With the January jobs report released Friday by the Labor Department, we know that President Trump took office amid a relatively low unemployment rate (it ticked up to 4.8 percent in January), strong job growth (227,000 positions added that month) and weak wage growth (average hourly earnings up 0.1 percent in January and 2.5 percent over the past year).
There was even some welcome progress in the number of people who count themselves as part of the labor force.
But Mr. Trump made amply clear in his campaign that he doesn’t care for the way that government agencies and mainstream economists summarize the state of the job market. The unemployment rate, he said in December, is “totally fiction.” He claimed at one point during the campaign that the real jobless rate was not the number below 5 percent widely cited by economists, but something like 42 percent.
There is certainly a filament of truth in that. The unemployment rate counts only people who say they want a job and have looked for one in the last month — meaning that millions of Americans who dropped out of the labor force during the aftermath of the 2008 recession do not count as unemployed by the standard definition.
But we have some exciting news for the president. The Bureau of Labor Statistics doesn’t just issue the widely cited headline unemployment rate (again, 4.8 percent in January). This month’s jobs report, for example, offers 42 pages’ worth of figures that allow anyone who knows how to read them an infinite variety of ways to examine the labor market. No one forces you to treat the standard unemployment rate as gospel; it’s just a convention, and any economic analysts worthy of their spreadsheets use a wide range of data to assess how things are going.
So if the White House wants to choose a different measure of job market health to emphasize on the first Friday of every month, we’re all ears. The nice thing about a convention is you can change it. Here, in fact, is a menu of labor market measures that might be plausible alternatives to the main number — and why the Trump administration may, or may not, want the world to watch it.
Underemployment. The Bureau of Labor Statistics publishes six different unemployment rates, from a relatively narrow definition to relatively broad. The jobless rate most often cited is the third of the six, known in bureaucratese as U-3, and includes only people who say they want a job and have actively looked for one in the last month. The broadest of these unemployment rates, U-6, also includes people who want a job but haven’t looked in the last month and people who have a part-time job but want full-time work. That rate: 9.4 percent.
Share of population not employed. If Mr. Trump really believes the 42 percent unemployment rate he floated last year is a useful measure of joblessness, here’s the math that almost gets him there. The proportion of the adult population that is working is 59.9 percent, so the remainder of that — the share of American adults not working — was 40.1 percent in January. Think of it as a measure of non-employment, not the more precisely defined unemployment.
There are some problems with that measure. It includes people who are not working and who have no desire whatsoever to do so. It includes 17 year-old high school students, 80-year-old retirees, disabled adults and parents who voluntarily stay home with their children. Moreover, if Mr. Trump wants to focus on this measure, he may not be pleased with the results. Because the baby boom generation is hitting retirement age and voluntarily stepping out of the labor force, this measure of non-employment is likely to rise in the years ahead.
Share of prime-age population not employed. One way to avoid the most obvious problems with that measure is simply to limit the age bracket you look at. Jordan Weissmann at Slate argues that a better all-encompassing measure of the health of the job market is the share of the population between ages 25 to 54 who are working. That number was 78.2 percent in January, so its remainder implies a 21.8 percent jobless rate. This still isn’t perfect — it includes people who fall in that age bracket who have no desire to work, and also fails to include people under 25 or over 54 who really do want a job but can’t find one. But as a quick-and-dirty measure of the health of the labor market, Mr. Weissmann is right that you could do a lot worse.
Average hourly earnings. A common complaint about media focus on how many jobs are being created is that it doesn’t account for whether they are good jobs. But this is a tricky concept to measure; everyone can agree that a minimum-wage, burger-flipping position is not as ideal as a highly paid computer programming gig, but it’s hard to get at more subtle differences. Is a job as a truck driver better or worse than one as a hotel assistant manager? It’s pretty subjective.
What isn’t subjective is compensation. And data on average hourly earnings helps convey not just whether people are getting raises for existing jobs, but also whether the composition of jobs in the economy is shifting toward higher pay. Average hourly earnings for all private sector employees was $26 in January, and for nonmanagerial private sector employees it was $21.84.
An economy in which those numbers rise significantly faster than inflation is an economy in which working-class people are feeling better about the world and are seeing greater rewards for their labor.
There are plenty more options for the Trump administration to consider as job market measures to emphasize. Unemployment in the construction sector (currently 9.4 percent)? The jobless rate among people who didn’t go to college (currently 5.8 percent)? The options are nearly endless.
But — and this is important — if the president wants to set a different measure of the job market for his administration to focus on and improve, it would be best if he could let us know what it is now, so that we can really assess whether things are getting better or worse during his presidency.