In 2005, the Dutch Central Bank published a paper aptly titled Payments are no free lunch74. The report mea-sures the costs of payments, defined as the sum of all the internal costs borne by the relevant stakeholders in the payment value chain. The stakeholders are the central bank, the commercial banks, the retail industry and consumers. The report concludes that the total costs of payment systems are considerable. In the case of the Netherlands they represent:
- 0.65% of GDP or €2.9 billion;
- €0.35 per transaction;
- €400 per household.
The study also compares the variable costs of different payment instruments per transaction and finds that cash is cheaper than debit cards, credit cards and electronic wallets.
The cost of payments has also been researched by the central banks of Belgium, Finland, Germany, Hungary, Sweden and Australia. The results, summarised on page 55, differ from country to country depending on the mix of payment instruments, the efficiency of the economy and probably the methodology used to measure costs. However, the studies reach four common conclusions:
- There are significant costs involved with making payments. They vary from 0.12% of GDP in Finland to 0.99% in Hungary. In a separate paper75, the ECB has estimated that on average, the cost of payments is 1% of GDP (based on a sample of 13 EU countries). If we extrapolate this to the world, the overall global cost of payments amounts to roughly $760 billion, and this is probably a conservative estimate. In comparison, this exceeds the revenues of the airline industry, estimated at $745 billion76.
- The cost of cash is in line with the share of cash as a payment instrument. In the ECB study, cash payments represent nearly half the total costs.
- Cash payments represent the lowest unit cost per transaction in all countries but Sweden and Australia.
- The structure of cash-related costs differs from that of electronic payments. The costs of electronic payments are essentially fixed as they are related to the infrastructure. On the other hand, cash-related costs are both fixed and variable. Items such as processing and transportation, as well as the opportunity cost of holding cash inventories, increase with the value of the transaction. As a result, cash is more cost-efficient for low-value transactions.
Malte Krüger and Frank Seitz77 have undertaken a critical review of payment cost calculations and conclude that “Great uncertainty is attached to estimates of the costs associated with the payment system. Due to the many unique aspects of different countries, we would especially warn against attempts to apply the findings for one country to another without making adjustments.”
Cash is not always the most efficient payment instrument but in some cases it is. And this is particularly true for low-value transactions which represent the bulk of retail payments. In the case of the Netherlands, the cheapest instrument, irrespective of the transaction amount, was e-money78
but the Dutch Chipknip scheme was abandoned in February 201579
The cost of a cash transaction remains a benchmark and is used to measure the efficiency of other payment methods. This is shown in the Merchant Indifference Test80
(MIT), which is used, among others, by the European Commission to establish the level of Multilateral Interchange Fees applied by card schemes. This test ensures that merchants do not pay higher charges than the value of the transactional benefits from using cards. Merchants enjoy such transactional benefits if card payments reduce their cost relative to cash payments.