As bad as the weekly jobless claims numbers have been, at least they have trended in the right direction. Until now, anyway. This week’s initial jobless claims number shows only a slight decline from the previous week, suggesting that the employment situation remains very unstable.

For the 13th straight week, the number comes in at over a million claims, a level not seen before the COVID-19 pandemic:

In the week ending June 13, the advance figure for seasonally adjusted initial claims was 1,508,000, a decrease of 58,000 from the previous week’s revised level. The previous week’s level was revised up by 24,000 from 1,542,000 to 1,566,000. The 4-week moving average was 1,773,500, a decrease of 234,500 from the previous week’s revised average. The previous week’s average was revised up by 6,000 from 2,002,000 to 2,008,000.

The advance seasonally adjusted insured unemployment rate was 14.1 percent for the week ending June 6, unchanged from the previous week’s revised rate. The previous week’s rate was revised down by 0.3 from 14.4 to 14.1 percent. The advance number for seasonally adjusted insured unemployment during the week ending June 6 was 20,544,000, a decrease of 62,000 from the previous week’s revised level. The previous week’s level was revised down by 323,000 from 20,929,000 to 20,606,000. The 4-week moving average was 20,814,750, a decrease of 1,092,000 from the previous week’s revised average. The previous week’s average was revised down by 80,750 from 21,987,500 to 21,906,750.

There are a couple of spots of relatively positive news. The four-week average still declined by 234,500, which was roughly the level of weekly claims before the forced shutdown of economic activity in the pandemic. The rolling average is usually more reliable than a week-on-week comparison. Also, it’s clear that people are coming off of unemployment at a rate faster than they’re being added, but that rate has slowed down this week, too.

Economists expected a lot more improvement, especially perhaps in light of the good retail-sales numbers and certainly after the May jobs report. This one-week result looks a little like a plateau at a level no one wants to see:

In addition to the standard filings there were 760,526 claims filed under the Pandemic Unemployment Assistance program, an increase of more than 65,000, according to unadjusted numbers.

The total of those receiving benefits to 29.1 million as of May 30, which was the most recent week for which data was available. That was a decrease of 375,522 from the previous week. …

The elevated claims number persists even as all states have reopened to various extents and nonfarm payrolls grew by 2.5 million in May. Before the coronavirus, the record for a single week was 695,000 in September 1982.

“The 58K drop in claims this week is very disappointing, given that the level still remains so high; the worst single week after the crash of 2008 saw claims at 665K,” said Ian Shepherdson, chief economist at Pantheon Macroeconomics. “It’s not clear why claims are still so high; is it the initial shock still working its way up through businesses away from the consumer-facing jobs lost in the first wave, or is it businesses which thought they could survive now throwing in the towel, or both? Either way, these are disappointing numbers and serve to emphasize that a full recovery is going to take a long time.”

It might take a while to get a clear picture of the labor market, too. A strange gap has opened up between the jobless claims and the Bureau of Labor Statistics’ numbers of sidelined workers. CNBC reported two days ago that economists aren’t sure what to make of it, other than as an indication that our systems of measurement might not be up to the stress of a black-swan event like COVID-19:

Amid the recent good news about stronger-than-expected job growth in May, some economists have eyed one labor statistic with a sense of bewilderment.

Specifically, the share of people getting jobless benefits last month exceeded the share of unemployed workers. …

There are currently 21 million unemployed Americans but roughly 30 million Americans collecting unemployment benefits.

“Now there’s a disconnect between those two numbers that didn’t exist before,” said Michael Farren, a research fellow at George Mason University.

This phenomenon does have some explanation. For one thing, the CARES Act expanded jobless benefits to some of the self-employed, which means that the two data sets aren’t identical. BLS requires people to be looking for work to be considered “unemployed,” but for now, the CARES Act doesn’t require a job search to be eligible for benefits. That means jobless claims and unemployment benefits are getting paid out to people in both BLS categories of “unemployed” and “discouraged” workers. That latter group might be as large as 6.3 million people, CNBC notes.

However one calculates it, the fact that we’re still seeing seven-figure reports for initial jobless claims means that we’re still seeing a meltdown in the jobs markets. Either we will need to accelerate the reopening, or we will have to find new ways to subsidize closure. This report shows we’re still doing too much of the latter.