On Tue Sep 05 2000, Seth Ackerman wrote:
> Yeah, but can't you do it the other way around?
[Here's the guess a letter writer to the FT came up with: a US real GDP growth rate 1/3 less than that posted for the last two years]
Financial Times, 05-Sep-2000
LETTERS TO THE EDITOR: Statistics explain divide between US and Europe
By KURT RICHEBACHER
Sir, In his speech given in Jackson Hole, Wyoming, Alan Greenspan claimed that Europe fails to reap the big productivity gains that the new economy has brought to the US because labour market inflexibility in Europe destroys the incentive to invest in cost-saving technology. Considering that European businesses have sacked millions, it is a strange argument.
At issue, however, is his assertion and general perception that high-tech capital spending is far lower in Europe than in the US and that this largely explains the tremendous difference in economic and productivity growth. In reality, the great digital and productivity divide between Europe and the US is least of all in the economies. It is overwhelmingly in the statistics and in the propaganda.
The decisive fact is that the US's method to gauge high-tech output and investment differs in two ways dramatically from that in Europe, and together these account for most, if not all, of the big difference. The one is the so-called "hedonic price indexing" of computer investment. Broadly speaking, it converts soaring "computer power" into steeply falling computer prices which, in turn, swell real gross domestic product growth. The highly important second difference is in the treatment of software. It is global practice to treat software spending as business expense, and as such it has no relevance to GDP growth. But last year, US statisticians decided to treat these as capital investment, and that happens to add also heavily to real US GDP growth.
To give an idea of these effects: Between end-1998 and mid-2000, business spending on computers rose in current dollars by Dollars 23.8bn to Dollars 114bn, accounting for 2.4 per cent of nominal GDP growth. But the hedonic deflator puffed this modest amount up to a massive increase by Dollars 120bn chained dollars to a total of Dollars 299bn. As to software spending, totaling now Dollars 226bn, it added another Dollars 95bn to GDP growth in chained dollars. Together, the two components swelled real GDP growth by Dollars 215bn, as against actual Dollars 23.8bn in current dollars. Note this: together, the two statistical "extras" accounted for fully one-third of US real GDP growth.
The Bundesbank estimates that if nominal spending on computer equipment would in Germany be deflated by the same price index as that used in the US, real IT investment would have grown 27.5 per cent a year since 1991, rather than 6 per cent recorded.
Finally, this raises the question of whether the American measuring method makes good economic sense. The answer is: no, not at all. The reason is that it creates more and more fictitious real GDP growth and in line with it more and more fictitious productivity growth, but this is growth that occurs without any correlated income growth. But is it not income growth that matters? Since the third quarter of last year, growth in real disposable income has been rapidly falling behind GDP growth. In June, it was zero and in July 0.1 per cent, and that in a booming new economy.
Kurt Richebacher 91 Boulevard de la Croisette F 06400 Cannes, France
Copyright © The Financial Times Limited