UK businesses can be forgiven for squirming at estimates of fixing the millennium computer problem. After all, the billions of pounds we are told they will need to spend is not going to improve productivity or customer service. Instead, the experience will be like flushing money down the toilet.
Last week, Accountancy Age published a calculation which estimated that u35bn would need to be spent. The source of the figure was Chantrey Vellacott’s head of economics, Maurice Fitzpatrick. Later, the head of the Government taskforce responsible for raising the profile of the issue, Robin Guenier, produced a calculation of his own for the Financial Times. He said UK enterprises would need up to u31bn to correct the problem.
Guenier said this week the figure was both ridiculous and conservative.
Ridiculous because the process of calculating the scale of the problem was like putting the tail on a donkey while blindfolded. And conservative because the figure could be at the top end of a scale which runs from u15bn to u50bn. ‘The number is very big and UK enterprises have little time left to make an assessment of their own situation. In fact, some have already left it too late to change their systems,’ he said.
The problem centres on the microprocessors found in everything from household appliances to PCs and mainframe computer databases. The two-digit date code which computer chips currently use will not be able to distinguish between 1900 and 2000, causing them to crash.
Last week’s story in Accountancy Age that companies for sale could have their price driven down by a prospective buyer if tests reveal millennium compliance difficulties, adds to Guenier’s point. Buyers, after all, are not going to be generous in their assessment of the cost of fixing systems by 2000.
But the fancifully large figures bandied about are convincing some commentators that Guenier is exaggerating to make his point and most companies can rest easy. This week’s Accountancy Age/Reed Accountancy Personnel survey of 200 FDs reveals a large minority of companies – 38% – have yet to assess their exposure, let alone do anything about it.
And there are still accountants who play down the importance of compliance when there is still just under three years to go before the deadline.
Michael Oaten, head of corporate finance at Arthur Andersen, said a company would need to be heavily dependent on IT to be affected at this stage, adding that there was enough competition among buyers to avoid dropping a sale price.
‘If anyone asked for u1 off the purchase price, I’d walk away from the deal and find another buyer – there are plenty of them.’ Oaten said he had recently completed the sale of railway leasing company Eversholt and the issue had not be raised, despite the company’s widespread use of IT.
But Guenier could be underestimating the cost of consultants and computer programmers. He has made all his calculations based on 1997 prices. As Maurice Fitzpatrick noted, US estimates have placed a $1 charge for changing each line of computer code. By 1999, some experts warn that the cost could rise to $8. By then, supply and demand for specialist skills will be out of balance and companies which have left making the changes to the last minute will pay through the nose.
THE PRICE OF THE MILLENNIUM DATE CHANGE
Maurice Fitzpatrick believes UK business will spend 1% of its u3.5 trillion annual turnover on the millennium computer problem. The percentage figure is an average based on reports of company spending plans already in the public domain. For instance, BT is set to spend 5%, or u700m sorting out its difficulties. A calculation based on estimates of executable lines of code (Elocs) would come out with a higher figure. US analysts reckon there are 600 billion Elocs worldwide that will need to be rewritten.
Fitzpatrick said estimates currently put the cost at $1 per line. But this is set to rise to $8 per line in 1999. He said if an average of $3 per line was accepted it would put the worldwide cost at $2 trillion.
The UK’s share would be $80bn, or u50bn, on the basis that it accounts for some 4% of global GDP.
Robin Guenier has based his calculation on the likely number of people companies will use to solve the problem. First, he excluded small businesses from the list of the UK’s 3.6 million companies. The one million left are allocated IT staff to fix the problem on a sliding scale. Businesses with 10-50 employees would need a tenth of a person over the next two years, while a company of 500-1000 will need ten people, and so on. This adds up to 266,500 IT people needed each year, which at an average u45,000 salary gives him the figure of u31bn.