How Hard Is It to Get Accurate Employment Numbers?

Every month, the Bureau of Labor Statistics releases a summary of how many people in the US are working and how many aren’t. You’d think that arriving at these numbers would be as simple as asking people to “Check ‘Yes’ if you receive a paycheck and ‘No’ if you don’t.” 

Turns out, it’s not that straightforward, and the BLS recently had errors in the March, April and May reports because some people who were counted as employed should have probably been considered unemployed (the BLS estimates this to have been about 4.9 million people in May).

The US government has been tracking unemployment for the last 104 years, so why can it be hard to nail down?

The confusion starts with the categories. When the BLS comes knocking in-person or by phone to take its mini household survey, you’re supposed to say that you’re employed if you were paid and worked at least one hour that week. If you didn’t work but looked for work in the past four weeks, or if you were temporarily laid off, you’re supposed to say you’re unemployed. Students, retirees and other people not in the workforce aren’t counted.

But there’s another part of the survey that may have been unclear and contributed to the recent errors: People who had been temporarily laid off or furloughed put “other” to explain why they weren’t working. This answer resulted in many of them being placed in the “employed” category, when they should have been considered unemployed. 

But wait, there’s more: not all jobs are counted. Job reports tell us, in theory, how many people are working and if they’re working in “farm” and “non-farm” jobs. In reality, the report doesn’t count certain employees, including some people who work for the government (appropriately, this includes CIA employees), private household workers and non-profit employees.

Why you should care: Despite the missing jobs and mix-ups about who’s employed and who isn’t, the data helps the Federal Reserve get a read on the economy’s health which has an impact on its monetary policy.

Plus, the report includes something called “continuing claims,” which has at least two uses. As described by Goldman Sachs researcher and chief US economist David Mericle, “continuing claims measure the stock of people receiving benefits, rather than just the inflow, so it does a better job of capturing both layoffs and hiring.” 

Another reason these claims are worth watching is their relationship to the GDP. “When continuing claims stabilize, when they reach their maximum and just kind of fall sideways, that’s probably when GDP has stopped falling,” he said.