Rankin On Thursday - Y2K
Keith Rankin's Thursday Column
Y2K: the Benefits exceeded the Costs
6 January 2000
Perverse creatures that we are, some people were looking forward to the world coming to an end on January 1.
Others were anticipating several days of infrastructure failure arising from the Y2K computer bug as their big new century thrill. Now, disappointed that the fireworks that they were looking forward to didn't materialise, some are claiming that the bug never existed and the whole scare was a grand conspiracy perpetrated by computer consultants on the make.
Did the money spent on finding and fixing date-related bugs represent a waste of resources, a necessary clean-up task (akin to the Exxon Valdez oil spillage cleanup), or something more?
Those who argue that there was no bug and therefore that Y2K expenditure was money wasted point to Third World countries, the former Soviet bloc, and to western countries like Italy who spent comparatively little on Y2K compliance.
One Third World country - The Gambia - did have major problems on the night. Indeed, it only takes one example to prove that the problem existed and therefore could have created panic if not chaos, had it not been attended to.
This kind of comparative analysis can be a very useful tool. For example comparative analysis shows that New Zealand's approach to monetary policy (through the 1989 Reserve Bank Act) achieves inflation rates no better than elsewhere, suggesting that the policy might represent, at best, a waste of effort.
However the comparative method must be used with care in an era in which countries are not independent of each other. While information technology (IT) is consumed globally, the design of hardware and software is dominated by the USA. Problems identified and fixed in the USA are in many cases world-wide fixes.
In a sense, countries like Italy escaped relatively unscathed because of Americans' efforts. That's fair, because it was disproportionately American programmers who gave the world the bug in the first place.
Not only was it a global effort of which most of the cost was funded in America, but also the USA, with more dependence on information technology than any one else, had more that could potentially go wrong than did any other country. Indeed the Pentagon did have a major problem with its spy satellites. While no other country except Russia has a defence establishment on the scale of the Pentagon, the American version contains many more computer systems than does Russia's.
The argument that the bug did not strike because most of the world's software was audited and fixed is much more sound than the claim that the bug was never there.
By December 1999, we knew that the chance of a major strike was at most 1 in 20. That meant that there were 19 chances in 20 that there would be no Y2K bug strike. So we should hardly be surprised at the outcome. (We do have trouble dealing with probability and uncertainty. For example, France probably had a 1 in 10 chance of winning against even a well-prepared All Black team on October 31.) Nevertheless, despite our reluctance to accept that chance is more intrinsic than will to the workings of the universe and all who dwell there, it is normal for us to prepare for life's booby prizes. It's called insurance, civil defence and social welfare.
I will argue further than others have done, and claim that the Y2K audit was an important item of social investment; more a benefit than a cost. Our computer systems were in need of a general review, regardless of Y2K. The prospect of a future (in, say, 2080) dependent on software for which we no longer had the source code has been a matter of concern to me. The perpetuation of crude 1960s' and 1970s' software within 1980s' and 1990s' software packages constituted a classic case (like pollution) of market failure; of cost externalisation. The private sector would never have initiated a general software audit if there was no private interest at stake.
The change of century forced the industry worldwide to take stock in a way that it would never have done under any other circumstance. That stocktake had major public good spillover effects; what economists call "positive externalities". Programs not up to scratch that are still running have been flagged for retirement as well as patched. Many other programs have been rewritten with the public consequences of bad software firmly in mind. Indeed computers talking to each other - the Internet - represent a community rather than the myriad of stand-alone individuals that existed in the 1970s.
The Y2K bug was the wake-up call that the computing industry needed. We can now be confident that real economic gains will arise from the IT revolution (the third industrial revolution). The bug, which is real, forced the private sector to fulfil its public duty. The resources used constituted a part of a process of mainly private investment in a knowledge society. The software foundations of such a society are now more like rock than sand.
The global software audit was an investment which will have a high (though unmeasurable) social rate of return. If that investment had not been made - ie if there had been no bug - then complacency would have set into the IT industry. And private complacency leads to public disaster, as the All Blacks discovered.
© 1999 Keith Rankin
Thursday Column Archive (1999): http://pl.net/~keithr/thursday1999.html