The Payroll Survey vs. The Household Survey

The payroll data is accurate, and the household survey isn't. Which is what most economists have been saying for months. But that doesn't stop the Conservative Lysenkos from saying that the household survey is the right one because it matches better with their idea that the economy is just fine.

They'll continue to say it, but Alan Greenspan and the Federal Reserve just pulled the rug out from under them by saying that "...the gloomy payroll data is essentially accurate and that the household survey is probably off base."

Economic View: Two Tales of American Jobs
February 22, 2004
ECONOMIC VIEW
Two Tales of American Jobs
By EDMUND L. ANDREWS

Washington

FOR more than a year, Bush administration officials and Republicans in Congress have seized on an intriguing statistical puzzle to suggest that job creation in the United States may be much stronger than it appears at first glance.

The puzzle is the enormous divergence between the two surveys that are used by the Bureau of Labor Statistics to measure job creation and unemployment. The payroll survey, which is based on a monthly poll of 400,000 employers, shows a loss of more than two million jobs since 2001. The household survey, based on questions posed to people in 50,000 households, shows an increase of more than 500,000 jobs over the same period.

If the payroll survey is correct, Mr. Bush is on track to be the first president since Herbert Hoover to complete a term in office with fewer jobs than when he started. If the household survey is correct, Mr. Bush can claim credit for creating jobs despite the blows of a recession, terrorist attacks and two wars.

The household survey also seems to support a political theory: that many people dropped from the company payrolls are not unemployed but rather self-employed. While the payroll survey suggests economic malaise, the household survey implies entrepreneurial energy.

"The household survey shows that we're at an all-time high in employment,'' said Senator Don Nickles, Republican of Oklahoma and chairman of the Senate Budget Committee, at a hearing this month. "It shows that, at least if you look at this trend, the employment situation has improved rather substantially.''

Administration officials are more cautious.

"At this point, the gap between the payroll and the household data continues to be a puzzle,'' said N. Gregory Mankiw, chairman of the White House Council of Economic Advisers, in a speech this month. But, he added, the number of self-employed workers has risen by 326,000 in the last three years and the "extent of self-employment has changed as the economy has changed.''

Unfortunately for the optimists, the Federal Reserve has just thrown cold water on the household data. It concludes that the gloomy payroll data is essentially accurate and that the household survey is probably off base.

"I wish I could say the household survey were the more accurate,'' Alan Greenspan, the Fed chairman, said in his testimony at a House hearing on Feb. 11. "Everything we've looked at suggests that it's the payroll data which are the series which you have to follow.''

To test the self-employment theory, the Fed adjusted the household survey by taking out all the kinds of workers who do not show up on the payroll survey - self-employed people, but also farm workers and family workers in family-run companies. Even then, Mr. Greenspan said, the discrepancy remains large.

The Fed's conclusion was that the household survey's results have been inflated by overestimates of population growth.

Because the household survey is a sample, the Bureau of Labor Statistics infers the total change in jobs by multiplying the ratio of employed to unemployed workers in the household survey by its estimate of the total population. If the population estimate is too high, the estimated number of jobs will also be too high.

THE Bureau of Labor Statistics bases its population estimate on the 2000 census, but it updates that estimate yearly with data on births, deaths and immigration. Immigration numbers are largely guesswork, however, because so much immigration is illegal. Fed officials suspect that the immigration estimate is inflated, because it fails to reflect tighter immigration controls after Sept. 11, 2001, as well as declines caused by the economic slowdown.

Indeed, the Bureau of Labor Statistics lowered its population estimate in January. Plugging the new estimate into the previous household surveys, the bureau found nearly half the apparent increase in jobs during the last three years vanished.

Not content, Mr. Greenspan also devised a "synthetic'' population estimate by crossing the household survey's ratio of employed workers to work force data in the unemployment insurance system. The results? "A significantly slower pace" of population growth, according to the Fed chairman.

The good news for the job market is that both surveys are now pointing to increases in employment. The bad news is that, compared with previous economic recoveries, both measures suggest that job growth remains well below par.

Background and notes: The payroll survey, also known as the establishment survey, also known as the Current Employment Statistics (CES) program, has shown a deep drop in employment, and only recently started showing new jobs in the economy. This program surveys nearly 400,000 establishments in the country asking how many people are employed at each one.

The household survey, also known as the Current Population Survey (CPS), is a monthly survey of 60,000 households. They telephone each household, and ask them how many people live there, how many of the residents are of working age, and how many of them worked that month. This survey showed a shallower drop in jobs, and a more robust recovery.