University of Chicago Professor Steven Davis, at an American Enterprise Institute seminar Wednesday, presented an examination of the Bureau of Labor Statistics figures on job creation and destruction. Most interesting was a difference between the early 1990s recession and recovery and that following 2000: In the early 1990s there was a temporary increase in the rate of job destruction while after 2000 the rate of new job creation dropped sharply and stayed down. This may indicate a structural problem, bad news indeed for the U.S. economy.
As Davis and the discussants Jared Bernstein of the Economic Policy Institute and Diana Furchgott-Roth of the Hudson Institute agreed, one of the great strengths of the U.S. economy is its fluidity. Approximately 8 percent of the labor force move jobs each quarter, which gives the economy exceptional flexibility in adapting to technological and organizational change. This turnover is significantly higher than in Europe; it means that the “headline” figure of around 500,000 jobs created in each quarter is actually a net of 8.5 million jobs being created and 8 million being lost.
Davis gave the example of WalMart, which in 2005 had around 1 million employees and 3,000 stores, having grown from zero in 1962. Research has shown that each time a new WalMart is opened, around 100 retailing jobs are created in the first year, the net effect of 150-300 jobs at the WalMart and jobs lost at its competitors. Then over the next 5 years a further 50 jobs are lost in competitors, and 20 jobs are lost at wholesalers, for a net gain per WalMart store of 30 jobs.
While on the subject of WalMart, Bernstein quoted some currently unpublished research that suggests that WalMart, while having no effect on wage levels in rural areas, tends to depress retailing wage levels in urban areas. This is an important data point towards the thesis that heavy immigration depresses unduly the living standards of less able U.S. citizens. Immigrants both legal and illegal are far more common in urban than in rural areas, so the competition for low end jobs is correspondingly more ferocious and WalMart’s wage-depressing effect more severe.
However, the most interesting finding of Wednesday’s presentation was that the private sector job creation rate (new jobs divided by total workforce) which averaged around 8.3 percent per quarter throughout the 1990s, with small fluctuations and no significant trend, dropped sharply in 2000-01 to around 7.1 percent and has remained around that level since, bottoming out at 6.9 percent in early 1993 and recovering slightly thereafter. Conversely the job destruction rate spiked to 9.4 percent for a brief period in 1991 then settled at 7.6 percent for the rest of the 1990s. It spiked slightly in 2001, but only to 8.2 percent, and has since dropped further, to a low around 6.7 percent in the middle of 2004.
The panel had no very satisfactory explanation for these changes, particularly the continued low level of job creation after 2000. Bernstein suggested that the bursting of the tech bubble after 2000 has had a long term effect, while Davis postulated that the rapid rise in productivity that he believes to have occurred since 2000 is restraining job creation (but you then need to explain the rise in productivity.) Other possible explanations are the continuing huge U.S. trade deficit and/or the rise in job outsourcing (which accounts for only about 300,000 jobs per annum, not enough to explain the difference.)
A further difference between past recoveries and this one has been the lackluster growth in wages, illuminated by the Bureau of Economic Analysis Friday, announcing wage growth of 0.7 percent in the last 3 months, lower than expected. Even though inflation has increased since 2000, wage growth has dropped steadily, from over 4 percent per annum throughout 2000 to below 3 percent today, with the decline continuing even after economic “recovery” began in 2002. This low wage growth ties in with anemic job creation to suggest that something is seriously amiss in the U.S. labor market.
As readers of this column will know, I don’t believe in the supposed “productivity miracle” and in general have problems with the happy-face approach to the U.S. economy so beloved by AEI and the George W. Bush administration. In 1996, the BLS changed its calculation methodology for price indices to a “hedonic” basis, under which the very rapid gains in nominal capacity in the tech sector were netted out from raw price data, producing a price series that rose more slowly than simple price movements would indicate, the difference being estimated at about 0.8 percent per annum.
Whether or not you think this change reflects reality appropriately (I don’t) the undeniable fact is that U.S. price indices from 1996 on have understated U.S. inflation (and correspondingly inflated real U.S. Gross Domestic Product growth and productivity growth) when compared either to previous U.S. statistics, or to statistics from European and Asian countries that did not adopt the hedonic approach. Netting out this difference in methodology removes the greater part of the rise in productivity growth since 1995, as well as the faster apparent rate of productivity growth in the United States compared to Western Europe.
