Because coronavirus jobs were lost this week vs last, monthly jobs data may get weird

On Thursday, we got the first jobless claims report that was noticeably affected by closures related to coronavirus. But it wasn't as much of a leap as you may think.

In the week ending March 14, the advance figure for seasonally adjusted initial claims was 281,000, an increase of 70,000 from the previous week's unrevised level of 211,000. This is the highest level for initial claims since September 2, 2017 when it was 299,000. The 4-week moving average was 232,250, an increase of 16,500 from the previous week's revised average. This is the highest level for this average since January 27, 2018 when it was 234,500. The previous week's average was revised up by 1,750 from 214,000 to 215,750.

That September 2017 figure was in the wake Hurricane Harvey, but that only hit 299,000 and quickly declined below 250,000 within a month. By comparison, the next jobless claims report seems likely to jump by a massive amount

In Ohio, more than 48,000 people applied for jobless benefits during the first two days of this week. The tally during the same period the prior week: just 1,825.

In neighboring Pennsylvania, about 70,000 people sought unemployment aid in a single day — six times the total for the entire previous week.

Jobless claims are surging across the U.S. after government officials ordered millions of workers, students and shoppers to stay at home as a precaution against spreading the virus that causes the COVID-19 disease.

And the same is happening here in Wisconsin.

The effect of the coronavirus outbreak on businesses is perhaps most seen in the effect on the state's workforce. Nearly 63,000 Wisconsinites filed unemployment claims this past week, according to preliminary figures by the Wisconsin Department of Workforce Development.

That's 12 times the number of claims the state received during the same time period in 2019, when 5,190 claims were filed.

About 15,000 claims have been filed every day since Wednesday, according to the data.

What's worth remembering in the coming weeks is that the timing of the closings and new jobless claims may lead to some misleading numbers in the coming weeks, where a severe virus-induced recession will be apparent to most Americans, but might not completely show up in the data.

Why do I say that? First, because the monthly survey for the monthly jobs report took place last week, since it takes place during the week that has the 12th day of the month. Many of the new layoffs won't be reflected in the March jobs report as a result.

In addition, March was a month where hiring started for the 2020 Census. If we take 2010 as an example, federal government employment rose by around 50,000 for this month, and then rose by another 65,000 in April before the big jump of 435,000 happened in May.

Also you will notice that what goes up goes back down, which means that job losses in June, July and August may end up being larger than normal. I have a hard time believing that the ripples of the upcoming job losses will be over by then, so the end of Census work will likely make the jobs figures even worse at a time when it needs to start bouncing back.

There's a lot going on, as you are well aware. But the shuffles in jobs and the wild changes in the data are things we all need to have some perspective on, because there will be a lot of spin that will confuse a lot of people that want an idea about just how bad things are in the real economy, on top of the plummeting stock market.