In a major embarassment for the editorial page of the Wall Street Journal, the Journal op-ed page carries an opinion piece by economist Edmund Phelps which in content is something like a biochemists raging against evolution -- but with the evolutionary theory completely mistaken. Imagine, for example, if the Journal opinion page ran an editorial by a biologist which dismissed the theory of natural selection because the author believed that the theory of natural selection was a theory of the inheritance of acquired characteristics and was incompatible with modern genetics. The Journal and that biologist would be laughed off the stage of serious scientific commentary. And this is just what should happen to the Wall Street Journal and Edmund Phelps when it comes to economic commentary.
False Hopes for the Economy -- And False Fears
By EDMUND S. PHELPS
The towering investment boom of 1996-2000, now over, was a huge lift -- and not just for wages, profits and employment. There may never been a time in U.S. history when innovative activity was more engaging and working life more rewarding. The task now is not to create artificially a replacement boom by assorted stimulants, which would be hard at best to do anyway. It is to maintain and improve the vitality and creativity of the economy so that high performance is the norm rather than the exception.
The public, though, is not in a frame of mind for talk of fundamental reform. The loss of the boom has left people uncertain, uneasy. They know that after the 1920s boom came the Great Depression. Some know that the '50s boom was followed by the '60s boom. There is a cacophony of opinion on the economy's prospects and recent policy interventions. Conflicts among dormant theories have broken out in the open. We cannot get to the deep issues until we dispel the more unlikely Cassandra scenarios and Pollyanna theories that plague present discourse.
The False Hopes
The most mindless optimism speaks ritually of "recession" from which we can expect "recovery" -- in labor, product and capital markets. But what recession? Standard interpretations of the usual charts estimate that in 1995-96 the economy was at or close to its long-run normal state -- with monetary disturbances in abeyance and no big nonmonetary disturbances either. (The core inflation rate was steady, averaging the same rate as in 1993-94.) Of course, what is normal is always evolving. Yet, impressively, the period's unemployment rate (5.5%), the share of GDP going to labor (65.8%), and stock-market wealth relative to the GDP (about 115%), were on the whole not far from their levels in two other pretty normal periods, 1987-88 and 1971. In the next four years the economy boomed, posting records in all these respects and others. Since 2000 it has fallen back: labor's share to its 1995-96 level, the stock market to its 1997 level and the unemployment rate -- at 6.1% -- to its 1994 level. So we're more or less back to normalcy. It was the boom that was abnormal.
Yet some take the view, many unconsciously, that there is no normal state to which the economy tends to return, even given current institutions, tax rates and welfare rules, and entrepreneurs' spirits. The normal level of employment, they say, is a statistical artifact that moves up if the government takes any of the straightforward measures to raise the level of economic activity -- if the Fed gives support with low interest rates and ample liquidity or Congress does by spending more money.
Behind that view is the "aggregate demand" fallacy: the government can deliver high employment simply by pumping up high aggregate demand -- by easier money or bigger budgetary deficits. A.W. Phillips sensed the mistake here, arguing in 1958 that a pumped-up employment level typically brings a higher rate of inflation (illustrated with his famous curve). Milton Friedman and I corrected Phillips, explaining in 1968 that, to keep on doing the pumping trick, the rate of inflation would have to be driven higher and higher -- until the payments system broke down or the policy was halted. Unemployment cannot be kept forever below its "natural" rate or path, along which the rate of inflation is neither rising nor falling. In natural rate thinking, a tax cut boosts employment, as supply-side economists knew, only through its effects on incentives to work and to save, the cost of labor and the cost of capital -- not its effect on demand.
There is much overconfidence too that depressions cannot happen, now that proper monetary policy is understood. For one thing, natural rate thinking has a role for aggregate demand. ("It's a question," to paraphrase Humpty Dumpty, "of who is the master, that's all.") For actual employment to home in on the natural employment path, market mechanisms have to drive demand up or down, as needed, to tailor it to the GDP producible at natural employment. But economists don't know accurately where the natural rate is, so how can the market? The gyrations of the dollar show the markets struggling with uncertainty. If the natural rate is again 5.5%, its mid-'90s norm, the fall of unemployment to the natural level would seem to require that exports get back to the share of GDP they had in 1995-96. That would appear to require the dollar to fall to its mid-'90s level -- about $1.35 to the euro.
Moreover, bad things can happen to the natural unemployment rate. If CEOs, seeing more difficulties or risks for profits ahead, cut back budgets for new business assets, the natural rate will go up. There can be structural slumps -- with or without disinflation.
The False Fears
One of the most unreasoning fears, yet pervasive, is the nightmare of interwar Austrian cycle theory: "over-investment." Wall Street gurus say that the investment boom, with its extra purchases of capital goods (and extra jobs to make them), caused capital stock to get "ahead of itself," with the result that capital-goods spending (and output) will for years be below its normal path -- to give wear & tear and obsolescence the chance to work off some of the excess capital and to permit the economy to grow into the rest. If that is right, employment will also stay for years below its normal path. A slump after the boom.
