The smart card is clearly the most viable of alternatives for cards that might cut across various industries and functions
Plastic cards are evolving not just in the direction of new applications but also toward multiple applications, and the choice of technology becomes much more critical than it was for simple credit or debit cards.
Much of the discussion revolves around the integrated circuit card, or smart card, and for good reason. It is clearly the most viable of alternatives for cards that might cut across various industries and functions, of which traditional payment or credit will be only one.
The drawbacks of the legacy technology - the magnetic-stripe card - have become all too clear. As a payment system it is ubiquitous, allowing for transactions at retail stores, restaurants, supermarkets, the travel industry, gas stations, automated teller machines, etc.
But the magnetic stripe's Track 2 is not rewritable for security and reliability reasons. And authorizing a transaction requires an inquiry into an on-line data base for authentication, validation of the personal identification number, and validation of the transaction amount.
The traditional on-line solution - which works well only in countries like the United States with comprehensive availability of cheap telecommunications - requires that all transactions be funneled to a central site and then to application providers' data bases. After that, an authorizing or rejecting message must go back out to the transaction point.
The system architecture is 25 years old and takes up valuable time, central-site and card-issuer computer capacity, and network and communications resources. It also lacks sufficient security. The credit card networks still cannot handle and protect PIN codes, which, based on comparison between ATM transactions with PINs and point of sale transactions without them, cut losses by a factor of 50.
Some more advanced technologies have limitations, too.
Supermarket-type bar codes lack the capacity to handle the minimum 19- digit banking industry account number. High-density laser-optical cards require expensive read/write devices that cost more than $1,000, and data on the cards are not rewritable.
Though these cards have a huge information capacity, updating information can be difficult and unwieldy. The cards also must be carried in a protective casing to prevent surface damage - something bank card users have never been accustomed to do.
The smart card is most promising because of three major attributes: secure and portable data bases, programmable computer logic and control, and full communications management.
These attributes allow each of the anticipated multiple application providers to place its own control rules and values within the chip. This can allow 85% to 90% of transactions - the routine items - to be PIN- authorized, authenticated, and validated at the terminal site without requiring an on-line authorization call.
Transaction data accumulate in the terminal and are typically forwarded to a central repository in an overnight batch transmission. The in-card controls are updated continually and, once a month, synchronized on-line with the central accounting and payment records. In that monthly procedure, authorization parameters can be altered, and applications may be added, changed, or dropped.
On exception transactions, the programmed communication instructions assure that data are routed directly to the appropriate control point, no longer to a central funnel. After a certain number of failed attempts to enter a PIN, the card "locks up" until it can be returned to the issuer. A failure to pay bills prompts tightened authorization controls for each application in the card.
The integrated circuit, or IC, card has a lower cost per transaction than the magnetic-stripe card. This stems from the IC card's longer life and superior security, in-card controls, loss controls, and bad-debt controls. The biggest savings come from reduced network and control-center usage.
All customers can be given an IC card and, within it, a menu of services including the level of credit for which they qualify. Further revenue potential comes from the multiple applications; the card issuer can "rent" space on the card.
One European airline collects fees for placing nine different logos on its cards, including those of MasterCard and Visa. It recovers the cost of card issuance with only three such rentals. The returns per card can be up to 20 times that from magnetic-stripe credit cards.
More than half of current bank credit cards are actually offered by nonbanks. IC technology creates an opening for any type of organization to issue multiple application cards. A securities broker, for example, could easily be a card issuer and could contract with a bank for bank-related services.
What would those services be? I could now get any banking product from a local discount brokerage company. Other industries will not wait for bankers to wake up and seize the opportunities in multiple application cards.
Mr. Svigals, a consultant, heads Jerome Svigals Inc., Redwood City, Calif., and publishes Smart Cards & Comments newsletter.