(published by United Press International, February 7, 2005.)

Labor productivity figures announced by the Bureau of Labor Statistics Thursday were poor, surprising the market, which since 1997 has believed the United States to be in a new era of rapid productivity growth. Multifactor productivity figures, announced by the same agency Tuesday, drew little attention because they were two years old, but clearly indicated that the era of stellar labor productivity statistics may have ended.

Labor productivity in the non-farm business sector was up only 0.8 percent at an annual rate in the fourth quarter of 2004, but that was after two years of very good productivity growth, so at first sight there didn’t seem much to worry about. After all, the annual average productivity growth in the 2 years 2003-2004 was a superb 4.0 percent, far in excess of any requirement for a potential ”productivity miracle.” If that figure were to hold up over the long term, the U.S. economy’s performance would be stellar indeed, and all problems relating to budget deficits, trade deficits and the financing of social security would melt magically away.

However, labor productivity does not tell the full story; there are more inputs into a modern economy than labor. Multifactor productivity, including both labor and capital as inputs, is a much closer proxy for the real economic gains caused by technological and organizational change.

Multifactor productivity, calculated by the BLS with a substantial lag, Tuesday showed a positive tendency for 2002, the latest figures published, up 1.9 percent for the year, a figure well above its long term trend rate.

However, over the 10 years 1992-2002, when compared with the longer term trend, the figures tell a different story. The 1990s did not in fact bring a secular increase in U.S. productivity, as has been widely and triumphantly claimed, which might justify the sky-high stock valuations of the decade, the general optimism about U.S. economic prospects and the incessant sneering at supposedly “sclerotic” economies such as France and Germany. Instead, the rate of increase in multifactor productivity in the business economy actually SLOWED during the decade. From a rate of increase of 1.30 percent in the decade 1982-92, multifactor productivity growth almost halved in the decade 1992-2002, to a rate of 0.72 percent per annum.

For those of us who are capitalist fundamentalists, believing that President Ronald Reagan with his 1981 tax cut, reform of Social Security and robust free market policies changed something fundamental in the U.S. economy, but that subsequent Presidents George H.W. Bush, Bill Clinton and George W. Bush have done very little to it one way or another, except maybe making it marginally worse by bringing in a flood of low paid immigrant labor and encouraging a culture of “get it now” and never mind the funny accounting, this is a deeply satisfying morality play. Multifactor productivity growth, consistently high at 2.15 percent per annum in 1948-73, dropped like a stone in the semi-socialist Nixon/Ford/Carter years, averaging 0.11 percent per annum in the decade 1972-82. Then after the Reagan Revolution liberated the U.S. economy it rebounded sharply, to 1.30 percent per annum in 1982-92, before dropping back again under his flaccid successors to a mere 0.72 percent. The only unsatisfactory thing about this “morality play” is that the rate of increase in the miracle 80s was lower than that in 1948-73 but hey, every jewel has a flaw!

The difficulty then is how to reconcile this morality play with the alternative Clintonista New Democrat morality play that appears from the labor productivity figures. Labor productivity in the decade 1994-2004 increased by 2.91 percent per annum (or 2.2 percent per annum in the decade 1992-2002, to be comparable.) Even more excitingly for acolytes of the New Economy, the growth rate in labor productivity appears to be accelerating; in the five years 1999-2004 it was 3.70 percent per annum. Compare this with 2.28 percent per annum in 1982-92, 1.26 percent in 1972-82 and 3.31 percent in 1948-73 and, if you don’t think about it too closely, you see a disaster under Nixon/Ford/Carter, a modest recovery under Reagan, and a further spurt in growth under Clinton and (teeth grinding at this point) George W. Bush. Still, under this scenario, Baby Boomer presidents are good for the economy, U.S. productivity has accelerated to a new level that justifies fancy stock price valuations (although annoyingly, productivity growth is not yet reliably higher than in 1948-73) and Bill Clinton, in particular, was a genius.

The problem of course is that these two “morality plays” are incompatible with each other, and we thus need to examine why, and which one is a better representation of reality.

There are two underlying factors that explain the differences between the time series. First, there is the matter of hedonic pricing. The Bureau of Economic Affairs in 1996 changed the way it calculates the Consumer Price Index and other U.S. price indices, to reflect the “hedonic” benefit we all receive from having ever faster computer chips, and so on. On close inspection, much of this “hedonic benefit” is illusory – even if Moore’s law doubles the speed of computer chips every 2 years, our hedonic benefit from our PC doesn’t double every two years, it increases quite slowly. Most of us are still using our computers for much the same things in much the same way as 5 years ago – certainly we aren’t getting 4 times the benefit from them, as hedonic pricing says we should be. Furthermore, only the sectors that have improved are included in the hedonic calculations; there is no negative hedonic benefit produced by the time we have to wait around in computerized call forwarding systems, or the unnecessary costs built into our cars by environmental legislation.

The total effect of hedonic pricing has been shown to be about 1 percent per annum, and has been tending to increase as information technology has become a more important part of the economy. However much of the hedonic benefits are real, hedonic pricing has since 1996 deflated the consumer price index, therefore inflated real Gross Domestic Product calculations, therefore inflated productivity calculations, compared with pre-1996 figures. Multifactor productivity is less affected, because whereas labor is a “real” input (hours) capital is a money input and hence equally affected by fiddling with the price indices, so the effect on multifactor productivity of hedonic pricing must be at most half the effect on labor productivity.

