Consumer Cash Preferences over Time and the Factors Influencing Them: Examining 2012 and 2017 Cash Preferences in the United States

Abstract

I employ a multinomial probit model to analyze U.S. consumer payment diary data from the 2012 and 2017 Diary of Consumer Payment Choice surveys to determine the demographic and economic conditions that influenced consumer payment choice in these years and how these preferences have changed over time. I additionally run counterfactual simulations to determine overall changes in cash preferences between the two years in order to control for differences in demographic and economic conditions across the two data sets. I find that in 2012, age, income, transaction value, and merchant type influenced consumer cash preferences. For the 2017 data, I find that the aforementioned categories plus education and race influenced consumer cash preferences. In general, I find income, transaction value, education, and some categories of merchant type and race to be negatively correlated with consumer cash preference. I find age and gift or donation payments to be positively correlated with consumer cash preference. Overall, I find that cash preference decreased between 1.03 and 1.08 percentage points between the two years. These findings area important for policy considerations, such as possible cashless businesses bans or steering provisions, because they show that lower-income groups are those who generally rely on cash and are thus most vulnerable to harms from cashless business models and price increases resulting from high credit fees to merchants.