Association for Public Policy Analysis & Management

We report the results of a field experiment which overcomes previous data hurdles to link advertising to search and search to purchase for energy efficient home improvements.  We are thus able to test for asymmetries in how pecuniary versus non-pecuniary advertising content affect search behavior versus purchase behavior.  The experimental design allows us to do three things: first, we can directly assess how changing the price on search and the information in an ad affects search and purchase behavior.  We find that while non-pecuniary signals have a strong influence on search, they have no significant effect on final good purchases.

We show that the effect of a particular nudge varies significantly across observable characteristics, which is consistent with the existing literature. Our design also allows us to evaluate the effectiveness of different nudges across similar households. We show that different household types respond to different types of nudges in different ways. Nudges that may be counterproductive for one type of household can be effective for others. We use these results to estimate a counterfactual simulation with each household receiving the most effective nudge based on their pre-experiment observable characteristics.