Why are new unemployment claims so low?

Scott Sumner reports:

…the ratio of new [unemployment] claims to pop is roughly back to the boom levels of 1999-2000 and 2006-07.   And yet the other indicators (total jobs, unemployment rate, etc), remain deeply depressed.  I can think of two ways to interpret this data:

1.  Casey Mulligan is right, we have lots of structural issues that are causing high unemployment right now.  The job market’s not that bad, it’s just that lots of people don’t want to work at the wages being offered, or are frozen out by the 40% rise in minimum wages during the housing bust.

2.  AD is still the main problem, but since 1975 there’s been a long term downward trend in the claims/pop ratio, for some mysterious reason.  That trend would explain why (according to new claims) the labor market looked as good in 2006 as 2000, even though most people think it was not.

On this topic, here is a kebko post of interest; he argues that employment has more or less recovered, once we adjust for various obstacles.


2 doesn't seem that mysterious-- as manufacturing's share of the labor force falls, temporary layoffs due to strikes/retooling/overbuildup of inventory have fallen as well. Most non-manufacturing industries don't have temporary layoffs, but they are quite common in the manufacturing sector. Thus one would expect a long-term decline in initial jobless claims starting around when manufacturing employment peaked.

To fit into more of the ZMP model, an organization that layoffs of ten percent of workforce makes the other 90% of the force more important. Therefore that is one reason nomial wages did not fall.

Nominal wages have fallen for a huge percentage of the population.

They earned $100K, but now earn $50K.
They earned $50K, but now earn $30K.
They earned $30K, but now earn $12K.

Correct me if I'm wrong but didn't we previously learn that the new jobs have mostly gone to folks in their 50s? And aren't unemployment claims only good for people laid off? So this particular metric is not counting the legions of jobless college grads.

Or am I off base?

That's correct. The data shows stronger job creation for those 55 and older than 25-54. Part of it is demographics, but this has been a trend going back a decade so I don't think it can be linked to the current troubles.

What is the mystery? ~1975 is also the year wages started stagnating despite rising productivity. Once upon layoff meant temporary and most workers were brought back. Now layoffs mean you are gone for good. Anyway in productivity driven economy, when a firm layoffs their 10% of ZMP workers, the other 90% must now do more.

Not if you take the ZMP part seriously.

I've become much more willing to accept structural issues as an important problem. Complimenting Sumner's post, and contra Peter Orazag's latest, I argue the difference in shape of recovery in openings versus that in hires preempts supply-side explanations. More importantly, the deterioration to which Orszag refers as a cyclical phenomenon is both absent in previous recessions (to the extent JOLTS goes) and occurs too late in this one (late 2010) to be purely demand-side. More here: http://ashokarao.com/2013/08/14/a-disjoint-labor-market-and-the-two-recessions/

I liked your post and I tend to agree with you, which is probably why I liked your post. My hunch (I've seen no data on the topic) is that part of the problem is the gap between wage requirements and wages offered. My reason for thinking this is the number of middle-aged people I know who were laid off from a middle management job and either had to take a big pay cut or remained out of work for an extended period.

Yes – fact is it's rather absurd to claim supply-side factors are irrelevant over a fairly broad time period when technology and globalization exist. Cyclical downturn masked this for a while, but starting 2010-ish we're seeing some real changes that shouldn't be snubbed as a bad economy. Not because AD is healthy, but because these particular problems are not a sign of deficient demand.

Agree. The barriers to entry and re-entry into the state of being employed appear to me anecdotally to have risen substantially in recent years, faster than the marginal unemployed worker has been able to demonstrably change his or her marginal productivity. Obviously there is a theoretical argument that you keep increasing AD until it reaches a fever pitch where employers will do anything to hire the last marginal worker but a battering ram is a suboptimal way to open a locked door compared to a key.

Yeah, ignorant on these things but it doesn't seem that tough to me, to get unemployment insurance you have to have been employed. There are rules. If I'm fully employed, then lose my job, then take unemployment for two years followed by low wage work (not worth pulling unemployment for) or part time, or becoming a stay at home parent by default, or retiring early or getting disability, I'm not eligible for unemployment insurance again.

