Pulling apart job data

Tuesday’s news on Minnesota’s unemployment rate and job gains was welcome. Our state’s economy clearly continues to strengthen.

The dramatic drop in the state unemployment rate and record increases in some job numbers need to be taken in context, however. So must the national news that President Bush took the formal step of reappointing Alan Greenspan to another term as chairman of the Board of Governors of the Federal Reserve. In both announcements, there is both more and less than initially meets the eye.

Take Minnesota first. The news is dramatic. The unemployment rate fell from 4.8 percent in March to 4.1 percent in April. This drop is the largest on record. The number of jobs increased by 12,100, the largest month-to-month increase in nearly five years. Employment in manufacturing, a sector widely described as being on the ropes only last year, grew by 4,500, another record and one that nearly doubled the record set in 1995.

All of this is great news. Still, it may not be quite as dramatic as headlines make it appear. Minnesota employment statistics, like those of all other states and of the federal government, are estimates. They are not a precise verification of the employment status of every U.S. resident.

Carrying out an exhaustive verification would be like performing the U.S. Census every month. The effort would be enormously costly. Instead, employment numbers and unemployment rates are based on two surveys conducted by the federal government.

One, a survey of households, hits about 60,000 out of 120 million in the country. This is a very small sample. The other is a survey of 300,000 “establishments” that include virtually every large employer. Both surveys are meticulously designed and carried out. Together, they give as good an estimate as one could expect of labor market conditions without spending much more money. Still, they remain estimates — very good estimates, but not a perfect picture of reality.

Data from these surveys go into tabulations of unemployment rates and employment numbers for the nation and some large metropolitan areas. Each state then uses the same information with data gathered locally to make a statistical estimate of the same indicators at the state, county and major city levels.

The fact that the most commonly cited numbers are statistically adjusted to iron out seasonal variations is an additional complication. When people hear such news as Tuesday’s, they need to remember that the numbers are estimates — ones that are very carefully and consistently made from month to month — but still estimates.

Even the best estimation techniques are subject to minor blips that give a distorted idea of what is really occurring. Seasoned economists know not to pay too much attention to month-to-month changes, even when such changes appear to set records.

Trends visible over a series of months are much less subject to quirks induced by the tabulation process. The April numbers are a continuation of an improvement trend that has been going on for several months, so the record numbers cannot be far off the mark even if they are not perfect.

Minnesota’s economy clearly is improving.

People should realize, however, that while the economy is likely to continue improving, we already are at a level of unemployment that is very good by long-term standards. If you expect unemployment rates to drop another two full percentage points, to the levels experienced in the late 1990s, you will be disappointed.

The Greenspan reappointment is fine, but not important, news. If a beleaguered President Bush, in the middle of a tough re-election campaign, had not named Greenspan again, the uproar would have been enormous. No one in Bush’s position would rock the boat unnecessarily in such a way.

But, reappointed or not, the chairman is increasingly becoming a lame duck. No member of the Board of Governors can fill more than one full 14-year term. Being appointed to fill out the partial term of someone who resigns, as Greenspan was in 1987, is common. Once such a partial term is up, the individual can be given a full 14-year term.

In what turned out to be the famous “I appointed him and he disappointed me,” episode, former President George H.W. Bush so appointed Greenspan in 1992. But that term expires in January 2006, less than 20 months from now.

Thus, whoever wins the November presidential election will immediately face the question of whom to select as Greenspan’s successor. Greenspan already has served four years longer than Franklin D. Roosevelt served as president of the United States.

Whenever some individual serves a long term, such as J. Edgar Hoover at the FBI or Hyman Rickover with U.S. nuclear subs, the transition to a new appointee is never easy.

Tuesday’s announcement puts things on the back burner until 2005. The next president will have to take the lid off that particular pot soon enough. The decision of whom to appoint may well be the most important one he will make in his term.

© 2004 Edward Lotterman
Chanarambie Consulting, Inc.