Why is there confusion about the job market?

As a professional economist for almost five decades,  I am the first to admit that many economic numbers and measures are confusing.   How many people – besides economists – know what GDP, the federal funds rate, or the personal consumption expenditures price index (I have to pause midway to speak the last one) mean, or what the difference is between the budget deficit and the national debt? \

But for jobs, most people would think that’s the easiest statistic to understand.  Yet if you think that, you are wrong!  First, there are two different and independent measures of jobs.  There are also periodic revisions of the numbers.  Furthermore, there are multiple measures of the unemployment rate. With all these different statistics floating around, how’s any person – other than an economist – expected to make sense of the job market and decide whether it’s good or bad?

That’s what I’m here for in today’s column.  I’ll walk you through the major job and related measures and give common-sense explanations for their meanings.  Then, you’ll be prepared to understand the numbers as they are released so you can decide on the condition of the job market.

But before I begin, exactly who creates the job numbers?  While there are some private estimates, the official numbers are generated by the US Bureau of Labor Statistics (BLS).  The BLS has a reputation of high credibility

I’ll begin with the basics of simply counting jobs. The BLS conducts two monthly surveys.  One called the establishment survey counts jobs at businesses.  The second, the household survey, directly asks individuals about their work status.  The establishment survey also tells us where people are working, while the household survey provides the information to calculate the various unemployment rates.

As noted, both are surveys, meaning the numbers of businesses and households providing information are far less than their actual total numbers.   This is not unusual.  Polls about voting intentions, product preferences, and a myriad of other decisions are also done by sample polling. 

Which brings me to a recent debate. It was reported the BLS reduced the total national job count for the year March 2023 through March 2024 by over 800,000.  While the BLS regularly makes such revisions each year, this number was exceedingly high and sparked charges of possible manipulation. Yet, there are logical answers for the revision. 

The first is that when conducting the monthly surveys, all businesses in the sample may not be ready with their numbers. The BLS could wait until the tardy businesses had the data ready, but the agency is always under pressure to release the numbers on schedule. BLS also doesn’t want to substitute another business because each is carefully selected to represent a specific part of the job market.  So the BLS estimates each lagging firm’s job number, and in the case of this year, the estimates were too high.

A second logical answer is that, like so many things, the pandemic is having a lingering impact.  How so?  It’s highly likely the BLS has not yet fully adjusted their sampling models to account for the impacts of Covid.

Another confusing part of the jobs report has to do with what is interpreted as a good number or a bad number.   Like so much of business, it’s often a matter of expectations.  Each month before the official jobs report is released, there will be a “guess” for the change in the number of jobs based on estimates from private firms and economists. If the actual change in jobs is less than the expected change, financial markets can react negatively, even if the actual change was positive.  This happened for the recent August jobs report.   Even though the job gain in August was higher than the job gain in July, the August gain was less than expected.

Let me now move on to the important topic of unemployment.  The jobless rate is likely the most widely followed labor market number, mainly because it’s easy to understand.  People intuitively know that if the jobless rate rises, it’s not good for the broad economy and could warn of bad times ahead.

But, of course, economists aren’t content with one measure of the jobless rate.  Instead economists have created several rates, with each saying something slightly different about the job market.

The official jobless rate is based on a simple calculation.  It is the number of individuals who don’t have a job and who are actively looking for work and available for work as a percentage of the labor force.  The labor force is everyone aged 16 and over except for those in active military duty or institutionalized..  The national rate in August was 4.4% up from 3.9% in August 2023.

 But there are alternative jobless rates developed by BLS.  A closely watched rate is called “U-6.”  U-6 includes the above-mentioned unemployed but adds people who want a job but have stopped looking as well as individuals who have a part-time job but want a full-time job.  In August the U-6 rate was 8%, up from 7.2% the year earlier.

 Of course, care must be taken not to mix these different jobless rates, because they use different measures of unemployment.   The official jobless rate has the narrowest measure of unemployment, whereas the U-6 rate has the broadest definition.  I like to follow both.

There are many numbers constantly thrown at us about the economy.  There are so many numbers that non-economists need a guide to what they mean. Hopefully this column has given readers such a guild to make common sense of key labor market numbers.  But, you decide.

Walden is a William Neal Reynolds Distinguished Professor Emeritus at North Carolina State University.