A comment on "Examining Inequality in the Time Cost of Waiting"

Holt and Vinopal (2023) investigate whether there is inequality in how much time people spend waiting for services using the American Time Use Survey (ATUS). They find that (1) high-income people are both less likely to wait and spend less time waiting than low-income people, and that (2) this is true even after conditioning on observable differences. Further, they find that (3) income has heterogeneous effects on waiting time by race and ethnicity. First, we successfully computationally reproduce the paper's main claims, but uncover two minor coding errors. However, we find five parts of Holt and Vinopal (2023) where the description in the paper differs from what is implemented in the code. Fixing these impacts their results. We also conduct several robustness tests, including calculating standard errors using the replication weights provided by ATUS, updating the outcome variables to use all waiting time for services recorded in the ATUS, and using weekly earnings as an alternative measure to household income. Between fixing the differences between the code and the paper, and our robustness tests, we conclude that Claim (1) is robust, Claim (2) is robust in some specifications, and Claim (3) is not robust.