There is little agreement on the question of whether U.S. households are well-prepared for retirement. We revisit this debate using the Consumption and Activities Mail Survey (CAMS) linked to the Health and Retirement Study (HRS) to better understand the dynamics of pre- and post-retirement consumption. We first develop a dynamic model with home production that captures conventional reasons why consumption may decline during retirement: Time preference rates, a decline in work-related expenses, health shocks, and changes in home production. We also consider an additional factor—inattention—that allows for a gradual adjustment to new economic conditions. Even after retirement, households may not realize how much they need for both current and future expenses. Dividing the sample into terciles based on retirement adequacy, we find that differences in consumption paths diverge not immediately at retirement, but over a lengthy period of time; by 10 years post-retirement, those poorly prepared for retirement experienced a drop in consumption roughly one-third greater than those best prepared. This differential decline does not appear to be explained by health status, family size, marital status, time preference, risk aversion, or home production such as cooking at home or shopping for bargains. Instead, these patterns are most consistent with a model of household inattention.