Source: Data on attrition at the store-owner level come from the follow-up survey of the experiment. Data on attrition at the store level come from the administrative records of Casas de Pollo Rey (CDPR).

Note: Each column in this table presents the results from an ordinary least squares (OLS) regression of a different attrition measure on a treatment dummy. In column (1), attrition at the store-owner level is defined as a dummy for failing to complete the follow-up survey, while attrition at the store level is defined as a dummy for missing sales data in June 2022. In column (2), the outcome is a dummy for having attrited the sample because the owner closed their store. A dummy for having attrited the sample because the owner established a different business in the same location is the outcome variable in column (3). The outcome in column (4) is a dummy for attriting the sample because the name of the legal owner of the store changed to avoid personal income taxation. Finally, the outcome in column (5) is a dummy for changing the store code due to store relocation or for fiscal purposes, such as avoiding corporate or payroll taxation. Standard errors within parentheses are robust to heteroskedasticity of unknown form in panel A and are also clustered at the store-owner level in panel B. All regressions control for strata dummies.