Every year or so Bear’s Lair examines productivity, supposedly the engine of a remarkable economic revival in the United States in the last decade, and a rationale for ever rising stock prices. The publication March 24 by the Bureau of Labor Statistics of multifactor productivity trends leads once again to the inescapable conclusion: the productivity miracle is entirely a statistical artifact and in reality U.S. productivity growth is not at all special.

The BLS had been wholly silent on multifactor productivity for more than 2 years, and then issued 2 years’ statistics at once, together with an explanatory note revising their industry code system and casting 39 years of productivity history, from 1948-86, into the limbo of “incompatible” historical data, that they no longer presented. Nice try guys – this removed from the record the period from 1947-73 when labor productivity growth was somewhat higher than at present and multifactor productivity growth much higher. Since for the U.S. economy as a whole the BLS productivity figures from 1987-2002, available on both series, were very close I take it that the overall data are not incompatible, even though the industrial classifications are different.

Multifactor productivity statistics aren’t exactly treated with great fanfare by analysts either, partly because they contradict the happy-talk that recent administrations, the Fed and Wall Street emit about the new productivity era, and the justification it provides for excessive trade deficits, inflated stock prices, huge budget deficits and spiraling money supply growth. Publication of the March 24 data, for example, didn’t even register on Economy.com, the economy wonk website that devotes considerable analytical energy to statistics emanating from such economic powerhouses as the Kansas City Fed. In terms of the data’s publicity, one is irresistibly reminded of the notice about Earth’s destruction in “Hitchhiker’s Guide to the Galaxy,” which was on display in a basement on Alpha Centauri, in a disused lavatory guarded by a sign “Beware of the Leopard.”

The multifactor productivity statistics for 2003 and 2004, issued in March, were quite favorable. Multifactor productivity in the non-farm private business sector increased 2.8 percent in 2003 and 2.9 percent in 2004, the highest rates of increase since 1983, when productivity was recovering from the 1979-82 recession. (Productivity statistics are reported for the private business sector and the private non-farm business sector; I shall concentrate herein on the latter, though the statistics including farming show similar trends.) Not only did labor productivity increase rapidly in those years, by 4.0 percent and 3.4 percent, but capital productivity, which had been declining since the early 1990s, also increased, by 1.3 percent in 2003 and 2.3 percent in 2004. One can hypothesize that in those years the huge corporate over-investment of 1996-2000 began to be absorbed, particularly in such areas as telecom bandwidth.

Nevertheless, over a longer period the picture is more mixed. Multifactor productivity increased in 1994-2004 at an average rate of 1.32 percent per annum, up from a mere 0.89 percent per annum in 1992-2002. This is a relatively favorable rate; it is indeed the highest 10 year average since the period ending in 1973. However, it does not compare with 1948-73; multifactor productivity increased by 1.90 percent per annum over the 25 year period from 1948-73. Thus even though multifactor productivity, including the two recent favorable years, has been rising faster than in the late 1970s, the 1980s and the early 1990s, its rate of increase is still only two thirds of that in the 25 years following World War II.

In fact, even this picture is somewhat too rosy. In 1996, the BLS altered its calculation of price indices, to reflect “hedonic” pricing, by which improvements in quality, or especially in speed and capacity in the electronic sector, would be recorded as if they were price reductions. As I have discussed many times, this is largely spurious; the doubling of chip size every two years according to the well known “Moore’s law” produces nothing like a doubling in the functionality of that chip, and so hedonic pricing greatly understates “true” price increases. Whatever the reality, “hedonic” pricing is not used in Europe, nor was it used before 1995, so when comparing U.S. “real” output figures with those overseas or from an earlier period it needs to be corrected for.

In productivity figures, hedonic pricing affects labor productivity, because it inflates “real” output while keeping the number of hours worked constant. It does not affect capital productivity; because both capital and output can be expressed in money terms, the price deflators are the same. It affects multifactor productivity partially, the proportion by which it affects it depending on the weightings of labor and capital in the multifactor productivity calculation. To get an idea of its effect, you can look at the productivity figures produced by the Organization for Economic Cooperation and Development, which attempt to normalize between countries (their corrections for hedonic pricing may or may not be accurate, but they are at least the work of experts in this field.)

