Readers Sound Off on Payroll Survey

Last week's column on the two different employment surveys generated a ton of reader mail. Many of you asked good questions that took a little digging to get answers for, and there were some worthwhile comments, too. So let's address some of those questions now, since part of our charge here is to answer questions, rather than raise new ones.

What size businesses does the payroll survey cover? Doesn't the Household survey pick up more start-ups because the payroll survey only questions firms of 50 people or more?

This seems to be a very common misunderstanding about the payroll survey. Quite a few people suggested the 50 cutoff number. That turns out to be incorrect.

According to the Bureau of Labor Statistics, or BLS, the universe of firms covered by the establishment (payroll) survey is representative of 99%+ of all businesses that pay unemployment insurance -- regardless of size. So if a company has two employees or 2,000, they are eligible to be included in the survey, just as long as they pay unemployment insurance.

It makes sense. This survey seeks to measure the compensated, nonfarm payrolled employees, so a mandatory measure like unemployment insurance is a good way to ensure an accurate count.

What actual numbers are significant in jobs creation? Why are some numbers better than others, i.e., 200,000 was bad one month, but 150,000 was good another month?

It is a matter of expectations: Dow Jones, Bloomberg and CNBC gather the predictions from 27 or so major economists. The consensus is the average of those estimated numbers. (Some consensus estimates throw out the outliers.) When the consensus is higher than the actual number, economists are disappointed. When it is stronger, they are happy.

What payroll creation numbers indicate solid growth? Why is 32,000 new jobs a bad number?

Job creation turns on two concepts: absolute and relative numbers.

Looked at in terms of "absolute numbers," the economy has been creating a good number of jobs -- several million over the past few years. This is a good thing.

It's when we get to the "relative numbers" that problems arise: We need about 140,000 new jobs per month just to keep up with expanding population, due to immigration and new employees in the labor force.

Indeed, as the population has expanded faster than job growth, the percentage of the labor pool that is working went down. So while there are more jobs today than there were a few years ago, as a percentage of the labor pool fewer people are working. This, combined with the increasing number of part-time jobs, explains why payroll-service providers like Automatic Data Processing (ADP - commentary - Cramer's Take) and Paychex (PAYX - commentary - Cramer's Take) say that the number of employees per client is up considerably year over year.

And this is before we even begin to consider the impact of high productivity on job creation. If this sort of wonky stuff interests you, you can see the details in this column I wrote in June 2003.

---

Barry Ritholtz is chief market strategist for Maxim Group, where his research and market analysis are used by the firm's portfolio managers and clients in the U.S., Europe and Japan. He also publishes The Big Picture, his macro perspectives on the economy and geopolitics, entertainment and technology industries. At the time of publication, Ritholtz had no position in any securities mentioned in this column, although holdings can change at any time. Under no circumstances does the information in this column represent a recommendation to buy or sell stocks. Ritholtz appreciates your feedback and invites you to send it to [e-mail].