there was a high pollution day in the past week, and to report staying indoors on the last high pollution day. However, we observe no significant effect on the ability to correctly recall which specific days had high pollution. Similarly, households that received an N95 mask are more likely to report utilizing a mask with filter in the past 2 weeks, but we observe no effect on using a filter mask on the specific days with high PM. These results illustrate the difficulty of using information treatments to affect behavioral change that could significantly lower air pollution exposure.