Economists assess labor market conditions using two reports in particular: the Current Employment Statistics survey (also known as the payroll survey), compiled by the U.S. Bureau of Labor Statistics, and the Current Population Survey or household survey, produced by the U.S. Department of Labor. Conventional wisdom holds the former to have the most reliable numbers, and while the two sets of numbers usually differ, they nevertheless follow the same pattern.

At an AEI seminar on May 10, Tim Kane of the Heritage Foundation compared the methodology of the two surveys and challenged the accuracy of payroll survey numbers. This survey provides the larger sample and asks businesses how many people they employ, while the household survey uses a smaller sample but asks workers whether they are currently employed. Kane contended that the payroll survey double-counts those who change jobs during a single pay period and therefore currently shows job growth as artificially deflated during this period of low job turnover. He noted that the number of the self-employed has risen but is still not counted in the payroll survey, thereby making the household survey a better gauge of the labor market.

Tom Nardone of the Bureau of Labor Statistics disputed charges of inaccuracy on the payroll survey and argued that both surveys provide important information on labor markets. Factors such as the accuracy of population controls, misclassification of the self-employed, and changes in turnover cause inevitable variances in payroll data.

David Malpass of Bear Stearns pointed to positive job growth in both surveys and argued that since each survey measures job markets differently, trying to reconcile the numbers may prove futile. The payroll survey counted 288,000 new jobs in April and 958,000 new jobs over the last six months. Unemployment dropped to an historical low of 3.6 percent in 2000, which Malpass considered to be an unsustainably tight labor market. He put today's figures in historical terms with recessions generally peaking at 7 to 10 percent unemployment, whereas the last recession peaked at 6.2 percent and has steadily declined to its current level of 5.6 percent. There has also been a decline of 226,000 part-time jobs even as total overall employment grew by 958,000 jobs. He concluded that the economy is stronger than most people believe and that the current economic expansion will continue into 2005.