But, first, what if only a small part of '90s investment is proved to be overinvestment? No slump should follow an investment boom inspired by new expectations of some outsize future productivity gain if they prove correct -- a sort of rational boom. The capital stock surges up only in proportion to the expected rise of productivity, and when the productivity gain is realized annual investment drops back to its normal ratio to (increased) output -- never below its trend-path -- and employment falls back to normal.
On overinvestment, the Austrians really had no argument. Imagine a sudden addition of capital was heaped on the world. The growth rate of capital stock would have to drop. But, for that, it would be more than enough that gross investment not increase as a ratio to the (increased) capital stock; it wouldn't require a decrease of investment or of jobs, let alone collapse. Indeed, a model of mine says world markets would react to the addition of capital with a sharp drop of interest rates, a skyward jump of capital goods prices, and an immediate lift of real wages and jobs! The saving point for the Austrian contention is that if the overinvestment is concentrated in the U.S., the U.S. alone cannot achieve the full drop in interest rates required because it is just one part of the world economy. But that won't wash if overinvestment was not much less important in Europe and Asia.
Before finishing with overinvestment, consider a related fear. It is that, with the capital stock pushed far higher in the last decade than will be justified by productivity gains in this one, expected rates of return on business assets must have fallen, so the rates of interest required for unemployment to attain and stay at its natural rate must have fallen similarly. The worry is that, if a recession were to arise, the economy would be faced with the need for interest rates of close to zero, even below zero. Should that come to pass, vast parts of the financial system would go under, unable to provide a return competitive with that of good old cash. Then disinflation would pick up and lead to deflation, intensifying the problem.
But the present overhang of '90s overinvestment appears far too small to have caused a decline in rates of return large enough to put us close to that interest-rate trap. From 1996 to 2000, the extra investment going on was 2% of GDP. So the increase in the capital stock in those five years was about 10% of annual GDP. That would likely have boosted the capital-GDP ratio by about 5% -- from, 2.0, say, to 2.1. Hardly enough to push us to the edge of the interest-rate trap. Besides, while we never want to get too close to the trap, fear that the risk of falling into it has greatly increased seems overblown. Medium- and long-term interest rates today are like those in the '50s.
The question du jour is what the information revolution means for jobs now. Since the boom ended, the result is not just that the extra jobs created in the capital-goods sector are gone. It is that the newly installed software embodying the new information technologies seems to have opened a Pandora's box of cost savings that are now destroying old jobs in the consumer goods sector -- in securities, airlines and communications.
Mistakes have been made on productivity by several economists. I will say this: It has been a terrible mistake to see productivity increases as creating jobs. Distinctions are required. It is new expectations of future productivity increases that are a strong job creator. The left was dead wrong that dynamism spells wide joblessness. Actual increases in productivity are different. An unanticipated economy-wide increase could be approximately neutral for employment, raising wages and nonwage incomes about equally.
The lesson of rational booms is that realization of a long-anticipated productivity gain -- hence long reflected in the value CEOs put on a customer, an employee, an office, so it has already largely generated the extra investments that the gain justified -- brings a cut in investment activity, causing employment to decline. (The productivity gain raises the cost of acquiring more assets.) The gain represents an increase in the cost of keeping employees off production to train new employees; so that is the time to stop hiring, so jobs decrease and output increases. Such anticipated gains are why booms end, not how they are created.
But this gloomy outlook is not the only tenable one. Recall that current expectations of future productivity gains not already anticipated in past investing serve to boost current investment activity, which generally lowers the natural rate. The decade could yet see a revival in expectations for future productivity growth.
A real fear is that CEOs have turned sour on the prospective returns on investment or have hiked the risk premium on what they see to be their cost of capital. If they have, the causes of this loss of confidence must be identified and addressed. Another real fear is that maybe our economic institutions and government policies are subtly tilting toward old capital, entrenched management and CEO pay ungeared to results. We need to guard against European corporatism and old-fashioned cronyism. The real hope is that the enterprising spirit is so strong here that, even if the system is not tuned up for best results, there will continue to be enough upstart entrepreneurs and established ones that will hit upon ideas for new products and methods worth developing and trying to market. With just our accustomed dynamism we can look forward to normal times, with their ups and downs.
Mr. Phelps is McVickar Professor of Political Economy at Columbia University.
Columbia University should be embarrassed as well.
What does the "Austrian" theory really say? Go here to find out, and check out this for a bit of irony.
UPDATE: Hayek-Ler John Cochran has a response to Phelps. And Casey Khan response as well.
Posted by Greg Ransom