The full explanation for the job lag comes into focus when you look not at labor productivity but at multifactor productivity, and in particular at the productivity of capital, which has dropped by more than 20% since the early 1990s – in other words, you now get 20% less output for the same amount of capital as you did in 1992. Combine this with another statistic: real M3 money (deflated by the Consumer Price Index) in the United States grew by 5.51 percent per annum in the decade 1995-2005, compared with 0.27 percent per annum in the previous decade. The result has been first a huge and unprecedented stock market boom, followed by a lengthy period in which real short term interest rates were negative and real long term government bond yields barely positive, while the stock market remained far above the valuations that prevailed in the 1985-95 period. Capital has been exceptionally cheap – and therefore exceptionally badly allocated.
Very cheap capital affects long run job creation and destruction rates in several ways. On the creation side, cheap capital causes the development of a capital assets bubble, the assets created by which are then discovered to be far in excess of what’s needed by the economy, producing a backlash that causes a prolonged unwillingness to undertake new capital investment. If cheap capital continues available, it is diverted into housing and retail construction, largely unproductive investment, also uneconomic in the long run, that buoys the economy in the short term at the expense of raising the cost of living by a “stealth” process, as housing becomes an ever greater part of everybody’s budget. The result is to prevent the normal post-recession rebound in the creation of new businesses that add real jobs, while the cheapness of capital compared to labor spurs the substitution of capital for labor and depresses new job creation and wage growth.
On the destruction side, the Schumpeteran process of creative destruction by which non-viable businesses are weeded out is suppressed. Poor businesses find it easy to get new loan funding, since banks are awash with loan funds, and so survive longer than they should. Job destruction rates are thus also depressed well below their natural level.
While interest rates remain low, net job growth appears more or less satisfactory, as both job creation and job destruction are dampened below their natural levels. The long term damage to the economy appears only when interest rates rise towards a more normal level. Agonizingly slowly, this is now occurring. The Fed has raised the Federal Funds rate from 1 percent to 3.25 percent over a period of more than a year. This has had little effect on the economy because inflation has also risen, long term bond yields have remained low, and risk “spreads” have narrowed to unprecedented levels.
The stability in the long term bond rate has been a puzzle to Fed chairman Alan Greenspan; it should not be. Apart from the well publicized activities of Asian central banks in the Treasury bond market, it is very probably caused by the activities of hedge funds borrowing short and lending long. The size, rapid growth and aggressive investment policies of the hedge funds are due to the vast oversupply of cheap capital over the last decade. The popularity of the strategy is caused by the relative lack of investment opportunities for such large pools of money and the continuing premium of long term interest rates over short term rates.
However, as the short term rate finally approaches the long term Treasury bond rate, the “gapping” profits available by borrowing short and lending long will disappear. This will cause the hedge funds to unwind their bond positions, at which point bond prices will fall sharply and long term yields will lurch higher. This will occur with one of the next few ¼ percent rises in the federal Funds rate, probably (for psychological reasons) that which takes the rate to 4 percent, due on the current schedule November 1. This increase will make it clear to even the doziest trader looking at the 10 year Treasury bond yield around 4.3 percent that the profit in this trade is about to disappear. Since a rise in bond yields would give the trader a (large, because leveraged) capital loss, he will unwind his position, as quickly as possible, thus selling bonds into a falling market. Panic will ensue.
Once long term interest rates rise, the “destruction” side of Schumpeter’s equation will come fully into play, and job destruction rates will soar, far above the currently anemic job creation rates. Job creation, meanwhile, will face a double headwind of suddenly more expensive capital, combined with a huge overhang of capital investment that is still proving far less profitable than had been anticipated. It will thus remain depressed or even decline further.
The result will be a lengthy period of rapidly rising unemployment. For this, the unemployed and destitute should blame Alan Greenspan, but they won’t – by the time the full force of the downturn hits, he will be well into his January 2006 retirement.
-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.