Suppose half the hedonic effect is real, so that from 1996 inclusive productivity figures are ½ percent “too high.” Then equally, pre-1996 figures must be “too low” by an equivalent amount in the early 90s, but are probably more or less accurate before 1973. Suppose the effect on multifactor productivity is about 1/3 of that on labor productivity. Then the figures look as follows:

— Multifactor productivity growth averaged 2.15 percent in 1948-73, dropped to 0.28 percent in 1972-82, rose to 1.47 percent in 1982-92, and dropped back to 0.68 percent in 1992-2002. Same picture as before, in other words, but with a slightly less pronounced drop between the 60s and the 80s.

— Labor productivity growth, on the other hand, shows a changed picture. 3.31 percent in 1948-73, it dropped only moderately, to 1.76 percent in 1972-82, rose to 2.78 percent in 1982-92, then if anything dropped back a bit, to 2.51 percent in 1994-2004, although the 1999-2004 period, at 3.20 percent, is rather better. The Clintonista/New Democrat theory is dented, but hey, maybe W made the difference!

The other factor explaining the different tracks of labor and multifactor productivity is capital. Capital productivity can be tracked just like labor productivity, but shows a very different path. It peaked as long ago as 1966, and has since been showing a steady decline, which decline accelerated during the 1990s. In terms of growth rates, capital productivity grew at 0.31 percent per annum in 1948-73 (a peak in 1966 followed by a decline), shrank at 1.95 percent per annum in 1972-82, recovered at 0.45 percent per annum in 1982-92, and shrank again at 0.63 percent per annum in 1992-2002, so that in 2002 it was almost 25 percent below its 1966 peak.

To interpret these figures, we need to think about what was happening in the economy at the time. Capital productivity is the efficiency with which capital is used in the economy. In 1948-66, that efficiency was steadily increasing, as a stable United States recovered from the war and modernized its plant, retiring capital assets left over from before 1929. After 1966, the efficiency of capital usage began to decline, as a surge in inflation, the economic diversion of assets to the Vietnam War, and pollution control measures such as the Clean Air Act of 1970 took productive assets out of service. Then in 1973 the price of a key input – petroleum – suddenly quadrupled, making a huge range of capital assets suddenly obsolete – like the gas guzzling 1973 Cadillac I was able to buy for a song six years later! Gradually through the 1980s, the economy adapted to higher oil prices, but all the time the capital intensity of the economy was increasing, so capital productivity never regained its 1966 level.

Then in the late 1990s and following 2000, we had a surge of capital inputs into the economy. First, the stock market rose towards Pluto, so that any new venture could get financed, no matter how hopeless. Then, before that bubble had been dealt with, interest rates were cut aggressively to a short term interest rate well below zero in real terms, igniting a blaze of consumer “investment” in housing, automobiles, boats and other fixed assets. Consequently, the capital efficiency of the economy dropped to a level far below that ever seen in recorded history, though I would guess that of 1932, when assets were being shuttered or used at half capacity, may have been lower. There was a record level of economic waste. For confirmation of this, the BLS itself points out that in 2001 and 2002 the capital/labor ratio in the U.S. economy grew more rapidly than in any year since the deep recession of 1982, by 6.6 percent in 2001 and 5.3 percent in 2002.

We have only 2002’s figures for multifactor productivity, but at this point we can make a pretty shrewd guess about what 2003 and 2004 will show. The surge of cheap money and artificial consumer asset investment already visible in 2002 grew to extraordinary levels in 2003 and remained very high in 2004. Thus the high figures for labor productivity growth seen in 2003 and 2004 were simply an artificial surge, produced by increases in capital input that may well have exceeded 2001’s record. True productivity, multifactor productivity, increased very modestly, or even declined.

We can also see what’s likely to happen going forward. Interest rates are now rising significantly, the housing bubble is beginning to cool, and the 2003-04 tax benefit for capital investment has ended. Consequently, in 2005 capital inputs to the U.S. economy will slow their increase or even decline. If multifactor productivity continues to grow at the slow but constant rate it has exhibited since 1992, we are likely to see a very sharp decline in the published labor productivity increases, or even a fall in reported labor productivity as the economy corrects itself and begins to work off the excessive capital investment of 1996-2000 and the excessive housing investment of 2001-04. This will cause much wailing and gnashing of teeth among the pundits, and in the administration, but it will represent nothing more than a long overdue correction to the imbalances of recent years.

Very slowly increasing or even declining labor productivity has other implications. On the bright side, the monthly jobs figures will cease being unexpectedly poor, and will become more closely aligned with expectations generated by other economic data. On the negative side, this alignment will be achieved through deterioration of the other data. Economic growth itself will be very slow, or even negative, while corporate profits are likely to drop sharply, as productivity gains from a U.S. workforce suddenly become very difficult to attain. Needless to say, this should produce a deep and long-overdue fall in the stock market, which itself will correct imbalances that have been present for far too long.

There never was a productivity miracle. There was only a recovery, from the imbalances caused by the 1973 oil shock and the early 1970s environmental legislation, which took productivity growth back close to its historic level. That recovery took place in the 1980s, not the 1990s. It was followed by a surge in capital investment, caused by excessively cheap money, first in the tech sector and then in housing. That surge has now ended.

The years ahead, in terms of reported labor productivity growth, economic growth in general, and the stock market are likely to be grim ones indeed. If you must blame anyone for this, blame Alan Greenspan.

-0-

(The Bear’s Lair is a weekly column that is intended to appear each Monday, an appropriately gloomy day of the week. Its rationale is that the proportion of “sell” recommendations put out by Wall Street houses remains far below that of “buy” recommendations. Accordingly, investors have an excess of positive information and very little negative information. The column thus takes the ursine view of life and the market, in the hope that it may be usefully different from what investors see elsewhere.)

This article originally appeared on United Press International.