How many people in the US are unemployed?
Sean Spicer, the press secretary for President Donald Trump, refused to answer that question during a press conference Monday. It’s unclear why, since there are so many numbers to choose from.
The official answer, according to the Bureau of Labor Statistics, is 4.7 percent, or 7.5 million people. That’s the number of unemployed as a percentage of the labor force. The BLS defines the unemployed as “people who are jobless, actively seeking work, and available to take a job.”
Even though there is only one official definition and only one official rate, BLS also publishes five alternative unemployment rates.
These rates account for:
- people who have been unemployed for 15 weeks or more
- people who just lost their jobs or completed temporary work
- people who are unemployed plus the people who are too discouraged to look for work
- people who are unemployed plus discouraged workers plus people who are not actively looking for work but would take a job if it came along
- people who are unemployed plus discouraged workers plus people who are not actively looking for work but would take a job if it came along plus people who are only employed part time but would like to be employed full time
That last metric, which BLS calls U-6, is the most drastic definition of unemployment. It is sometimes referred to as the “real” unemployment rate. It was 9.1 percent, or 14.5 million people, in December.
The reason there is such a huge spread between this rate and the official rate is largely because of the underemployed. In December, there were 5.6 million people who wanted to be working more but couldn’t. While not technically “unemployed,” these people probably feel like the economy sucks right now.
Almost everyone uses the BLS rates. It takes a lot of work to crunch these numbers monthly. There are some lone wolf economists like John Williams, the proprietor of the subscription service ShadowStats, who calculate their own stats — but those seem to mostly be twiddling with different parts of the BLS numbers, adding different subsets together, with some other stuff tossed in. (Williams, who believes unemployment is at an implausible 22 percent, does not make his complete methodology public.)
The Hornstein-Kudlyak-Lange Non-Employment Index (NEI) was devised by a group of economists in 2014 who thought they could improve on the official federal calculation. It’s published monthly by the Richmond Federal Reserve, which covers the region from Maryland to South Carolina and helps set national monetary policy. The Richmond Fed still uses conventional unemployment statistics in its research, but it calculates the NEI on the side for contrarian economists to cite in their policy recommendations. The NEI was 8.4 percent in December 2016. Adding in people who are underemployed bumps the rate up to 9.5 percent.
This metric tends to be higher than the official unemployment rate because it includes the retired, disabled, students, and other people who are out of the labor force but could be working (but notably, still not the 2.3 million people who are in prison). However, it weights these people according to their likelihood of transitioning back into the labor force. In a way, this is a less humane metric since it tries to measure labor market “slack” — whether the US economy is pumping as hard as it can rather than whether people who want to work are able to find it.
Another interesting metric is the Gallup Good Jobs rate. “What the whole world wants is a good job,” according to the research firm. “This is one of the most important discoveries Gallup has ever made.” Gallup defines a “good job” as working 30 or more hours per week for a regular paycheck, calculated via 30,000 telephone surveys that update on a rolling basis. That rate in January was 45 percent. The Gallup Good Jobs rate is primarily cited by news media and ignored by the economics community.
There are a few other ways to break down unemployment. You could look at just the “adult unemployment rate,” which focuses on the 25-and-up category, or splice out the full-time unemployment rate, which is just people actively seeking full-time employment — both metrics that have been used at various points by developed countries. You could look at the opposite of unemployment, which is the employment-population rate. That’s the percent of people 16 years or older who are employed, which the BLS calculates at 59.7 percent. By that metric, 40 percent of Americans — in this strictest, most technical sense — are unemployed.
Spicer dismissed the question by saying that the president doesn’t care about statistics, he cares that Americans are hurting. That’s reasonable, although statistics might be helpful in determining which Americans are hurting, in which ways they’re hurting, who is hurting the most, and therefore how to prioritize them. However, if the Trump administration wants to claim the unemployment rate is higher than 9.1 percent in order to score political points, it will be hard pressed to find a reputable source.