I would guess there will be a peak on people retiring and on people drawing disability (which will lag because people delay applying as long as they can and then the process takes years), after which you'll see a decline in new cases because everyone who can be on board is already on board.

But you'd think that would be accounted for by someone looking into it, wouldn't it?

This seems like a pretty good candidate. After 4 years of employment carnage, those who still have jobs are on average pretty smart about avoiding getting fired and employers have staffed as well as they can to meet the currently low level of aggregate demand.

That would fit pretty well with a Keynesian model of this recession.

Another possibility is a very slow recovery will mean that new unemployment claims will come down first but slow acceleration of job growth will mean that you won't see a rapid drawdown of the currently unemployed. Again that would fit will with a 'demand is the problem' view of things.

"That would fit pretty well with a Keynesian model of this recession."

Again, how is that exactly?

Keynes argued an economy did not necessarily gravitate to full employment. It could very likely find itself in an equilibrium below full employment for a very long time. What would it look like if such a thing happened?

Well as the initial transition from full employment to below full employment happened you'd see a lot of new unemployment claims. But as time wears on you'd see few new claims. If the 'new normal' is 10% unemployment for the next 20 years well you won't see lots of new claims. 90% of people will have jobs for 20 years after all! But the labor market will look horrible (after all, only 90% of people can find work leaving 10% out int he cold for the next 20 years!)

I can think of two ways to interpret this data

I can think of a few more:

3. AD is too low but after 4 years of carnage most people who were going to be laid off have been laid off. Those left with a job are good at not getting fired and employers are good at maintaining the right number of workers for the current low rate of demand. Hence you see low new claims for unemployment but not a huge pickup in currently unemployed being pulled back into the working universe.

4. AD is too low, we are in the 'long run' type of recovery Keynes said was possible where the slow depreciation of capital and assets causes demand to creep back up. This means essentially a long, slow recovery. So in a slow motion recovery from a deep recession you'd expect to see new lay offs stopping first. Between ceasing to lay off people and starting to actually hire some new workers, you're going to see a prolonged period of time so you get the supposedly paradoxical 'no new unemployment claims, but crappy labor market' observation.

5. There's nothing mysterious at all. When a recession stops, employers first stop firing current workers. They then will stretch their current workforce as much as possible to meet the demand of a recovery (two reasons...one is to make sure that the recovery is for real before hiring new workers, another is since they are presumably left with their most productive workers, they can make more product with their current workforce before demand overwhelms and forces them to expand hiring).

What's interesting to note is that theories 3,4 and 5 are consistent with a demand view of the problem. They also would tend to predict that one would expect to see current corporate profits being rather healthy. (In case 3 and 4, companies have already been wiped out by the collapse in demand, those left are able to sell to current demand. In case 5 you have companies enjoying a bit of a lag where they can meet the increasing demand without increasing payroll and hence reap more profits for a time).

Claims as a % of covered employment is 2.3%, near the historic lows.

This ratio has been below 2.0% only twice -- right before each of the last two recessions.

The only other times it was lower than this was right before the 1970 recession and in 1988.

The ratio of claims to population is a biased ratio.

The ratio of claims to population is a biased ratio.

Unpublished BLS data shows that over 100% of the recent pop in part time workers was federal government employees working part time because of the sequester.

Just curious, but how is it over 100 percent?

I've no idea about the details in this case, but if the non-fed-seq number was a decline, and the net inclusive of fed-seq was an increase, then "over 100% of the recent pop" was due to fed-seq.

AD? Can someone please spell it out for us non Econ Ph.D. types? Thanks.

Aggregate Demand. How much everyone in the economy wants stuff.

No, they can't.

No, they can't explain it because there is no model for unemployment insurance initial claims.

Indirectly, initial claims are driven by the unemployment rate. Simple enough.

But later, claims are driven eligibility for UI compensation and/or the belief in an entitlement to such benefits. Initial claims include claims that are denied. Te peak of unemployment, many people will file claims with no eligibility hoping to sneak into the system. Ineligible people try more than once. Eventually they tire of it. Eventually, eligible people begin to lose eligibility.

