The nature of current unemployment

Dylan Matthews writes:

…Short-term unemployment is actually lower than it was in 2007. Indeed, the percentage of the labor force that had been unemployed for five weeks or less didn’t grow all that much during the economic meltdown. What changed was what happened after or within those five weeks. In 2007, they typically ended with a job. In 2009 and 2010, they more often ended with another few weeks of unemployment. The result is that if you break down the unemployment rate by duration, the problem appears to be almost entirely about long-term unemployment

There is more here, including a good picture.


I can't believe that he wrote that entire article without mentioning Extended Unemployment Insurance.

Considering that we have a regime shift with this recession where we had 99 weeks of UI available - the average duration of unemployment with extended insurance was 90 weeks, compared to 40 weeks or less in previous recessions - the burden of proof to explain the odd behavior of unemployment durations over 26 weeks is on how it could have resulted from something other than EUI. EUI is the obvious change in context.

It's a shocking example of the poverty of political discourse when we enact a policy that clearly will increase unemployment durations over 26 weeks, then when we get oddly lengthened durations over 26 weeks, we get a bunch of political discourse that doesn't even mention the policy, but frames the issue as if it requires some new solution, frequently in the form of more public policies that will increase long term unemployment durations.

Most people you speak of are simply choosing to run out the clock at 40% of their previous income for what reason exactly?

I wouldn't put it that way. I would simply say that most people 50 years and older have more discretion over their month to month employment status than they did when they were 30 or 40. This creates some natural pro-cyclical employment behavior, which plays out both through measured labor force participation and unemployment statistics. EUI is increasing this effect, which is making production and employment even more pro-cyclical.

But it's the 55+ who are increasing their labor force participation rates.

There are a lot of interesting moving parts there. 55-64 year old LFP had been increasing for 20 years or so, but then seemed to plateau at about the time of the recession. There might be a decent cyclical drop in LFP, but it's hard to know for certain if the timing of the plateau was coincidental. LFP behavior over time is anomalous in this age group. LFP for 65+ year olds continues to rise. The other age groups have very long term down trends. And, the much more important trend is that 50 year old baby boomers are becoming 60 year old baby boomers, and the LFP drops tremendously over these age ranges.

Older workers tend to have lower unemployment rates, but when they are unemployed, they tend to be unemployed for longer.

So, aging boomers are causing a natural decrease in LFP and there is some cyclical decrease also. But EUI causes the LFP and Unemployment Rates to increase. And, a high number of older workers who already have a tendency for long unemployment durations combined with extensive EUI policies probably also increases the cyclical portion of the Unemployment Rate.

"And, a high number of older workers who already have a tendency for long unemployment durations combined with extensive EUI policies probably also increases the cyclical portion of the Unemployment Rate."

So, by cutting consumption of unemployed workers, the decreased demand for existing production will increase the labor employed to produce less production because older workers will take jobs at much lower wages out of desperation which will cut both labor costs over all, and thus cut demand over all.

Talking about unemployment without talking about consumption is simply waving one hand; economists are supposed to keep in mind both the one hand and the other hand - production and consumption.

Economics is zero sum. labor plus capital equals production which equal labor consumption plus capital consumption.

The idea that capital consumption can go down to increase profits because government unwisely invests in such things as public health and bridges so tax should be cut, but business should not consume to provide bridges and public health because there is no way to profit, somehow results in a better sustainable economy.

Re: Considering that we have a regime shift with this recession where we had 99 weeks of UI available -

In some states we do, and in some we don't. And those extensions of UI are nothing new: they've happened in previous recessions too.

By the way (Re, your comment below), a 30 years old probably has more discretion about working than a 50 year old does. The 30 year old likely still has living parents he can move home with. The 50 year old may not. And the 50 year old probably has a mortgage and other obligations (maybe even minor children) that the 30 year old does not.

And given that there are still three job seekers for every job opening (normal is 1 to 1) why is it hard to believe that there is still a job crisis, and UI is not what is causing it?

There has been a recession, there are demand and supply side problems, etc. Nobody is denying that.

But, UI extensions averaged 90 weeks this time, compared to around 40 or less in previous recessions - it's a big difference.

