TCS Daily

The Case of the Employment Forecast

By Arnold Kling - March 11, 2004 12:00 AM

"Is there any other point to which you would wish to draw my attention?"

"To the curious incident of the dog in the night-time."

"The dog did nothing in the night-time."

"That was the curious incident," remarked Sherlock Holmes.

-- From "The Adventure of Silver Blaze" by Arthur Conan Doyle

Partisan economists are attacking the Administration's forecast of rapid job growth this year, as Paul Krugman did recently. In fact, however, those attacking the Administration forecasts appear to believe them. If you were a left-leaning economist who thought that employment growth was going to be feeble, you would be making an urgent case for more economic stimulus. You would call for tax cuts and/or increases in government spending. At the very least, you would call for more interest rate cuts from the Federal Reserve.

The Administration's opponents do not have a stimulus package. From that, it is logical to deduce that they judge the economy does not need one. In that case, they must believe that something close to the Administration's employment forecast is probably correct. Elementary!

The Blindfold Test

Maybe I'm too old-fashioned. Or too idealistic. But I believe that the way an economist looks at the data should not depend on the political party that is in power. I expect to be held to that standard. And I think that it is reasonable for other economists, even Krugman, to be held to that standard.

Call it the blindfold test. Put a blindfold on an economist, show him some data without telling him which year it comes from, and ask him to do his analysis not knowing which political party is in power.

Below are some data on output and employment in the nonfarm business sector of the U.S. economy, available from the Bureau of Labor Statistics.

Percent Growth, fourth quarter 2000 to fourth quarter 2003



Output Per Person




Now, suppose you were to do a blindfold test. Give an economist the actual output growth of 7.8 percent over three years (roughly 2.5 percent per year) and ask the economist to "predict" the breakdown between growth in employment and growth in output per person. Almost any economist would have said something like "Well, the output growth sounds pretty close to trend. So, assuming that trend productivity growth is about 1.5 to 2 percent per year, I figure 5 or 6 percent total productivity growth and 2 or 3 percent total employment growth."

No economist in his or her right mind would have predicted that we would have gotten 7.8 percent growth in output with 11.3 percent growth in productivity and a loss of 3.1 percent in employment. That is completely outside the experience of history.

By attacking the Administration's credibility in forecasting employment, Krugman and others are implicitly criticizing its failure to forecast the unprecedented burst in productivity. In fact, Krugman's tone all but suggests that the Administration knew about the coming productivity miracle and was engaging in a cover-up!

Here, we have another curious incident of a dog that failed to bark. In 2001, were the Democratic economists warning that productivity was going to take off, which would mean that the economy would require more stimulus to reach full employment? I have not found any evidence of such warnings.

Alternative Policies

The Administration's opponents argue that the President's fiscal policies should have been changed in three ways.

  1. more revenue sharing with state and local governments
  2. use temporary tax cuts in the near term to cope with the recession, rather than enacting permanent tax cuts
  3. skew the tax cuts away from the rich and toward people with lower incomes

None of these alternatives would have addressed the problem that we were just discussing, which is that the relationship between output growth and employment growth was distorted by unusually high productivity growth. Even the proponents of these alternatives do not argue that they would have increased employment by lowering productivity, or even that it would be desirable to do so.

From the standpoint of dealing with the recession, the argument for these fiscal policy alternatives would have to be that they deliver "more bang for the buck," meaning more stimulus to spending. However, even if one had known that more would be needed, there is no evidence to suggest that these alternatives would have provided additional stimulus. State legislatures are not quick to act. If the tax cuts went to "the wrong" people, who did not spend them, then this would show up as an increase in the personal savings rate. However, the personal savings rate has remained at low levels over the past three years. By the same token, there is a theoretical case to be made that temporary tax cuts would have been less stimulative than permanent tax cuts, but, once again, the empirical difference does not seem to be as large as theory might suggest.

I have a different argument for using revenue sharing and short-term tax cuts, which has nothing to do with their effect on employment. My argument is that revenue sharing and short-term tax cuts are faucets that easily can be turned off when the recession is over, so that there is less damage to the long-term fiscal outlook. The Administration's opponents make this argument, at least on Tuesdays and Thursdays when they focus on deficits. But on Mondays, Wednesdays, and Fridays, when they focus on jobs, they have no case.

The Spoiled Kid

If you gave me a blindfold and asked for my analysis of the Bush tax cuts, I would say that they trouble me in the context of the Administration's failure to address entitlement spending. Because the President has not touched the long-term Medicare and Social Security deficits -- other than to add to them with the prescription drug benefit -- I find it quite offensive that he wishes to claim credit for cutting taxes. Long term, he has done no such thing.

A President who has only added to future entitlement obligations ought to be judged as having acted to increase taxes. To call this Administration a tax cutter is like taking a spoiled kid who does not touch dinner but takes a double portion of chocolate cake for dessert a "good eater."

Unfortunately, this is not the criticism that is being made by the economists who are partisans for the Democrats. Once again, they are dogs that fail to bark.

Arnold Kling recently wrote for TCS about "Red Sox technologies."


TCS Daily Archives