So initial claims have come down primarily because the unemployment rate is down and the economy has added jobs. Next, it is down because many people are not or are no longer eligible for benefits and do not submit claims. Finally, claims are down because illegitimate claims are drying up.

Changes in continuing claims is a better way to track net new eligible claimants. But continuing claims are also subject to a specified maximum duration of benefits.

Another issue which is more subtle is the ratio of benefit payments to the wage of an easily obtainable job, tempered by expectations. If you were at minimum wage and got laid off, unemployment is a large percentage of your previous wage for doing nothing. As you get closer to the end of your benefits, working looks better and better because you might not find a job the day after you lose benefits. Also, as wages rise, the value of working goes up relative to your fixed UI income.

Initial claims is too messy to draw too many sophisticated inferences as people here like to do.

One factor kebko does not consider is SSI. SSI disability claims have almost doubled over the past ten years. It's difficult to believe there's really that much more disability, and it's plausible that some older long-term unemployed might find SSI an attractive alternative to possibly futile attempts to re-enter the labor force.

Of course, SSI has existed since 1974, and the standard for SSI disability is very high. Nonetheless, there are no objective tests for some types of disability (for example, back pain and some psychological conditions such as PSTD).

Now, I'm NOT claiming that no one is totally disabled by back pain, or by PSTD- just that there's no objective way to test for these conditions. And therefore, it's to be expected that at least some claimants for SSI due to these disabilities either do not have them, or do not have them in a severe enough form to qualify under SSI's rather strict definition of total disability- and yet their applications are likely to succeed. And there seems to be increasing popular awareness of this, perhaps due to the incessant advertising by attorneys offering to represent SSI claimants on a contingency-fee basis.

A fundamental problem with the hypotheses that increasing claims for SSI disability are at least partly responsible for reduced labor-force participation is the difficulty in testing it, as claimants are unlikely to voluntarily admit fraud and yet fraud cannot be detected through objective testing.

BUT the difficulty in measuring this contributor to reduced labor-force participation does not invalidate it.

How about this, I bet there are tons of people who are now on SSI who were not on it before who have genuine and verifiable and qualified disabilities -- they just never applied before, they were working through it -- e.g. the 50 year old woman with MS who loses her job and finds she can't find another one. She uses up her two years of unemployment, but she legitimately is limited in her job prospects now, in part because of the economy but in part because of the disability. So now she is out of options, she applies.

I'd bet you could count them. And I bet that number is substantial.

That would require empathy from upper-middle class white men who haven't worked manual labor for 30 years and whose social network doesn't have a single individual without a advanced degree. It ain't gonna happen.

My post is a bit rambling, so you can be forgiven for missing it, but I do mention that many workers who used up EUI subsequently transitioned to SSI benefits of some kind. I think this is mainly a demographic issue, and the disabilities rolls will bulge and then decline again as baby boomers move through their 50s, 60s & 70s.

According to the Social Security Administration's statistics, easily available on its website, more than half of new disability claims involve one of just two types of disability - musculoskeletal conditions, which would include back and joint pain, and psychiatric conditions. These also happen to be the easiest types of disability to fake.

what's the big change that occurred over the past 6 years?

good question. Housing prices got clobbered and along with it consumer confidence and consumer willingness to spend based on debt. Feeds into AD component. I'm not sure how the financial crisis contributes to this on the corporate side, but possibly considerable deleveraging? (I don't know in aggregate, but my firm - insurance - went through years of writing bad debt off the books and getting financial ratios back in order).

Simplest answer is probably that what people think is a baseline isn't. Second simplest is that it is just math. You have to have gotten a job in order to have lost it. So, usually that doesn't dominate the new UE equation, but in a long recession it may.

I combined current levels and trends of LFP with census forecasts for the next 50 years to show how the LFP rate will behave purely as a result of aging issues.


kebko - thank you. Very interesting explanation using the "mix" of demographic profile. It reminds somewhat of an econometrics study I did back in college in early 90s looking at crime rates - and demographics shifts explaining much of the forward projection of falling crime levels. It was interesting that when crime rates kept falling in the 90s so many news stories follow attributed new law enforcement programs and resources when projections based mostly on demographic trends would've expected similar results.

