Once again there is no new economic data, so let me follow up some more on the issue of longer term unemployment.

Earlier this week I pointed out that just as initial claims lead continuing claims, so does short term unemployment (under 5 weeks) lead long term unemployment (15 weeks and over). Think of unemployment as a pipeline, and the intake flows before the main body of the pipeline.

Yesterday I followed up by noting that just as initial claims lead the unemployment rate, so continuing claims lead long term unemployment.

Because one of my persistent themes over the years has been that consumption leads employment, let me take a little time and briefly examine a variant of this: does consumption lead *unemployment* as well?

The answer is, generally speaking, yes it does.

First, let’s look at the YoY% change in real retail sales going back 75 years to the inception of the series in 1948 (red) and compare it with the YoY% change in unemployment for 15 weeks and over (inverted, blue):

(Click on image to enlarge)

(Click on image to enlarge)

With the sole exception of the 2001 recession, where the consumer barely participated at all, the answer is pretty clear: while real retail sales are noisier, they clearly turn up and down before longer term unemployment turns.

Now let’s look at the post-pandemic record:

(Click on image to enlarge)

Once again real retail sales turned down YoY about 6 - 9 months before longer term unemployment rose. 

Next, let’s compare real retail sales with continuing claims, which as noted above, lead longer term unemployment as well:

(Click on image to enlarge)

Although sales are much noisier than continuing claims, and there is clearly no one-to-one relationship, in general sales turn up or down coincident with to slightly before the turn in  continuing claims.

Here is the post-pandemic look at that relationship as well:

(Click on image to enlarge)

Sales decayed first, although both crossed the zero line coincidentally. 

YoY real sales have been getting “less negative” and turned slightly positive in September. October’s retail sales will be reported next Wednesday. If the recent improving trend continues, this would be more confirmation suggesting that long term unemployment is not going to significantly worsen in coming months, contra the historical pattern highlighted by Cullen Roche and others.

More By This Author: