March 2012

A New Flavour for eCash

mcw 2012

At the Digital Money Forum in London this month Marc Brule the CFO of the Royal Canadian Mint outlined the research they have been doing on the evolution of currency. He explained the need to be able to handle cash like transactions on the internet and how it might be achieved by an asset transfer model which for those who were around in the 90's might appear to be reminiscent of Mondex. There are still a lot of people around who think Mondex got it right but just 20 years too early.

Electronic cash has long been the holy grail of payment systems largely because it's the one area into which the current payment system players can expand. This of course includes PayPal as well as Visa and MasterCard. The properties of cash such as anonymity and immediacy without transaction fees are of course diametrically opposed to the classic credit and debit card systems for which the players have to validate the authenticity of each transaction and are so involved in the transactions that they also have to manage disputes and chargebacks. It goes without saying that this will incur fees.

There are three ways to address a cash like electronic alternative,

  1. Minimize the costs associated with the classical 4-Party model

The Royal Canadian Mint has gone for the third approach by adopting a direct asset transfer model. There are advantages and disadvantages of each approach which we will look at in more detail.

The 4-Party model is that commonly seen today when paying with a credit or debit card. The bank holds the consumer's accounts for which funds can be either pre-paid or post-paid. The bank that holds the account is called the Issuer because it traditionally issues a card which allows the consumer to access their account. The consumer is called the CardHolder because they traditionally hold the card that gives access to the account. The Merchant is the provider of goods or services to the cardholder in exchange for payment. The fourth member of the set is the merchant Acquirer which is the bank or service provider that acquires the transactions for clearing and settlement within the networks of the financial institutions. In other words the Acquirer has to get the funds from the Issuer on behalf of the merchant.

It doesn't take very long to appreciate that all these various parties have costs and therefore want to charge fees for their payment services. The Issuer charges an Interchange fee for the costs of managing the cardholder relationship, the acquirer charges for the costs of managing the transaction (which will include an authentication and authorization process) and the network processing costs applied by the payment operators such as Visa and Mastercard. All these fees of course are accumulated and get applied to the merchant, sometimes called the merchant discount fee. These fees are applied per transaction and for a small merchant may be 3% or more for a credit card transaction. Debit card transactions are usually a fixed fee per transaction and for a small merchant may be 25 pence or more per transaction.

It is immediately obvious that this 4-P model is unlikely to be economically feasible for very low value transactions of a few pounds or less. In the cash world these small value payments dominate the market.

So the question is how can we reduce the cost of these 4-P model transactions. That's not easy and the approach adopted by the payment operators to date has been to remove the on-line authentication/authorization step. Effectively what the Issuers are doing is taking the risk on an off-line transaction if the value is low enough (e.g $20 or less). However ignoring the details for the moment it is clear that there is an increased risk and only a part of the costs have been reduced. In addition you would want to minimise the risk of card fraud and at the least would want the consumer to be using a secure smart card. This is really not a place for a magnetic stripe card. From the consumers point of view there is an advantage because the Issuer often has to bear part of the transaction risk under the various consumer protection laws and of course if you lose your card you don't lose your money (assuming you report it to the bank).

You don't need a degree in economics to realize that this still isn't going to work for transactions with a value of say less than say $5 and in the virtual world of the internet there is a need to handle transactions perhaps less than $1. The ultimate test is whether you can handle a transaction of 1 cent? Not by the 4-P model! The second approach to handling low value payments is to have some form of aggregation. What you want to do is to avoid the 4-P overheads of handing individually small value transactions. Cardis is a company that has been working on this for some time with a scheme that does effectively aggregate small transactions leading to reduced costs. They have recently announced that Austria's Raiffeisen Bank International is to use Cardis software plug-in for low value payments in Q2 this year.

The other approach to aggregation is effectively a subscription model where you lodge funds with a service provider on a pre or post-paid relationship. What is clear is that you are effectively setting up an account with the service provider and although it is a closed system (only you and the service provider) there are still the overheads of the service provider having to authenticate the consumer requests. This part of the transaction is no different to the 4-P model and if the scheme is post-paid then there is also the risk overhead of the service provider not being paid. Apple's iTunes is effectively a subscription model. From a consumer's point of view this works with any intermediary you regularly use such as Amazon or iTunes, we will however save discussion on the 30% fee taken from the service providers by Apple for another day. Can you do a 1 cent transaction? Mmmm now that also seems unlikely because you still need to cover the overheads.

And then not last is the third model where the transaction does not require intermediaries. This is the approach being adopted by the Royal Canadian Mint in their current R&D project. This is of course exactly analogous to the use of cash. I can make a payment to you (in the physical space) where no third party is involved in the transaction. In this case the asset is being handed from person to person. What the Mint has been working on is the ability to do this electronically in the internet world so again you can pass 1 cent from person to person. Now of course nothing really comes for free, in the UK it has been estimated that it costs £4 Bn per year to manage cash. It is the cost of collecting it all up, counting it and securely distributing it between merchants, banks and consumers.

Now this is the $64K dollar question (just changing currencies for the moment), can you electronically manage electronic cash without hidden overheads? Mondex actually had a good go at this back in 1990 (can you believe it's that old) and suffered with a lack of infrastructure. There were few mobile phones in those days and who do you know that was using the internet. So let's see what happens to a 21st century approach to electronic cash.

Conie Mutters Analyst, (Electronic Payment Systems)

Ed: Since this article was written the Royal Canadian Mint has gone public with a developer Challenge to test and gather industry ideas for its MintChip eCash system. They have provided a lot more information on their web site which we will review next month.


Unable to open RSS Feed with error SSL certificate problem: certificate has expired, exiting

Video Interviews

Tim Jones talks on the wealth of networks

Christophe Dolique of Gemplus talks about ·SIM

Dominique Brule of Philips Semiconductors talks about Near Field Communication