It's interesting that as age increases the labor force participation rate of women seems to perhaps exceed that of men at the high end 65+. Seeing the projected increase in age 65+ working makes me think there's political hope for social security to be able to raise the withdrawal ages as live-spans increase. Fingers crossed anyhow.

Thanks, Shane. But, I might not have been clear about the future trends in LFP. For the other age groups, I followed current trends that are pretty stable (the trends for middle aged men have been linear for about 80 years). But, for the 65+ trends, I just made them up. And, I chose especially aggressive trends, so as to say, even if 65+ LFP increases at this aggressive pace, total LFP would still show marked declines. I was trying to bias the result against my preconceptions in order to be conservative. So, those trends don't represent any sort of analysis or prediction.

But, that is basically the rate at which 65+ LFP is increasing currently, mostly because the bulge of baby boomers is pulling the weighted average age of the group down, so I don't have any reason to believe it will continue for 50 years. You are right, though. If the baby boomer trend of working more into later years continues, it will help some of the entitlement issues tremendously.

"The job market’s not that bad, it’s just that lots of people don’t want to work at the wages being offered, or are frozen out by the 40% rise in minimum wages during the housing bust."

The job market may be "not that bad" for the currently employed looking for another job, or for people who've been unemployed for a year or less.

But I t think the job market's terrible for many people. That would certainly be true of the very long term unemployed who haven't worked since, say 2008-2011. The resume filters that companies use surely filter out their resumes in most cases -- and if they don't, what HR dept is going to bring these people in?

We also have high school dropouts and high school graduates who now are competing with people with more formal education.

I think there's a lot to worry about here. People refusing to accept lower wages after being unemployed for a year or two is the least of it.

I refuse to even address the idea that a $7.25 minimum wage is perceived as too high by employers.

It's pretty nasty, I agree. What bothers me about the high numbers on SSI, food stamps, early retirement, etc. is that while this was noted for awhile as a worrisome trend, it seems like in the "normal" news it's no longer treated as such. What you get is the outrage stories about abuse, which we've always had but which are returning now as our sense of being able to fix the situation gives way to our desire to find misbehavior that we can address. But other than that, we seem to be settling in to the feel that there's just going to be this segment that wants to work and won't be able to. Not that it hasn't always been so, it just goes invisible, but the segment is pretty substantial now. I am unhappy that we are becoming reconciled with our state as a nation, lots of government dependent wealthy, a mediocre and cowed middle class often doing make work and combating anxiety disorders, and a "working" class with increasingly no work.

On the up side, I have noticed an interesting trend, many of the un or underemployed have become tired of waiting, and they are getting out and doing a bunch of stuff. Volunteering at food banks, helping neighbors with kids, getting involved in local governance, etc. One family I know suffering from chronic unemployment (resolved now) put huge amounts of work into church and other community efforts, honestly did more work in the area than most employed. People will find a way.

People will find a way.

Yes, they will -- to survive.

But I want all of us to do so much more than that. I think you do, too. As prices continue to go up and wages and household income for so many stays stagnant or goes down, we may see things like groups of middle aged and older singles living together, we may see more and more long term multigenerational family living situations.

In other words, we will see a throwback to things we haven't seen while America has been a First World country, and it won't be on a voluntary basis.

Accepting the current trends, in my opinion, means we accept the seeds being sown for future social unrest and maybe eventual revolutionary change that the .01% and .001% will not like one bit.

Why is there any mystery here? This simply means that layoffs have returned to their pre-crisis levels. The fact that there are still a good deal more people unemployed than there were in 2007 has no sort of logical connection to that fact which should cause any head-scratching. Anyone with a job knows that there's a bit of trimming here and there, and the occasional firing for cause, as happens even in boom times, but we are certainly not seeing mass layoffs as we did in 2008-09.

A dramatic increase in the min. wage being the cause of a 'bad job market' while at the same time there are less new claims for unemployment would make sense only if you saw an increase in the labor force participation rate. If there really was a mandated 40% increase in wages you would see people who were not in the labor force (housewives/husbands, students, those living off their parents or friends) decide to come out and start applying for jobs. Yet we have just the opposite, a lower labor force participation rate.

Comments for this post are closed