Could The Way Unemployment Is Reported Change Under Trump?

The president doesn’t trust the unemployment numbers, but the stats he’s looking for have been a part of the report for 70 years.

Could The Way Unemployment Is Reported Change Under Trump?
[Photo: Flickr user mathiaswasik]

When it comes to certain issues, we know the truth is complicated.


Take the gender wage gap. It’s commonly quoted that women make, on average, 77¢ for every dollar that a man in the same job earns. That number comes from U.S. Census Bureau data that indicates what the typical American woman working at least 35 hours a week, year-round, is paid. But it’s not accurate to cite that figure as the ultimate indicator, as the actual figure varies between white women versus black and Hispanic women.

Nor is it fair and accurate to cite the unemployment figure as the chief indicator of the health of the job market or the economy.

When the Bureau of Labor Statistics (BLS) released its monthly jobs report on February 3, the unemployment rate ticked up slightly to 4.8% with 227,000 new jobs added for the month of January. Taken alone, that figure looks remarkably healthy for the new administration. By comparison, when President Obama took over in January 2009, he inherited an unemployment rate of 7.6%, representing a continued loss of jobs across all industry sectors, according to the BLS’s report that month.

During his campaign, Trump at times has called the unemployment figure (which decreased during Obama’s eight years in office) “one of the biggest hoaxes in politics,” as well as a “joke,” claiming the rate could be as high as 42%. Others have called the statistic into question, including Jim Clifton, chairman and CEO of Gallup, who titled a blog post he wrote “The Big Lie.”

Who’s right when it comes to the numbers?


“Saying that the BLS unemployment rate is “misleading” could mean two very different things,” Katharine Abraham, PhD, the commissioner of the BLS between 1993 and 2001, tells Fast Company. Abraham, who is currently director of the Maryland Center for Economics and Policy, says it could mean that you think the unemployment rate doesn’t measure what it claims to measure, or that the unemployment rate measures what the BLS says it measures, but that it isn’t the best or only measure of what is going on in the labor market. “I interpret most of the comments that have been made by people associated with the Trump administration as being of the latter sort,” she says.

“I’d take strong issue with anyone who questioned the integrity of the BLS numbers,” Abraham continues, but agrees that it makes sense to be paying attention to some other labor market statistics the BLS produces. She’s referring to the labor force participation rate and the share of the workforce who are working part-time but would prefer full-time work–in addition to the unemployment rate.

How The BLS Gets Its Numbers

What many people don’t know is that since 1940, the BLS has been producing the kind of in-depth report that covers all the areas Abraham points out.

The current report, as well as each previous one, is over 40 pages long and includes a variety of measures both in chart and text form that analyze employment numbers by industry, location, etc. At the end of each report, as well as in a dedicated FAQ page on the BLS website, there is a clear explanation of what the numbers are, how they are gathered, and how each category is defined.

For example, the basic concepts involved in identifying the employed and unemployed:

  • People with jobs are employed.
  • People who are jobless, looking for a job, and available for work are unemployed.
  • The labor force is made up of the employed and the unemployed.
  • People who are neither employed nor unemployed are not in the labor force.

To find these numbers, the BLS takes surveys every month called the Current Population Survey (CPS) to measure the extent of unemployment in the country. It went through a major modification in 1994 to computerize the interview process and get more information.

The BLS doesn’t count every individual every single month to see if they are working or not. They take a sample of around 60,000 eligible households (about 110,000 individuals). This is much greater than the average public opinion poll, which tops out at around 1,000. To get a representative sample, the BLS divides by geographic area that reflects both rural and urban centers as well as different industries. To further mix it up, every month, a quarter of the households rotate off the survey so no one household is interviewed more than four months in a row. As the BLS notes, “This procedure strengthens the reliability of estimates of month-to-month and year-to-year change in the data.”

Related: This Is The Hidden Challenge In The Future Of Work

Bob Murphy, a professor at Boston College who previously worked as a White House economist under President Bill Clinton, calls survey and statistical analysis the “rock-solid gold standard.” Furthermore, he says that the BLS is a world-class independent statistical agency that is inherently nonpartisan, and challenges to their reporting is “complete nonsense.”

“They’re very clear about the margin of error,” he points out, and the BLS has noted in the past if the unemployment rate falls by as little as 1/10th of a percentage point, it is not statistically meaningful. “Politicians and the public don’t appreciate that,” says Murphy.


Why Is The Administration Suspicious Of The Unemployment Numbers?

Despite the proliferation of data, just over one-third of executives (38%) had real confidence in the insights their analysts were providing, according to a survey by KPMG and Forrester Research. Bill Nowacki, the managing director of decision science at KPMG in the U.S., told Fast Company in a previous interview about that survey that the mistrust could be the result of traditional thinking clashing with new technology.

Murphy says that the mistrust of government officials goes back to Watergate. “Also, social media has a huge multiplier effect,” he says. “Something that would have never gotten to see the light of day gets sent around a lot,” Murphy explains. And while he encourages skepticism toward officials, Murphy says the spillover to mistrust the BLS is “not justified.”

Some economists have pointed out certain statistics that need to be taken into account that don’t necessarily get widely reported in the media. For instance, on the morning the report was released, Justin Wolfers, PhD, an economics professor at University of Michigan, tweeted: “Be careful: Household survey estimates can’t be compared with previous month because of changing survey weights.” Wolfers also observed, “Benchmark revisions were relatively minor, but suggest job growth through 2016 was a bit faster than previously believed (85k over the year).”

Another reason is that the labor force participation rate and other more nuanced data aren’t headline makers. As Murphy says, the BLS is already putting out a “whole slew” of numbers including broader metrics that include people who are discouraged and have left the workforce. That number is considerably higher, at 9.4%.

Then there’s the labor force participation rate, defined by the BLS as the measure of people 16 years old and over who are either working or actively seeking work. Murphy notes that the steady drop in the labor force participation rate was a previous concern. “That seems to have bottomed out,” Murphy adds. Indeed, the labor force participation rate rose by 0.2 percentage point to 62.9% in this last report. That technically leaves 37.1% of eligible workers out of a job.


However, that’s not always the case, as Murphy says. He explains that part of it could be due to a long-term shift in demographics. “We have more people retiring and choosing to leave the workforce,” he explains. Boosting the economic literacy of the general public would help, Murphy says, so that people could understand what’s behind the unemployment percentage.

Can The BLS Numbers Change Under Trump?

The BLS overhauled some of its statistical analysis during the 1990s, in part to dig deeper into the joblessness rate, recognizing that unemployment alone isn’t enough to measure how the workforce is doing. The new reporting came out in 1996. It’s important to note that was an election year, when the incumbent administration had a stake in the changing results. Yet because the BLS is an independent nonpartisan organization, it stayed on schedule.

Fast Company asked Wolfers if he thought that the BLS could change the way it reported its findings under the current administration. “My guess and hope is that it won’t change,” he replied, “but many people’s faith in this administration has been proved wrong so far.”

About the author

Lydia Dishman is a reporter writing about the intersection of tech, leadership, and innovation. She is a regular contributor to Fast Company and has written for CBS Moneywatch, Fortune, The Guardian, Popular Science, and the New York Times, among others.