We estimate a two-stage Heckman selection model of credit card adoption and use with a unique dataset that combines administrative data from the Equifax credit bureau and self-reported data from the Survey of Consumer Payment Choice, a representative survey of US consumers. Even though the survey data from the borrowers vary somewhat from the data provided by the lenders, the results based on the merged data are qualitatively similar to those based exclusively on self-reported surveys. This finding suggests that if administrative data are not available, it might be sufficient to use survey data to estimate consumer behavior. We find that credit card revolvers have lower income and are less educated than other cardholders. Although consumers who carry credit card debt might be liquidity constrained and not have cheaper borrowing alternatives, the high cost of paying off credit card debt could exacerbate existing inequalities in disposable income among consumers.