In the good old days, the large one-stop-shop smart card manufacturers would hide large mark-ups inside their smart card prices. They would procure the chips, add their proprietary operating system and applications ROM mask, print the plastic, embed the chip module and slap on a high price tag. They could do this because they had a niche market and nobody knew any better.

A major shift is under way as the value chain splits into more parts with more players [1]. However, what is bad news for some is good news for others. New opportunities are appearing particularly in the software-related segments of the new value chain such as operating system and application implementation, and smart card management systems.

The price of a chip card is generally inversely proportional to the volumes ordered, which is why we see the same chips being sold into different markets for very different prices. A good example of this is the Visa Low Cost Card Programme where member banks can obtain “white plastic” Global Platform, 16K EEPROM EMV cards (SDA only) for US$2.70, or Global Platform, 8K EEPROM EMV (DDA) cards for US$2.60. The latter card is capable of public key cryptography that is the basis of emerging network ID applications. However corporate network ID cards typically sell for considerably more, due to the lower volumes.

At the very bottom of the scale is the US$1 native EMV card (SDA only). This is possible due to a deal struck by Visa with chip provider, ST Microelectronics, and operating system and application provider, IBM, who will do very nicely out of the guaranteed banking volumes due to the EMV liability shift in 2005.

I imagine that the card manufacturers providing these cards will receive very little from embedding the modules into white plastic. Of course, their revenues will come more from the physical personalisation requirements of banking cards (such as printing, embossing, holograms and magnetic stripes) and electronic personalisation (loading the chip and magnetic stripe data relating the bank and the customer). And these revenues will be worthwhile due to the shipment volumes which Consult Hyperion forecasts to grow at an average of around 35% annually over the next five years.

Pump up the volumes

Until the lower security EMV SDA cards are routinely cracked (say, two years?) and the higher security EMV DDA cards become a banking requirement, the US$1 card will do very nicely for getting hundreds of Visa member banks around the world to put their toes in the EMV water. DDA cards are required to perform Public Key cryptographic calculations and therefore on average cost US$1 more than less functional cards. Most of this goes to the chip providers to cover the cost of the necessary co-processor.

Let’s look at another example, SIM cards. From the manufacturer’s point of view, these are even worse than banking cards. Yes, the volumes are high, but not growing significantly like forecast banking processor card volumes. To make matters worse, they require virtually no personalisation. Each card has a unique cryptographic key or two, but this data is not linked with a customer until the SIM card is finally sold to the end user; there is no personal information stored on SIM cards until the user starts filling up the phone book. Like Visa and the other payment schemes, the Mobile Operators are grouping together to drive down the price of SIMs.

So where is the money to be made? Clearly the chip providers will do very well since their shipment volumes are forecast to rocket and the kind of chips in demand in most sectors are increasingly expensive due to complexity and amount of free application and data memory they contain.

Operating systems and application providers should also do well. Someone has to develop the EMV card masks for the payment schemes and the SIM card masks for the mobile operators. For example, there are several new players in the Java Card and SIM implementation market who have nothing to do with card manufacture. They tend to have more of a software background; some smaller companies will be happy to take a small piece of a very large pie.

Smart card management itself will be lucrative as smart card shipments escalate since the management system suppliers are likely to operate on per-card license revenue scheme. Interestingly enough, Consult Hyperion predicts that most of this revenue will not be associated with post-issuance downloads to multi-application cards. Even if multi-application card usage takes off (and you never know, it might in, say, the cellular mobile and national ID sectors) post-issuance downloads remain a feature of a far-distant future. On the other hand, post-issuance technology can work well, as Consult Hyperion showed the world for the first time in 1998 [2].

Mobile operators want to modify their SIMs in the field in order to offer value added services, but their bandwidth is too low for load new applets over the air. Instead they are settling for sending an SMS to tell the SIM to change a bit or a byte to modify the behaviour of a pre-loaded SIM Toolkit applet.

Perhaps all is not lost for the manufacturers. After banking, our processor card market analysis suggests that the national ID card sector will show the biggest growth over the next five years. These cards will require special production techniques that might allow card physical personalisation specialists to do well. The plastic and the printing techniques used will have to be considerably higher quality than for, say, bank cards in order to satisfy government demands for cards which cannot easily be forged and will last up to ten years before wearing out.

Volumes will dictate that there is still money to be made in card manufacturing. Margins, however, are in decline and what we are seeing is organisations changing how and where they operate in the value chain in order to chase those profits.

References
[1] M Christensen, M Raynor and M Verlinden, “Skate to where the money will be”, Harvard Business Review, pp 72-81, November 2001. [2] J Elliott, “Lifestyle card games”, Card Forum International, pp 36-39, Vol 3. No 3, May/June 1999 By John Elliott, Principal Consultant, Consult Hyperion.
mailto:john.elliott@chyp.com
www.consult.hyperion.co.uk
http://www.cards-worldwide.com