How did the closely watched jobs report get its start?
Its technical name is the Employment Situation Summary. Business journalists call it the monthly jobs report.
But it’s actually two reports, based on two different surveys. The Current Employment Statistics is a statistical sample of employers gathered by the Bureau of Labor Statistics. The Current Population Survey is a statistical sample of American households conducted by the U.S. Census Bureau for the BLS.
They get mashed together, giving us, from the CES, the number of nonfarm payroll jobs added to or subtracted from the economy each month, including details on payroll jobs in specific industries, average hourly earnings and hours worked. And from the CPS, we get levels of employment and unemployment among American households broken down by demographics, including age, sex, race and ethnicity, education level, veteran status and disability.
“If you want how many jobs there are out there, what kind of hours, what wages are being paid, geographic and industry detail, then the payroll survey’s the place you want to go,” said Cornell University economic adviser Erica Groshen, who served as BLS commissioner from 2013 to 2017.
“But you can’t find out who’s unemployed by talking to businesses,” Groshen continued. “So if you want to know how many people are looking for work, who’s not looking for work and why, then you need a household survey.”
So how did we get these two different datasets?
Let’s go back 140 years to 1884.
Americans were leaving farm work behind and moving to towns and cities to take up jobs in new factories. Work hours were long and working conditions were dangerous in many industries. Labor unrest was spreading.
“Because workers — in the form of nascent unions — and their employers were in the midst of a lot of industrial strife,” Groshen said, “the policymakers of the time realized that one important step towards resolving this tension was to create trustworthy data.”
So in 1884, Congress established the Bureau of Labor Statistics.
But at first it didn’t produce monthly counts of employees — now BLS’ bread and butter.
“When it was created, its main work was what they called special investigations, to look at what was called the ‘labor questions,’” said Thomas Stapleford, an economic historian at the University of Notre Dame. “These were puzzles about urbanization, growth of unions, rise of factory work, industrial unrest, waves of immigration and what today we might call business cycles — fluctuations with major depressions, and then booms in the economy, and then collapses again.”
It took one of those boom-bust cycles, the recession of 1913-1914, to kick the BLS into gear doing regular monthly surveys of employers.
Stapleford said the first report, based on data gathered from approximately 200 employers in late 1915 and published in early 1916, looked at one industry. “They had clothing and textiles — this was by far the largest. There were a couple of subdivisions within that,” including boots and shoes, cotton goods, cotton finishing, and hosiery and underwear.
(By the way, this is the industry my ancestors would have fallen under — around the turn of the 20th century they had a small pants factory in Philadelphia.)
By mid-1916, the BLS was surveying employers in several industries, said Stapleford:
“Iron and steel; automobile manufacturing and repair; and the last one is a weird one — cigar manufacturing. Between 1916, when they started, and 1919, employment in cigar manufacturing had declined by about 60%. So this wasn’t a growth industry.”
So now the BLS was gathering hard data on employment. But policymakers and economists could only guess about the level and impact of unemployment.
And then, in 1929, the stock market crashed.
“The initiation of the Great Depression really changed things,” said Harvard University labor economist Larry Katz. “There was a profound sense that we knew something very big was happening in the economy, that there were lots of people out of work, but we didn’t really have a good way of measuring how serious it was.”
So the government turned to another statistical survey.
“Questions were added to the Census of Population in 1930 to try to measure unemployment,” Katz said, “to learn about whether people were working, whether they were looking for work.”
Through the 1940s, measurement of unemployment improved as statistical methods for sampling and extrapolation were developed and refined by statisticians, economists and polling firms.
Nearly a 150 years after the BLS was founded, the payroll survey now includes 119,000 companies and government agencies covering approximately 629,000 work sites. The household survey taps 60,000 individuals and families.
There are now multiple categories of “unemployed” people. They include the “official unemployment rate,” people who don’t have a job and have actively looked for work in the past four weeks; involuntary part-time workers who can’t find full-time work (those “employed part-time for economic reasons”; and “anyone who’s out of work and tried to look in the last year, often called ‘discouraged workers,’” Katz explained.
When work patterns changed in the pandemic, BLS responded with new survey questions and statistical series.
“Since 2022, we now have very good data on the importance of remote work and how that differs across workers,” Katz said. “Alternative work arrangements and gig work is much tougher, and the BLS is trying to improve the questions to get at that.”
He said the goal is to better measure and track how Americans’ work lives and income are being affected by the growing gig economy.
There’s a lot happening in the world. Through it all, Marketplace is here for you.
You rely on Marketplace to break down the world’s events and tell you how it affects you in a fact-based, approachable way. We rely on your financial support to keep making that possible.
Your donation today powers the independent journalism that you rely on. For just $5/month, you can help sustain Marketplace so we can keep reporting on the things that matter to you.