Digital Payments Need Discipline

The rise of cashless payments has raised the concern that some consumers will be induced to spend beyond their means. As monthly income is spent over the course of a month, if consumers are aware of their reducing balance, they are likely to constrain their spending. The Swiss National Bank has published a paper that explores differences in how freely people spend between people who pay in cash and digitally and how their awareness of their balances affects their cash management 1.

Central banks, and others, have concerns about digital payments, particularly as they dominate payments and are now squeezing out cash. Can central banks continue to guarantee safe and reliable payments to the universe of consumers? Can they maintain monetary sovereignty?

There are also concerns about the privacy of consumer data, the potential market power of payment service providers and, which is the focus of this paper, vulnerable groups in society spending beyond their means.

The model and its assumptions

Traditional economic theory suggests that the form of payment can influence spending behaviour, with cash transactions often associated with greater psychological pain of parting with money compared to electronic payments. To investigate the relationship between payment methods and spending behaviour, the paper using a proprietary dataset that links a payment-methods survey, a paymentdiary survey and a behavioural survey for a representative sample of 1,138 Swiss consumers.

The paper assumed that consumers make their payment choice and monetary management decisions in a planned period and can perfectly commit to these decisions. The model included payment choice, money management and intramonth spending. It assumed there were two types of goods in each subperiod of the month.

The paper looks at two types of consumers. Those who are ‘present biased’ (PB) and those who are ‘time consistent biased’ 2. Present bias is the tendency to settle for a smaller present reward rather than wait for a larger future reward, in a trade-off situation. Impulsivity and procrastination are key behaviours associated with the present-focused preferences of consumers.

The behavioural survey included indicators of ‘present bias’, measures of discount rates, risk aversion, trust, memory and conscientiousness.

In a planning period consumers could either pay only with cash, only with a card or with a mixture. When card payments were used, consumers could only pay up to their account balance. There was no liquidity constraint and they had total access to those funds. With cash they could only pay up to the point where they had money in the wallets. Going to an ATM to get more was a self-control mechanism.


Consumers with high PB bias spend more if they pay more frequently by card, or another cashless method. The average discretionary spend per day was CHF 78.5. A one standard deviation increase in the frequency of cashless payments saw a 6.5% increase in spending.

The daily spending of those with low level PB was unrelated to their preferred payment instrument.

The paper found important heterogeneities in the impact of cashless payments on spending across age groups and income groups. Cashless payments are associated with higher spending by PB consumers amongst both low and high income households but not among younger consumers.

There was no robust evidence suggesting consumers strategically choose payment instruments and cash management to self-control spending. PB is not associated with a more frequent choice of cash over cashless payments nor is it associated with higher frequencies of cash withdrawals.

However, the sensitivity of the frequency of cash withdrawals to income is less for households with higher levels of PB than those with lower levels of PB, ie. there is a weak take-up of costly commitment devices among consumers with selfcontrol issues.

While digital payment platforms often offer features such as budgeting tools and real-time transaction monitoring, there is evidence to suggest that some consumers may struggle to manage their finances effectively in a cashless environment. This is particularly true for vulnerable populations, including low-income individuals and those with limited access to traditional banking services.

The prevalence of automatic payment methods, such as subscription services and recurring billing, can lead to a ‘set it and forget it’ mentality, where consumers are less mindful of their spending decisions.


Research about India’s 2016 demonetisation exercise has found that moving to digital payments resulted in increased supermarket spending. However, the 2021 staggered roll out of debit cards to low income bank clients in Mexico saw spending fall and savings rise. The 2022 staggered roll out of contactless debit cards in Switzerland saw no impact on spending.

The opening hypothesis was that ‘naïve’ PB consumers who predominantly pay cashless will overspend, while PB consumers who pay by cash will not. There was also the hypothesis that sophisticated PB consumers may choose costly payment behaviour and cash management to selfcontrol their spending.

The paper found that consumers with high levels of present bias spend more when they pay frequently by card as opposed to paying frequently by cash. Digital payments may, therefore, induce higher discretionary spending by impulsive consumers. For consumers with low levels of present bias, spending is unaffected by payment instrument usage.

The heterogeneity tests reveal that the effect of cashless payments on spending is strong both for low- and high-income consumers but not for young consumers. This conclusion comes with a caveat about the small size of subsamples in the dataset.

Final word

The findings of this paper make clear that even without the risks that come with access to easy credit (think ‘Buy Now Pay Later’), digital payments remove a budgetary constraint that those with present bias are ill equipped to handle. Another largely hidden strength of cash. The design of CBDCs might benefit from considering this.

1 - Cashless Payments and Consumer Spending. Martin Brown, Yves Nacht, Thomas Nellen, Helmut Stix. SNB Working Papers 6/2023.

2 - Time-consistent: The situation arises when someone makes a commitment to take an action in the future. If the incentive to keep the commitment is the same as the incentive to make the commitment, then the example is time consistent.