Smart card introduction a serious worry

Recently, the Federal Government has been talking of a new "smart card" to access Centrelink and other benefits. Supporters of the idea have envisaged something of a one-stop shop where all the information is carried on the card, with the risk of duplication between agencies and possible fraud being reduced.

Articles can be found here:

ABC 26/4/06 - Smart card gets cabinet approval (7:30 Report)
ABC 24/7/06 - Banks warming to smart card idea: Hockey (ABC Online)

The federal services minister, Joe Hockey, has been pushing the idea in recent days of making the cards usable from bank EFTPOS machines, but has yet to resolve issues regarding interchange fees and the like with the banks despite what by all accounts appear to be quite forceful negotiations.

It is well documented that Centrelink has real trouble coping with its caseload as things stand - their IT systems often screw up or produce unpredictable results, especially with cases operating intermittently over a number of years, and understaffing and lack of training have been problems for years, although to be fair, Centrelink has been working on these issues and many of the most critical ones that plagued them in the late 1990s are now consigned to history. Also, the EFTPOS system is operated by banks which have been increasing both fees and profits over the entire 10½ years that Howard has been in office.

From an ICT transition or risk management point of view, there's a significant logic gap in the Minister's statements.

Firstly, integration, while a highly desirable aim, comes with considerable risks, especially when large amounts of old data are involved. Risks inevitably mean time and money, and few people outside the ICT industry seem to appreciate this when dealing with ICT projects - the belief is that hiring a large enough contractor like Oracle or HP will resolve this problem all by itself.

Secondly, when developing additions to an existing system, it is essential that the existing system works within its existing brief and determined requirements.

Thirdly, change management issues are often not well understood by government - they confuse having legislative or executive power with the actual ability to govern, and their capacity to pass laws and to publish or air advertisements as being a public education campaign which people will carefully follow and obey. However, expecting millions of users and thousands of operators to uncritically adopt any new system, especially one that is hard to use or is accompanied by penalties for accidental misuse, presents a change management challenge. Many large projects have been derailed and/or dropped (at a cost of millions to the community) by the failure to anticipate consequences or risks at the implementation stage.

The consequences of poorly-planned public-sector projects can be clearly observed with another smart card project being developed by Transperth for handling public transport fares. The project, SmartRider, is already more than 6 months overdue, has had considerable cost overruns, still has implementation issues, and a confusing mesh of the old and new systems are still in operation. The new system introduces requirements on ordinary public transport patrons which will most likely result in chaos (or even electoral consequences) if the system ever goes fully "live".

Before jumping in the deep end, the Minister should determine whether such a project is in fact in the best interests of the community, given the likely duration and cost of development and implementation, as well as the unanswered questions about the banks and what happens if a card is lost or stolen.