According to the OECD, U.S. labor productivity increased by 2.54 percent per annum in 1996-2004, while the BLS reported the increase at 3.00 percent per annum; we can roughly assume therefore that labor productivity figures from 1996 on are artificially inflated by about 0.46 percent per annum compared with earlier years, or compared with other countries. For multifactor productivity, the correction would be correspondingly less, perhaps 0.2 percent per annum. Thus the “true” increase in multifactor productivity in 1994-2004 was about 1.1 percent per annum, and in labor productivity was around 2.3 percent per annum – almost exactly the levels of 1982-92.

Productivity hasn’t lagged in the last decade, but there’s been no “miracle.”

Comparing the United States to other OECD countries in the OECD, U.S. multifactor productivity growth of 1.1 percent per annum in 1994-2004 was better than that for the same period of Austria, Belgium, Canada, Denmark, Italy, Japan, the Netherlands, New Zealand and Spain, but not as good as Australia, France, Germany, Greece, Ireland, Sweden or the United Kingdom. (French and German multifactor productivity growth was only a little faster than U.S. growth, but it was faster, not slower as almost all of Washington believes.)

U.S. labor productivity growth in the same period was also around the middle of the pack, very close to G7 average growth. There is thus no “Euro-sclerosis” and very little “Japan disease” – Japan did much better than the United States in the 1980s and only modestly worse in the 1990s so that over the 25 year period to 2004 its labor productivity grew at 2.7 percent per annum compared with 1.7 percent per annum in the United States. For the period for which OECD multifactor data is available for both countries, 1985-2004, Japanese multifactor productivity growth was also faster than U.S. growth.

Since the beginning of 2005, U.S. labor productivity has begun to slow markedly, as this column forecast in February 2004 would happen as interest rates began to rise, so that the rapid labor productivity growth of 2002-04 is more clearly seen as a blip rather than a trend. In any case, since capital per unit of output remains well above its level of the middle 1990s, it seems unlikely that the middle-of-the-pack behavior of U.S. multifactor productivity will have improved significantly. If and when recession hits, U.S. multifactor productivity is likely to drop sharply, as it did in the recessions of 1974-75, 1979-82, and 1991-92 (it was flat over the year in the modest recession of 2001.)

The lack of a “productivity miracle” has a number of implications:

• The U.S. balance of payments deficit is very troubling; if indeed desperate foreigners are stuffing their money wildly into the United States they are completely irrational to do so, and must be expected to stop sometime.
• There is no law of nature that says the United States will always innovate rapidly enough to prevent job losses through outsourcing. Thus heavy immigration, if accompanied by a recession, is likely to lead to startlingly high unemployment as in parts of Europe.
• The federal budget deficit needs to be brought under control as quickly as possible, since it will spiral out of control in a recession. Social Security and Medicare are un-financeable in the long term, since productivity growth is central to the economic forecasts underlying government projections.
• The United States is economically in a multi-polar world, and France and Germany are serious economic partners, not basket cases. There are more successful economic partners and competitors in Asia: Japan and South Korea, but also doubtless China and India (who are not OECD members.) Also Ireland, but that may be a special “catch up” case that will not extend long term.
• Mexico, astoundingly, had negative labor productivity growth in 1991-2004, the period for which figures are available – so much for the supposedly revolutionary effects of NAFTA. Latin America is thus economically and politically a big problem, not an opportunity for the United States.
• As inflation rises, and interest rates rise, capital productivity in the U.S. will continue to increase as the capital to labor ratio drops. Labor productivity, on the other hand, is likely to decline, particularly if there’s a recession – good news for the unemployed and very bad news indeed for corporate profits.
• The surge in corporate profits of the last few years has been due to low interest rates and easy capital availability. These conditions will reverse, and profits will drop sharply, to a level below rather than above their long term trend. Wall Street has not discounted this.
• The stock market remains hugely overvalued, since the theories that have been upholding it are hogwash. The “chartist” prediction of a Dow Jones Index level of 2800 that this column made a few weeks ago is thus confirmed by this “fundamental” analysis.

As I said, it’s no surprise that the BLS, the George W. Bush administration and the bulls of Wall Street want as little attention as possible paid to these statistics.

-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.