Working Paper: NBER ID: w30740
Authors: Hunt Allcott; Daniel Cohen; William Morrison; Dmitry Taubinsky
Abstract: We use public finance sufficient statistic approaches to characterize the welfare effects of “nudges,” such as simplified information and warning labels, in markets with taxes and endogenous prices. While many studies focus on average effects, we show that welfare also depends on how the nudge affects the variance of choice distortions, and average effects become irrelevant with zero pass-through or optimal taxes. We implement the framework with experiments evaluating automotive fuel economy labels and sugary drink health labels. Labels decrease purchases of low-fuel economy cars and sugary drinks but may decrease welfare, because they increase the variance of choice distortions.
Keywords: nudges; welfare; nonstandard policy instruments; behavioral economics
JEL Codes: D90; H0
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| increase distortion variance (C46) | total surplus (D46) |
| average treatment effects of NPIs (I12) | welfare implications (I30) |
| NPIs reduce distortion variance (C45) | total surplus effect (D46) |
| NPIs (L31) | beneficial behaviors (D64) |
| NPIs (L31) | distortion variance (C46) |
| NPIs (L31) | total surplus (D46) |