When the U.S. jobless rate dipped in September from 8.1 percent to 7.8 percent, opponents of President Barack Obama decided the books had been cooked.
“In a move that is simply beneath the dignity of the Oval Office,” declared conservative talk-show host Sean Hannity, “it appears that the Obama administration altered, for political gain, the monthly jobs report.”
The fact that economists on the left and right dismissed that notion made no matter. Obama’s critics said the fix was in.
“When you look at the hard data,” said Hannity, “there is no way to explain how the rate decreased from 8.1 percent to 7.8 percent.”
Actually there is, but we’ll get to that in a minute.
As the nation looks to goose a sluggish labor market, few pieces of economic data are as scrutinized — and misunderstood — as the monthly jobs report.
Calculated by the Bureau of Labor Statistics, the unemployment rate is the measure that grabs headlines and resonates most with Americans worried about economic security. In the hands of politicians, it is an effective cudgel, used to batter opponents.
The rate is produced each month by government economists working under tight security.
The Washington Post reported this year that in the week before the release, there are daily confidentiality agreements, computer encryptions and a work space that is off-limits to anyone who doesn’t have clearance.
When BLS staffers step out to use the bathroom, they lock their data in a safe.
Two different surveys
The agency relies on two surveys to produce its report — and this is often where confusion begins.
One survey samples about 65,000 households, asking people about their job status. Data from the household survey are used to calculate the unemployment rate: the percentage of people who are out of a job and looking for work. That last part is crucial: You’re not considered unemployed if you’re not searching for a job.
The second survey focuses on 140,000 businesses, asking them about total jobs, earnings and hours worked. From this, BLS estimates the number of jobs added each month.
Typically, the different surveys move in concert. Job gains push unemployment down. Job losses nudge it higher.
But sometimes, the numbers are funky — which brings us to September.
The government reported 114,000 new jobs, a modest increase.
But the household survey found the number of people who said they were working jumped by 800,000. That’s what pushed the jobless rate down and prompted accusations of tampering.
Economists say it’s not unusual for the two surveys to produce different numbers because they measure different things.
For example, the self-employed don’t show up in the survey of businesses, but they are counted as working in the household survey. And respondents are counted as employed in the household survey even if they’ve worked just an hour during the survey period.
The monthly numbers are a first pass, and they’re often revised. September’s job creation, originally set at 114,000, was revised to 148,000. August’s figures were revised to 192,000, up from 142,000.
Mark Vitner, a senior economist for Wells Fargo, said he thinks the September report overstated the strength of the labor market. But he has little time for conspiracies.
“I am absolutely certain,” he said, “that the unemployment-rate numbers were not tampered with in any way.”
Even so, some conservatives have suspected as much for years.
They have pushed the idea that Obama changed the way unemployment was calculated to artificially lower the jobless rate. Economists say that charge is untrue.
The most common myth? That only people receiving jobless benefits are counted as unemployed. This one is so widespread, BLS debunks it on its website.
BLS points out that many laid-off workers aren’t eligible for unemployment or they exhaust their benefits before finding a job. So “clearly,” the site says, “UI [unemployment insurance] information cannot be used as a source for complete information on the number of unemployed.”