‘How do they do that?’: The behind-the-scenes on the unemployment report

Editor’s note: The following is an excerpt of a story first published at TalkBusiness.net, a content partner with The City Wire.

Pop quiz: For May, the U.S. unemployment rate was:


A) 4.5%, 
B) 9.4%
. or C) 16.4%

While the official answer was B, you could very well have added “D) All of the above” and chosen it. It all depends on how you – or, more precisely, the government – define the terms “employed” and “unemployed.”

DEFINITIONS
Employed is the easier of the two. For starters, if you’re at least 16 and have a job, you’re employed.

Simple, yes?

But you can also be considered as employed if you’re not drawing a paycheck. For instance, according to the federal Bureau of Labor Statistics, if you did at least 15 hours of unpaid work for a family enterprise owned by someone in your household – hung sheetrock after school for Dad’s construction business, say – then you’re considered employed.

Likewise if you have a job but are temporarily off work because of illness, bad weather, a strike or “various personal reasons,” you are still counted among the employed.

Now, for the other side of the ledger: In the simplest of terms, a person is counted as unemployed if they had a job, lost it through no fault of their own, and are trying to find another. Such a person is usually also receiving unemployment insurance benefits from the state.

That said, it is also possible to be without a job and not be counted among the rolls of the unemployed.

For starters, discouraged workers – those folks who are not working and are also not looking for work because they believe there are no available jobs for which they qualify, are not counted among the unemployed because they aren’t looking.

There’s also the much larger class known by the Bureau of Labor Statistics as those “marginally attached to the labor force.” That includes people who want a job, are available and have looked in the past 12 months, but are not currently employed.

THE UNDEREMPLOYED
There’s even a category of job-holders that some economists argue ought to be included in the ranks of the unemployed: The underemployed – people who want full-time employment but have only been able to secure a part-time job.

Some economists believe the monthly unemployment figures are inaccurate because they leave out some or all of these neither-fish-nor-fowl categories. In fact, were you to round up all of those folks in the above examples and count them with the unemployed, the national jobless rate for May would have been a whopping 16.4%.

It’s also possible to have a much narrower interpretation of unemployment. For example, the 4.5% figure referenced above was arrived at by counting just the number of persons jobless for 15 weeks or longer and applying that figure as a percentage of the labor force.

WHO’S COUNTING?
Another key to understanding the jobless rate is knowing how the government goes about collecting unemployment information.

Here’s a hint: They don’t actually go out and count everyone who’s not working.

“There’s a perception some people have that the numbers are calculated from [unemployment benefits] claims or a survey of everybody,” said Greg Kaza of the Arkansas Policy Foundation. “That’s not accurate.”

Rather, both federal and state unemployment figures are based on much smaller monthly samples, gathered during telephone surveys conducted by the U.S. Census Bureau for the Department of Labor during the week of the month that includes the 12th – the “reference week,” according to Kimberly Friedman, communications director for the Arkansas Department of Workforce Services. She said about 800 Arkansas households are called monthly and asked who in the household is or is not employed.

Speaking of, the state is paying out between $12 million and $13 million in unemployment benefits per week to some 49,000 people, Friedman said. That’s double from a year ago, when it was about $6 million a week to about 25,000 people.

The telephone survey isn’t flashy, but it’s serviceable, and Kaza said “they’ve been doing it that way since 1940.”

But even if you understand how the data is gathered and how the government defines joblessness, Kaza said, there are some other important things to consider when you look at the unemployment rate. Number one: Don’t think it’s a real-time measure of the economy.

“The unemployment rate is a lagging indicator,” he said. “I don’t encourage people to rely on the unemployment rate as an indicator of where we’re going in the economy.”

Drawing upon records from the Bureau of Labor Statistics, Kaza produced a chart showing the monthly national unemployment rates from 1948 through May of this year.

After going through and highlighting the months during which the economy was in recession, Kaza noted that unemployment generally didn’t hit its peak until after the recession had ended — sometimes a couple of months after, sometimes a year or more later.

“Over-relying on the unemployment rate is akin to a hitter standing at the plate and wondering how that third strike got by him,” he said. “He was expecting a fastball or curve, and a slider came by.”

Kaza instead looks at the payroll numbers released on the first Friday of every month, as well as the data on industrial production, when he wants to find clues about trends in the economy.

It’s hard to argue that getting more data is a good thing. But here’s a kicker: Sometimes when unemployment is on the rise, businesses are simultaneously reporting an increase in the number of people they’ve hired. What gives?

Bernadette Coleman, manager for Bureau of Labor Statistics programs for the state, said the conundrum happens more often than most people realize. And it’s not really all that much of a contradiction, she assured.

“It’s coming from two different places,” Coleman said.

In addition to the household survey that asks people if they’re currently employed or not, there is another survey done by the Bureau of Labor Statistics where employers report if they’ve increased or decreased the number of people on their payroll from the last month.

That employer survey can seem to contradict the official unemployment number, but that’s because it is only one facet of the equation used to determine the jobless rate.

(Link here for the full story as originally posted at Talk Business.)