2 February 2004, Doug Henwood, The Guardian
In the late 1990s the US was famous across the globe for its New Economy. Computers had unleashed a productivity miracle, recessions were relics of a transcended past, ideas had replaced things as the motors of economic life, the world had become unprecedentedly globalised, work had evolved into something deeply meaningful and mutual funds had put an end to class conflict.
That miracle did not quite work out as hoped, but now the US economy is working a new kind of miracle: clocking near-Chinese rates of GDP growth while producing hardly any new jobs. In the third quarter of 2003, the economy grew by 8.2% while employment rose 0.1%.
In the fourth quarter Canada, despite being about one-eighth the size of its southern neighbour, produced more new jobs than the US - not in percentage terms but in absolute numbers.
Strangely, the two miracles are not unrelated. The economic heart of the 1990s miracle was the productivity revolution. There is no doubt that the official productivity statistics shook off their 20-year-old torpor in the mid-90s and accelerated significantly. But what does that mean?
There are at least two ways to approach that problem: the technical and the philosophical. Let us take the technical approach first. Labour productivity measures real output per hour of labour. There are serious problems in estimating both the numerator and denominator of the productivity equation.
The labour inputs to the productivity calculations are not hours worked but hours paid, as reported by employers to the bureau of labour statistics (BLS). That is no small distinction.
One of the undisputed stars of the productivity revolution is the huge retail group Wal-Mart, which has repeatedly been sued for requiring its "associates" to work long after they have clocked off for the day. The BLS does not have firm estimates of how long management work weeks are; essentially it makes a guess.
At Wal-Mart, many store managers work 60- or 70-hour weeks, but the productivity statistics assume far less. The same goes with the computer industry. The BLS assumes that executives in the hi-tech sector work normal 35- or 40-hour weeks. To anyone in the industry, that assumption is hilarious.
There are plenty of problems with measuring output, too. Take, for example, the computer. If today's $1,000 PC is twice as fast as last year's then, according to conventional economic logic, its "real" value is twice the 2003 model's. Who knows if that is true? If you are typing letters and sending email, the speed increase hardly makes a difference.
But to the official US accountants, the matter is settled. That logic ripples throughout the statistical apparatus.
According to economist Robert Gordon, one of the few mainstream sceptics on the productivity revolution, most of the acceleration in productivity has occurred in the manufacture of computers and similar devices. Gordon's conclusion is controversial but even enthusiasts concede that productivity in heavy computer-using industries - such as finance, business services and communications - has either been increasing very slowly or declining.
But the technical argument needs a broader context. The point of increasing productivity is, or should be, to improve our material standard of living and make our lives a little easier. The American productivity miracle has done neither. Even at the peak of the boom, more than 60% of respondents to a Business Week poll said the miracle had done little to raise their incomes or improve their job security.
Over the longer term, productivity gains have done little to ease the work effort, either. A worker paid the average manufacturing wage would have had to work 62 weeks to earn the median family's income in 1947. In 2001, he or she would have had to work 81 weeks. So, despite the fact that productivity was up more than threefold over the period, the average worker would have to toil six months longer to make the average family income. Americans work more hours a year than just about anyone else on earth, and things only got tougher in the 1990s. Some revolution.
During the boom, there was plenty of irrational exuberance and spare cash to spice things up a bit. American workers saw some real wage growth in the late 1990s, a welcome change from the previous 20 years of stagnation alternating with decline.
With the bursting of the bubble and the emergence of a jobless recovery (with no real wage growth), the underlying reality of the productivity revolution has been revealed: wage squeezing and extra pressure in the workplace.
The essence of the US economic model was nicely encapsulated in a 1997 article by New York Times reporter Alan Cowell. In the midst of lecturing the Germans on the need to give up their long weekends and long vacations, Cowell recommended that they adopt the American approach, which he defined as "working longer for less."
Doug Henwood is the author of After the New Economy (New Press)
