Journal of Financial Market Infrastructures

As we start the third volume of The Journal of Financial Market Infrastructures I would like to take a look back at the field's most important innovation from the standpoint of the end user. On the ground floor of our warehouse (1) are located all systems, institutions and payment instruments that are needed to make a payment from the consumer/payer to the end beneficiary.

In most countries worldwide it still takes too long to process these payments from end to end and the time it takes greatly depends on the day on which the payment is executed and at what time.An interbank payment submitted after the Friday afternoon cutoff time may have to wait until Monday morning before it is executed. And even if the interbank system is open for business, it may still take hours or a whole day before the end beneficiary can spend the incoming amount again. In the internet age such delays in payments are no longer acceptable.

I see two requirements with different levels of priority. The first priority is that noncash retail payments should be processed twenty-four hours a day, every day. The second priority is that the processing time of a noncash retail payment should be faster than handing over a banknote from your wallet to the cashier. (Incidentally, note that the bitcoin network at the present time has realized the first goal but not the second.(2)) The reasons for speeding up those payments are partly connected to the characteristics of the Internet itself: always on (done) and available everywhere (coming soon). As the payments industry also provides services to E&M commerce, the infrastructure should follow suit.

In some countries, such fast(er) payments are already a reality. This issue's Forum contribution by Kristian Kjeldsen, Lars Egeberg Jensen and Tommy Meng Gladov, entitled "Faster payments in Denmark", discusses the Danish case. The authors describe the gradual approach that has been followed in the renewal of the Danish retail payment infrastructure. It elaborates on the adaptations necessary to the Danish RTGS system, which also affects the second floor of the warehouse. In November 2014 Denmark's instant payments system ("Express Clearing") is scheduled to go live.

Turning to another innovation area - contactless card payments - we come to the paper "The impact of retail payment innovations on cash usage", by Ben S. C. Fung, Kim P. Huynh and Leonard Sabetti. The authors investigate the effect of contactless credit cards and stored-value cards on the use of cash at the point of sale. The data comes from a survey on methods of payment conducted by the Bank of Canada. One obstacle in measuring the effect of innovations is the selection problem: typically, early adopters are not representative of the wider population. The authors use propensity score matching methods to decompose the impact of these payment innovations on cash usage into selection and causal effects. Their results show that the share of cash payments is reduced in favor of contactless credit and stored-value cards.

The other two papers in this issue deal with large-value payment systems. One paper, "Analysis of the use and impact of limits" by Martin Diehl and Alexander Müller, focuses on a particular liquidity management feature of TARGET2: bilateral or multilateral debit limits. The analysis reveals that such limits are rarely used and are not actively managed by participating banks. The authors quantify the first-round effect of limits (longer queues and more delay) and the partially offsetting secondround effect, which is caused by the liquidity redirection of effective limits. It is shown that the net effect is significantly smaller in the case of a more severe stress scenario. Furthermore, in applying limits to late payers, the burden of additional delay is systematically shifted toward the late payers, which clearly benefits the limit setters and punishes free riders.

The other large-value payment systems paper, "Inferring unsecured interbank loans and interest rates from interbank payments: an evaluation" by Q. Farooq Akram and Casper Christophersen, investigates Norwegian overnight unsecured interbank loans using payments data. This paper fits into a line of research started by Furfine in 1999 in which the goal is to infer volumes and prices on the unsecured money market. The authors were able to test the results of their application of the Furfine algorithm using the actual data that banks use to calculate the Norwegian overnight weighted average interest rate. In so doing they provide an estimate of the accuracy of the Furfine algorithm. They find that interbank payments provide reliable information about overnight lending and overnight interest rates at the market level, and even at the bank level for relatively small overnight lenders and large overnight lenders that mostly lend on their own behalf.

I hope you enjoy reading this issue of The Journal of Financial Market Infrastructures.

(1) This warehouse was introduced in the volume 1, issue 2 of The Journal of Financial Market Infrastructures. It represents all the financial (market) infrastructures within a given currency.
(2) See my editor's letter on bitcoin in Volume 2, Issue 3 of the journal.

Ron Berndsen
De Nederlandsche Bank and Tilburg University