Five Reasons Why the Unemployment Rate Is a Bad Statistic

Politics Features Unemployment Rate
Five Reasons Why the Unemployment Rate Is a Bad Statistic

The jobs report for April came out Friday morning, and one of the highlights was the 3.6 percent unemployment rate, which, as outlets like the Times pointed out, is the lowest since 1969. It didn’t take long for Trump to crow about the numbers on Twitter:

What follows is not meant to diminish the April report, which showed the U.S. economy beating expectations with 263,000 new jobs, though wage growth was disappointing. Instead, it’s a critique of “unemployment rate” in general as a measure of the economy’s health.

Before I go into why it’s a flawed statistic, ask yourself if you really know how the unemployment rate is calculated. I say that without condescension, by the way—to me, it’s a deceptive term, and any reasonable person would answer the question with “the percentage of working-age people who are out of a job.”

In fact, that’s not the case, which leads to the first reason why this is a bad statistic:

1. Very few people know what it means

The truth is, unemployment rate is limited to considering those Americans who are in what’ called the “labor force.” That means, in order to count, a person has to either have a job or be actively looking for one. What percentage of potential workers do you think that would entail? The latest report paints a dire picture:

There was a big drop in the number of people who said they were looking for work. The labor-force participation rate, which measures the share of people 16 and older who are employed or seeking a job, fell to 62.8 percent, from 63 percent in March.

As you see from this graph, that number has been consistently dropping since it peaked roughly at the turn of the Millennium:


source: tradingeconomics.com

Which leads to our second reason…

2. Unemployment rate doesn’t count some critical demographics

When your statistic only counts 62.8 percent of the working-age population, it’s reasonable to believe you’re probably omitting some people you shouldn’t. Which turns out to be true: While certain groups like stay-at-home parents aren’t rightfully left out, so too are unemployed workers who have given up looking for a job, even if they want one. That can leave out millions of people and paint a falsely optimistic picture.

Or take all the U.S. citizens who are incarcerated. They’re not counted at all, but how many would be employed if they weren’t in prison or jail? How many were driven to crime in part because of unemployment? To simply write them off as though they’re independent from the numbers—a totally separate entity—is short-sighted and deceptive.

3. A potential worker disappears from the “labor force” after just one month

In order to be considered someone who has given up on a job—a “discouraged worker”— you only have to be dormant for one month. That means if you’re sick and can’t work for longer than a month, or have to take care of a loved one, or get pregnant, you don’t get counted in the unemployment rate. It’s very hard to read that one-month period as anything but an arbitrarily short term designed to juke the stats.

But it’s not just the people who aren’t counted…

4. Unemployment rate counts some people it probably shouldn’t

Seasonal workers and part-time workers count as “employed,” even if they would rather be full-time workers, or their jobs will disappear after a certain date. Seasonal hiring may have had a significant on this month’s numbers, the Times reports:

One-time factors may have stimulated hiring in some parts of the economy. Gregory Daco, chief United States economist at Oxford Economics, noted that the 34,000 increase in leisure and hospitality jobs might suggest that restaurants and hotels were staffing up earlier in the month for Easter, which fell late in the month.

And while, sure, both of these categories of people are technically “employed” at the time of the report, many are in fact under-employed, and counting them in the same way that full-time employees are counted is not useful or accurate.

5. There’s a better stat that we should all be using

“Real unemployment” accounts for discouraged workers (although after 12 months, they too disappear from this metric) and those with part-time work who would prefer full-time employment. This is still not a perfect measure, as it doesn’t account for wage growth, and it tends to follow a similar path as the unemployment rate, but the higher number—7.3 percent in March—gives a much more accurate picture of the national state of employment.

In every case mentioned above, the flaw tends to present a lower number—probably not an accident, since it comes from a Labor Department that doesn’t want to spread bad news if they can help it—and while the statistic has the appearance of being vaguely useful in historical comparison to itself, it says very little about the state of our economy.

Nevertheless, it’s the leading story in every jobs report, and in the Times piece today, for instance, wage growth takes a back seat and there’s not a single mention of the real unemployment rate—just as there isn’t in the Bureau of Labor Statistics summary. There’s an appropriate quote here about lies, damned lies, and statistics, but the better advice is just to be careful of what you trust.

Share Tweet Submit Pin