They're Going to Try to Scare You Yet Again. Don't Let Them.

The first installment of this blog topic was related to the reporting of initial jobless claims. I explained that the measure represents the number of people who filed for unemployment insurance for the first time. I also noted that the media would try to aggregate the number across multiple weeks, and misrepresent it as “total unemployed.” They did indeed do that, and they continue to.

Note again that total unemployed is the number of people in aggregate that are out of work as of any reporting period. Initial claims are reported weekly, every Thursday, for the week ended the previous Saturday. Total unemployed are reported monthly, as part of the overall jobs report that includes the unemployment rate, total non-farm payrolls, and other key labor market metrics. That release generally prints on the first Friday of each month, for the prior month.

As a reminder, the reason aggregate initial claims over a period of time don’t represent total unemployed, is that some of those people may have found jobs and are no longer unemployed. That’s especially true in the current environment, in which many service sector employees were laid off or furloughed with the promise of being re-hired when restrictions were lifted. That happened relatively quickly, thus initial filings have been dropping steadily since peaking the second week after the shutdown. (For that matter, so have total unemployed, as people have gone back to work.)

Tomorrow morning, Thursday, July 16, initial claims for the week ended July 11 will be released. Last week’s number was 1.314 million. That’s down substantially from the peak level of 6.867 million which was reached the week ended March 28. (You may recall that I initially predicted that the curve for initial claims for this downturn would resemble El Capitan, with a sharp, cliff-like spike to the peak over a short period of time, followed by a relatively sharp decline that might become more gradual over time. In most economic downturns, that curve looks like the Matterhorn, with a much more gradual rise to the peak, followed by a more gradual decline.)

The consensus (of a bunch of economists) forecast for the number to be released July 16 is 1.250 million, which would mark the 15th consecutive weekly decline in initial claims. This makes sense, as more and more reopening businesses are hiring people back. The employed suddenly became unemployed; the unemployed are now becoming employed again.

Here’s where the media will try to scare you (besides aggregating total initial claims and falsely claiming that it represents total unemployed, as I promise they will).

The number may come in higher than the previous week, instead of declining further. The media will then claim that that means the economy is worsening again, and that the nascent recovery has “fizzled” (I’ve already seen that claim made by one media outlet). But if the number does come in higher, that’s not what it means.

Several states have recently rolled back the reopening of some businesses in their state. California closed indoor dining, bars and movie theaters. New Mexico closed indoor dining. Other states have followed suit, but on a far more limited basis than the initial complete and total shut-downs of virtually all service sector businesses. (These moves have been made in response to increasing confirmed cases in many states, which are for the most part a function of increased testing, and without regard to declines in hospitalizations and deaths, reduced hospital stays, increased capacity of the health care system, and other positives. However, let’s not scamper down that rabbit hole in this post.)

So a higher number for the next release does not mean that the economy is getting worse than it was back in March. It simply means that people are re-filing first-time claims. Let me explain.

Sarah is a hostess at the fictitious Osteria Mezzaluna in Santa Monica. While she’s a great employee, she was hired in January, so she has less tenure than any other employee. When California closed all restaurants for inside dining in March, Sarah was laid off, but with the promise of hiring her back when the restaurant reopened. So the week of March 28, she made her first filing for unemployment benefits, and thus was counted among the 6+ million initial filers that week.

Subsequently, she filed claims each week until the restaurant reopened and she was hired back. Thus for those weeks she was no longer counted in the initial claims number, but was instead counted as part of continued claims, which tracks ongoing filings after the initial week. She was also counted among total unemployed for each month she was out of work.

Once the restaurant reopened for indoor dining, Sarah was hired back. At that point, she was no longer unemployed, so she was no longer eligible to file claims for unemployment benefits. Thus, she would no longer be counted as part of continued claims (nor, for that matter, among total unemployed). At this point Sarah is part of the improving trend we’ve seen in these measures: she is back among the ranks of the gainfully employed.

Then, in July, Gov. Newsom closes restaurants for indoor dining again. Still among the least-tenured employees of the restaurant, and given her position as a hostess, Sarah is once again laid off temporarily, with a promise of being re-hired when the restaurant again is allowed to resume indoor dining.

At this point, Sarah will again file a first-time claim for benefits. How is that possible?

The definition of a first-time claim is that it is the first time the claimant files for that period of unemployment.

During my youth, I once found myself let go from a job. So I filed for unemployment benefits, and the first week I filed, I was counted among initial claims. I have not found myself eligible for unemployment benefits since. However, were I to lose my job tomorrow, I could file for benefits, and that first week’s filing would counted among initial claims for the week. It’s not the first time in my life that I filed a claim, it’s the first time for that period of unemployment.

This is how it usually goes, as most people don’t go from being employed, to being unemployed, to being re-employed, to being unemployed again, in a matter of fifteen weeks. However, this situation is very different, and much more fluid.

Thus, if we do see initial claims rise with the next release (or the next, or the next), it’s related to the retrenchment of those states’ reopening plans. What it means is that, for the most part, the same people who made first-time filings in late March or early April went back to work, but now find themselves temporarily unemployed again, and are filing a first-time claim for a new period of unemployment.

Is this good for the economy? No, it most surely is not. However, does it mean that an entirely new cohort of people are now unemployed, and the recovery is over (as the media will most surely try to convince you)?

No. It means that some states have re-closed some businesses (on a far more limited scale than what we saw with the initial shutdown in March), which has resulted in some of the same people who were laid off at that time, being laid off again. Temporarily. The extent to which this is bad for the economy is dependent on how long “temporarily” is, as well as how many other states may follow suit. But the initial number will not mean what the media (and politicians of a certain persuasion) will have you believe. And note that, as this is happening, many businesses in other sectors of the economy that remain open and continue to recover are hiring back staff, so we may continue to see net improvement in the weekly claims numbers, in spite of these states’ retrenchment.