Payment Habits During COVID-19: Evidence from High-Frequency Transaction Data

This paper uses high-frequency transaction data to assess how consumers’ payment habits have adjusted to the COVID-19 pandemic. Understanding changes in payment behaviour is important for monitoring cash demand. It also allows us to shed light on the reliability of these transaction data in assessing the state of the economy.

We use a comprehensive dataset with data for automated teller machine (ATM) withdrawals and point-of-sale (POS) debit card transactions from two payment infrastructure providers in Canada, Interac Corp. and the Automated Clearing Settlement System. We construct daily measures of payment habits, such as the share of cash transactions and average transaction values. The dataset allows us to measure the number and value of transactions in which the customer or card holder and the acquiring machine, like ATM or POS, belong to the same bank. Using simple dummy regressions and local projection models, we assess how these indicators of payment habits have changed throughout the COVID‑19 pandemic.

We find evidence that during the pandemic consumers adjusted their behaviour to avoid frequent trips for cash withdrawals and POS purchases. Instead, they performed fewer transactions but for higher amounts. Consumers also made smaller-value cash withdrawals compared with the value of card payments, and they performed a smaller number of cash transactions than card transactions. This could reflect a reduced use of cash for POS transactions. Consumers also made more withdrawals from ATMs that are linked to their financial institution than from other ATMs. Finally, we show that estimates of economic activity based on debit card data alone could be misleading if shifts in payment habits are not taken into account. We estimate that debit card payments might have overstated consumer expenditure growth by up to 7 percentage points over the course of the pandemic.