In countries, where a substantial proportion of retirement income rests on savings, there is much concern that a sizeable fraction of the population reaches retirement with insufficient financial resources. We define saving regret as the wish in hindsight to have saved more earlier in life. We measured saving regret and possible determinants in a survey of U.S. households in which respondents were aged 60–79. We find high levels of saving regret, affirmed by some 58%. Saving regret exhibits significant and plausible correlations with personal characteristics and wealth: Married, older, healthier and wealthier respondents are less likely to report saving regret, suggesting the measure’s validity. We find only weak evidence for correlations between saving regret and measures of procrastination: persons with traits associated with procrastination express saving regret about as often as those without those traits.