GDP growth rates are like a standard transmission in a car. For practical purposes, you get them in three speeds forward and one in reverse. The three speeds forward are 0-5 per cent, 6-10 per cent and probable nonsense. Don't try to put numbers on the reverse. My advice is to work on the basis that the first number to the left of the decimal in GDP growth rates is uncertain, let alone the first to the right. The reason, particularly in so open an economy as Hong Kong's, is that most of the components of GDP cannot be based on hard numbers. They exist in the merchandise trade figures, but for such things as changes in stocks the numbers come largely from surveys, and these are at times open to a huge margin of imprecision. This has become moot again with the announcement that GDP contracted by an estimated 2 per cent in the first quarter. The breakdown of this number into the components of GDP has not yet been published, and the figure is almost certain to be revised over coming months. But there is no denying that there has been a slowdown, and the published figures for the components up to December last year show some interesting trends. For one, the slowdown in tourist arrivals has hit harder than many people realise. Visitor expenditure in Hong Kong accounted for the equivalent of only 5.5 per cent of GDP last year, down from 7.1 per cent in 1996. Remember that these are annual figures. The slowdown hit only in the second half of the year. The month-to-month trend was worse by the end of the year and likely to have become even more so in the first quarter. Given that the counterbalancing outflow of expenditure by Hong Kong residents abroad was still holding firm at the year end, this alone could have tipped the first quarter growth numbers into negative territory. Building a cable car to the tourist trap at Ngong Ping will not go far to reversing the trend, but it is understandable that the tourist industry should be thrashing around to look for a solution. Another trend of interest is that construction activity was moving at a faster pace in the private than in the public sector. Public sector construction accounted for 4.2 per cent of GDP in last year, down from 5 per cent the previous year while the private sector share rose to 6.5 per cent from 5.8 per cent. The lower numbers for the public sector can in part be attributed to the slowdown in construction activity at the airport as it finally neared completion. It is also open to some quibbles as to what is private and what is public expenditure. But, superficially at least, it suggests that the Government can point a finger at itself too when asking why growth is down. Overall, the breakdown for last year indicates the problem components were still the smaller ones. The big components, such as personal consumption expenditure and fixed capital formation, were relatively firm. We now need later figures to see if this is still the case.