Beware False Claims about Jobless Numbers

The Department of Labor just released its weekly update on jobless claims, reporting that:

For the week ending Jan. 24, the advance figure for seasonally adjusted initial claims was 588,000, an increase of 3,000 from the previous week's revised figure of 585,000. The 4-week moving average was 542,500, an increase of 24,250 from the previous week's revised average of 518,250.

In December 2008, weekly initial claims (4-week moving average) averaged 549,062, compared to the January 2009 average of 526,625. That decline of -22,438 on a monthly average basis represents the largest monthly decline since November of 2005 (-50,425).

The chart above shows average monthly initial claims as a percent of the labor force, from January 1980 to January 2009. (For January 2009, I have assumed that the labor force remained at the same level as December 2008, 154.447 million.) In previous posts (here and here), I have documented how the increasing size of the labor force over time distorts the frequent comparisons of today's jobless claims to the number of claims in previous years. For example, the labor force has increased +45% since 1980, from 106.78 million in 1980 to 154.5 million today, an increase of more than 48 million workers.

As bad as the average 526,625 jobless claims in January might seem, we're still nowhere near the jobless claim levels of 1982 or 1991, as a percent of the labor force. In 1991, jobless claims as a percent of the labor force peaked at .3915%, which would be the equivalent today of 604,660 jobless claims. And to be equivalent to the .6067% reached in 1982, we'd have to have 937,000 claims today, or almost double our current level.

WASHINGTON, Jan 29 (Reuters) - The number of people remaining on the unemployment benefits roll after drawing an initial week of aid, or continued claims, rose 159,000 to a higher-than-forecast 4.776 million in the week ended Jan 17, the most recent week for which data is available. The Labor Department said this was the highest reading since its records on this series began in 1967.

There's a little problem here. The size of the labor force has doubled since 1967, which distorts the comparison of today's continued claims to past years (the chart and post above illustrate this issue using the labor force vs. initial jobless claims). Consider 1982, when there were 4,713,000 continued claims (lower than today), but there was also a much smaller labor force (110.744 million) than today (154.447 million). As a percent of the labor force, the continued claims in 1982 represented 4.256% of the labor force. Given our labor force today, it would require 6.57 million continued claims to reach the same 4.256% level as 1982, or an additional 1.8 million people.

Bottom Line: Adjusted for the size of the labor force, we're still nowhere close to a record for continued unemployment claims. But that reality won't stop the media from reporting "record jobless claims," there are already dozens of new stories with that "false claim" about "record claims."

Comments

[handout:]

The difference between now and then is the fact that households have borrowed so they are dependent on duo income rather than single income

Jan 30 04:04 AM

[Steven Hansen:]

Mark, your point is well taken comparing unemployment numbers between periods.

note: total non-farm employment is currently 135,000,000 which is considerably lower than your number.

Jan 30 04:44 AM

[Al-USA:]

handout

Writer handout has a good point. I just read in a local newspaper the the up to 30% of the increase in consumer discretionary income was DEBT. Absent more jobs at higher paying wages, the only way out is even more DEBT for "joe sixpack". Well, when "joe sixpack" can be kicked out on the street at any moment, more debt is the last thing "joe sixpack" wants or needs to do.

Otherwise a good article, thanks.

Jan 30 06:07 AM

[Don Eremin:]

Your point is well taken, but you fail to recognize that a much smaller part of the labor force is currently eligible to collect unemployment benefits as a result of changes in the law over time.

Jan 30 09:51 AM

[knowwhere:]

Jobless claims don't take into account those that no longer qualify for unemployment. So, in reality, unemployment is higher than jobless claims. When you fall off the dole, or don't qualify because you just graduated from college and can't find a job (early 90's repeat), it isn't reported. Beware of statistics - they easily mislead

Jan 30 10:45 AM

[Stephen Kanitz:]

Mark, do us all the favor. Recalculate joblessness as % of labor force, call it the Mark Perry Index, otherwise the mistake will be repeated ever ana ever.

Jan 30 12:54 PM

[John Lounsbury:]

Mark - - -

You might want to look at a recent article I wrote comparing the jobless claims to a 52-week moving average. It gives some addition perspective to your discussion.
seekingalpha.com/artic...

Jan 30 01:26 PM

[Moon Kil Woong:]

There is a slew of reports touting the real jobless number. I suppose, with government statistics getting more and more unreliable due to changing factors and how it is calculated maybe the best is to take the number of working age people and subtract it from the ones who aren't employed. Wow you might say, that's a huge number. You would be right.

But, you might say, what about those people who want to be housewives or not work? They still are potential labor supply which we are really calculating. Rename it tappable labor supply for all I care. At least then you have a straightforward metric over different periods of time to measure.

If you want to factor in the population gowth just figure out the percentages. Unfortunately, no matter how we calculate the number the media will always try to sensationalize it. Real economists are interested in the real numbers, not the local TV commentator screaming the sky is falling every day it's not sunny outside.

Jan 30 01:49 PM