Whatever the cause, in all economic contexts, unemployment duration rises with age, education, and income. An objective review of the labor market should account for this.

Since EUI and the current demographic makeup cause the number of job seekers to be statistically overstated, it is important to try to account for these effects if we are referencing the statistical number of job seekers as a measure of the crisis. Making these adjustments doesn't make the entire recession go away, and it might not even make the recession comparable to previous recessions, but it's a big enough effect that a discussion of the scale of this recession without these adjustments is suspect. And, to the extent that there is a positive feedback between these measures of labor market distress and some of the proposed policy reactions, it is especially important to get these adjustments right, because if we don't we're likely to be like the 19th century doctor who decides that we need more blood letting.

Why did we see unemployment lengthening out to 40 weeks in previous recessions? During the most serious recession before this one, the early 80s recession, the median duration of unemployment was a mere twelve weeks-- less than the 6 months that even standard UI provides.
Also to be considered: fewer than 40% of the unemployed receive any benefits at all, and for those that do the typical benefit is about 1/3 of their previous salary-- with no healthcare or other benefits. The notion that people are living it up under those conditions is quite frankly absurd.

When you only get 26 weeks of unemployment benefits you find a job in 5 weeks or less, but if the unemployment rate is really high and the benefit is extended to 99 weeks, you decide to stay unemployed to drive up the unemployment rate so you can be in the long term unemployed without any unemployment benefits?

Obviously economist understand individuals are irrational when it comes to making life decisions related to long term financial well being, so nothing economists propose that puts the responsibility for paying for retirement or illness has any chance of working.

"if you break down the unemployment rate by duration, the problem appears to be almost entirely about long-term unemployment"

OR maybe the problem is that we had a MASSIVE demand shock in the financial crisis/Great Recession five years ago from which the economy has yet to recover from. The long-term unemployed (and policies like extended UI benefits) are not the primary cause of our current labor market problems -- more like symptoms of a problem that is ongoing. I agree the chart is an important one, but it needs careful unpacking.

I would argue that we still face a shortfall in demand and should not view the long-term unemployed as particularly unemployable/unproductive (not a bunch of ZMPs). Here's one recent study that finds manufacturers don't face a shortage of skilled workers but rather a shortage of demand:

The article in the post closes with a discussion of hysteresis, which is a serious concern, but there is no indication that the US economy is now operating back at its full potential. I do agree all this underscores how important a rapid response to a short fall in demand is. Different policies for those who lost or might lose jobs, such as wage insurance or Germany's employment stabilizers might have reduced the long-term unemployed, but that's second order to an economy getting back on track.

In any case, I think the idea that 'all is back to normal but for the long-term unemployed' is misleading ... All is not back to normal and the long-term unemployed are one of our many reminders.

Great points, Claudia. I think your evidence on demand shortfall may be closely related to your work on income expectations (

If lower income expectations are temporary, maybe a recovery in demand is likely and, perhaps, not too far away. But what if, instead, "the large, unexplained shock to income expectations might suggest a permanent change in households' views-a phenomenon that would continue to weigh against a recovery in consumer spending"? What is the policy prescription in this case?

Is that consistent with investment being down but spending being at all time highs and basically back to trend?

We already know that most employers treat long term unemployment as a strong negative signal, even to the point of overpowering relevant credentials. Thus when the Fed failed to get NGDP back to trend within a couple years, many continued to camp on the floor without their musical chair. This just goes to show how important it is to return to trend as soon as possible.

How could the Fed have gotten individual borrowing to fund consumption they could not afford to pay for then or in the future restored to what it was in 2005?

Doesn't the Fed need the Congress to borrow trillions more than they have in the past five years and then do helicopter drops of the money from the Fed monetizing all the borrow and spend recklessly to hike GDP?

Is the Fed responsible for the Republicans suddenly deciding that a Democratic president means Republicans must stop being "tax cut, borrow and spend"-ers and switch to "cut taxes and stop spending anything"-ers? The latter is the policy of Republicans in 1993-2000 and 2009-present, and the former from 1981-1992 and 2001-2008. How does the Fed control the Republicans?

Which is trend: "tax cut, borrow and spend" or "cut taxes and stop spending anything"?

Comments for this post are closed