The JOLTS data provide hree major challenges to any disruptions hypothesis of the Great Recession. That is, any hypothesis that says the United States was engaged in some form of economic production that was disrupted by event X and that led to a huge recession until we could get back on our feet. This includes the vast majority of structural theories.
The first problem is that job destruction fell dramatically during the Great Recession.
Second, Job Hires are typically very close to separations. However, in late 2008 the two broke apart.
Even still the gap, though enormous up to 700K was still small in comparison to the low point in hires 3600K.
What’s more is that third, if you look closely at the pattern of separations and hires, it looks like what happened was that separations stalled in late 2008 while hires continued its downward trajectory.
Here I just used a straight line but if you were to follow the logarithmic path of hires you would wind up with an even smaller gap and even fewer job losses to make up for during the recovery.
So, to really get at the heart of everything we not only have to explain why overall churn including separations and hires started to fall – that is why the employment structure of the economy started changing at a slower rate. We also have to answer why the stall
7 comments
Comments feed for this article
Sunday ~ January 15th, 2012 at 4:21 pm
teageegeepea
This post is incomplete.
Sunday ~ January 15th, 2012 at 4:36 pm
rjs
he does that a lot…thats his style…
Monday ~ January 16th, 2012 at 4:07 am
Economist's View: Links for 2012-01-16
[...] JOLTS Data and Theories of the Great Recession – Modeled Behavior [...]
Monday ~ January 16th, 2012 at 4:25 am
Links for 2012-01-16 | FavStocks
[...] JOLTS Data and Theories of the Great Recession – Modeled Behavior [...]
Monday ~ January 16th, 2012 at 10:51 am
Rick Russell
I don’t understand the complaint. In a healthy economy, most unemployment is frictional or structural, so you expect hires to closely follow separations.
In times of trouble, you would expect separations to remain high while new hiring falls, and that seems to be what your graphs show. Microeconomically, I think it’s easy to understand: hiring freezes come from executive mgmt in a panic, managers respond by trying to keep the workers they have but eventually the employees are lost without replacement, resulting in lag of the macroeconomic labor indicators.
Monday ~ January 16th, 2012 at 5:24 pm
scott f
Doesn’t the second chart show a similar pattern during the 2001 recession? As Rick pointed out, your first response is not to fire 1/2 your employees because the economy may turn around any minute and replacing workers is expensive. Instead, you hold your breath and avoid hiring any extra workers. There doesn’t seem to be any story here.
Tuesday ~ January 17th, 2012 at 7:27 am
RGregory
If the cause was structural i.e., current workers no longer having the correct set of skills given the new technology, then I don’t think it is amiss to expect that we would see fewer hiring freezes and managment more likely to get rid of the current work force (i.e., creative destruction). The data don’t seem to show that.