TCS Daily

Would Keynes Change His Mind?

By Arnold Kling - May 6, 2003 12:00 AM

"Practical men, who believe themselves to be quite exempt from any intellectual influence, are usually the slaves of some defunct economist. Madmen in authority, who hear voices in the air, are distilling their frenzy from some academic scribbler of a few years back."
- John Maynard Keynes

I recently characterized the United States as an elastic economy, meaning that there are more opportunities for substitution and adaptation in response to economic shocks. As I will explain below, John Maynard Keynes described an economy that is inelastic in several key respects. Does this mean that government attempts at macroeconomic policy reflect an outdated distillation of Keynes' own academic scribbling?

Keynes vs. Elasticity

In the 1930's, the United States experienced persistent rates of unemployment of 15 percent or more. This would be impossible in an economy in which the assumptions of classical economics hold. So Keynes broke with some of those assumptions. One way to describe Keynesian economics is that it includes several important inelasticities.

One feature of Keynesian economics is that the labor market behaves as if the demand for labor is highly inelastic in comparison to the supply of labor. Inelastic labor demand means that when a firm suffers a drop in demand, only a dramatic reduction in wages would allow it to maintain the same level of labor input. Workers will not accept such wage reductions, and the result is a drop in employment. Stories like this are the result. During the Depression, this phenomenon occurred in many industries at once.

Keynes argued that both saving and investment are inelastic with respect to the interest rate. He viewed saving as being determined by a basic hoarding propensity, which does not depend on interest rates. He viewed investment as being governed by "the state of long-term expectations." He would have described the dotcom boom as a case of "animal spirits" on the part of venture capitalists and Wall Street speculators, with the current situation as being one of pessimistic long-term expectations. Changes in interest rates are not sufficient to offset these mood swings.

The Elastic Economy Meets Keynes

Today, the economy is more elastic than it was in the 1930's. Today's recession is a far cry from the Great Depression of the 1930's. Of course, some of this may be due to a difference in the severity of the shocks in the two periods (making that comparison would be a difficult task). And much of it is due to a better policy regime, particularly relative to money and banking. But I believe that some of the credit belongs to the more elastic economy.

Labor demand seems to be more fluid than was the case in the 1930's. The economy is less concentrated in the manufacturing and agriculture sectors. Firms are more adaptable, and the employment base is more diverse in terms of types of work and variety of industries.

The Internet and increased foreign trade contribute to the elasticity of labor demand. The ability to shift production overseas enables firms to reduce wages for some jobs. With the Internet, the work that can be relocated now includes computer programming, customer support, and clerical functions.

In fact, these days labor supply may be less elastic than labor demand. A reader of EconLog wrote this comment:

"I was talking with the Executive Director of my alumni association this afternoon. He notes that none of the current class of computer science graduates has a job offer as yet. In contrast, each of the Petroleum Engineers has an average of seven. The school has a 4-6 year pipeline for graduates in each department. Students tend toward those departments that have good offers for their graduates when the students enter the program.
This would indicate at least several years worth of lag in responding to a changing labor market."

When it is labor supply that is inelastic, and there is a mismatch of skills with needs in the labor market, economists refer to this as structural unemployment. (Note, however, that there is still a question of why these mismatches cause unemployment as opposed to adjustments in relative wages.) Structural unemployment is a supply-side issue, not necessarily amenable to Keynesian demand-side policy intervention.

Even though the stock market continues to be at the mercy of the "state of long term expectations," overall saving and investment do respond to interest rates. Lower long-term rates stimulate the demand for housing construction. Moreover, homeowners who do not buy new homes can refinance their mortgages, giving them the ability to increase consumer purchases.

One of the mechanisms for balancing the supply and demand for capital is international trade in goods and assets. Even if domestic saving and investment were not responsive to a drop in interest rates, a decline in rates here relative to overseas would cause a decline in foreign investment in the U.S., which in turn would lead to a weaker dollar and a lower trade deficit. Lower interest rates can increase the demand for domestic output through this channel.

The Policy Challenge

The good news is that the elastic economy is more resistant to economic downturns, because labor markets, asset markets, and capital markets have become more complex and more diverse. The bad news is that by the same token the elastic economy is more resistant to economic policy. Any given amount of fiscal or monetary stimulus will have less "bang for the buck" than in the past.

For monetary policy, the problem is the high elasticity of supply and demand for long-term financial assets. As a result, the Fed can manipulate short-term interest rates without necessarily having any effect on mortgage rates, corporate bond rates, or other interest rates that affect saving and investment. The ability of conventional monetary policy to affect real economic decisions is weakened. (See Can Greenspan Steer?)

I suspect that fiscal policy, too, is less effective. My guess is that as labor markets become more complex, they tend to follow their own dynamic. An increase in spending or a cut in taxes provides less of a "quick fix" when unemployment is a varied mix that includes hotel workers, telecommunications managers, and web programmers, with some industries in cyclical slumps and others in secular decline. It was probably easier for stimulus to work when unemployment consisted of laid-off auto workers ready to return to the assembly line.

For the elastic economy, the options for dealing with our current unemployment problem appear to be these:

  1. Increase the dosage of fiscal stimulus. The disadvantage of this is that it would put taxpayers deeper in the hole in the long run.

  2. Increase the dosage of monetary stimulus. Even though the Federal Funds rate is already less than 2 percent, we are not in a liquidity trap where there is no room for rates to fall farther. Instead, as Federal Reserve Governor Ben Bernanke has pointed out, the Fed has available to it the option of buying long-term Treasury bonds or even mortgage-backed securities. The disadvantage of this is that it might prove to be destabilizing, because we do not have any experience with which to calibrate the degree of novel forms of monetary intervention.

  3. Wait for the private sector to right itself. The disadvantage of this is that it has become politically unacceptable to "do nothing" about recessions.

No doubt there are demagogues out there who will claim to offer a risk-free, pain-free way to restore full employment. But to me, all of the options appear to have down sides.

Perhaps even Keynes would tell the public today that the Federal government is less well positioned to solve the problem of unemployment than are the individual decision-makers operating in a decentralized, elastic economy. After all, it is Keynes who reportedly once remarked, "When the facts change, I change my mind - what do you do, sir?"

1 Comment

Has anyone ever proved Keynesian theory (Keynes effect, liquidity trap, investment trap, sticky wages) correct? Has anyone ever verified the model?

TCS Daily Archives