The Washington Post published “5 Myths about unemployment” by Heidi Shierholz of the Economic Policy Institute. The article is indeed full of myths, though not in the intended sense.

“There are now roughly five unemployed people for every available job,” says Ms. Shierholz, adding “there literally aren’t jobs for four of every five unemployed workers.” That statement has been repeated endlessly− in recent columns by Paul Krugman and Art Laffer, for example, and in a July 20 Wall Street Journal editorial which said “there are still five job seekers for every job opening.”

Regardless how often you hear this, the statement is completely false. After all, the same survey that showed only 3.2 million “job openings” in May also showed 4.5 million people were hired that month. If 3.2 million “openings” measured all available jobs, as Ms. Shierholz claims, then how did 4.5 million get hired? I exposed this myth and others in my June 10 Wall Street Journal article. “The myth that there are nearly six job seekers for every available job,” I wrote, “arises from the misnamed BLS ‘Job Opening and Turnover Survey’ (JOLTS), which asks a few thousand businesses how many new jobs they are actively advertising outside the firm. But note well that this concept of ‘job openings’ does not purport to include ‘every available job.’ On the contrary, it is closer to being a measure of help wanted ads. ‘Many jobs are never advertised,’ explains the BLS Occupational Outlook Handbook; ‘People get them by talking to friends, family, neighbors, acquaintances, teachers, former coworkers, and others who know of an opening.’ Because many jobs are never advertised they are also never counted as job openings! The BLS Handbook also notes that, ‘Directly contacting employers is one of the most successful means of job hunting.’ Those jobs are also not counted as job openings.”

Unfortunately, I apparently failed to persuade even the Wall Street Journal editors about this statistical hoax. So, let’s get a second opinion.

The Minneapolis Fed recently interviewed Stanford economist Robert Hall, the famed co-author of Hall-Rabushka flat tax and (as he once told me) a “Clinton Democrat.” For 32 years he has chaired the NBER committee that defines the dates of business cycles. If he’s not an expert, who is?

Hall noted that “there’s been a decline in the profitability of hiring a worker without a corresponding decline in the wage. The incentive to create a job is the difference between what a worker will contribute to the business and what the worker has to be paid.” But he also noted that the difficulty of finding a job is not just because fewer jobs are created, but because employers “do relatively little to try to recruit workers” when unemployment is high: “Interestingly, the number of people who find jobs each month is more or less a constant…,” said Hall, “So, something like 4 million people find jobs every month. Even with 10 percent unemployment, as recently, we’ve still seen the same thing. A very large number of people looking, very low job-finding rate for each individual, but the product—the number of jobs filled—is roughly a constant. It’s a very important fact about the labor market. Think about a slack market from an employer’s point of view. They see there are all kinds of highly qualified people out there they can hire easily, so they don’t need to do a lot of recruiting— people are pounding on the door.”

When job seekers are pounding on the door, the number of advertised “job openings” is a useless indicator of the much larger number who actually find jobs. If the Washington Post were really interested in exposing myths about unemployment, they could start by debunking the myth that the “job openings” survey means “there literally aren’t jobs for four of every five unemployed workers.” That is literally hogwash.