Structurally impaired jobs vs. non-impaired jobs

Here is some text from the post:

One way to look at the US job market is to break it up into two components: jobs generated by structurally “impaired” and “non-impaired” sectors. Credit Suisse defines structurally impaired sectors to “include real estate related industries, finance, manufacturing, and the state and local government sector.” These are the sectors that at least in part rode the “bubble” economy wave. Many of these jobs were credit dependent, with growth beyond what the economy could sustain naturally.

The chart below shows the job creation and loss of the two components. The structurally impaired sector jobs created during the period of over-capacity growth simply never returned.  The sectors were highly credit dependent and with all the deleveraging taking place, the jobs are not likely to come back any time soon…

On the other hand job growth of the non-impaired sectors has almost returned to the pre-crisis levels.

Could it be that 2011 is what macroeconomic recovery looks like, minus the remaining structural problems?  Hat tip goes to FT Alphaville for the pointer, and here is my earlier post on the disaggregated aggregate demand.


This sounds a bit tautological.

Yes. Impaired and non-impaired are sufficiently vague that creative definitions will align the curves to the mood of the author.

Also, what are the units on the first plot? His impaired defination seems expansive enough to include more than half the economy.

Impaired jobs are jobs that are not recovering therefore they are not recovering
Non-impaired jobs are jobs that are moving towards recovery

Nothing tautological about that...........

Why wouldn't that be the exactly right way to do it? The structural changes are the ones that change structurally. That's the way it is because that's the way it is.

No, "structurally impaired" jobs are those most closely related to the housing bubble and bust. There may be some ex post categorizing, but the point is very clear - we heavily invested in real estate and that malinvestment reverberated through the economy. In a "normal" recovery, the FIRE sector would usually lead us out, but this recession is very different.

"Structural" is the operative word. These are jobs whose skills are not desired by the economy, and those former workers need to get trained to do something else. We have an excess inventory of homes, and the financial sector is cutting overhead costs (consisting mostly of salaries and benefits). The quote also makes it clear that these jobs were "credit dependent," and in a balance sheet recession, we shouldn't expect those jobs to come back.

This isn't tautological at all, but it does involve a good amount of hindsight.

But if the size of the housing bust was affected by unduly tight monetary policy (cf. Scott Sumner) then we are overestimating the "malinvestment" component. Marginal but perfectly reasonable housing investments (and banks, etc.) will go south if monetary policy in the crisis was too tight. You can debate this point but you can't just assume it if you're going to make a structural argument.

A monetary story would say that a lot of the so-called "structural" hit would never require restructuring if monetary policy were optimal.

The housing bust wasn't caused by tight monetary policy. The housing bubble was caused, in part, by prolonged loose monetary policy. The size of the bust was based on the size of the bubble. One could argue that the Fed should have raised interest rates much sooner than they did - that's what the Taylor Rule suggested. That would have limited both the size of the bubble and the bust, eliminating much of the malinvestment. Nobody had the courage to take the punch bowl away from the party until AFTER housing construction peaked. Developers were the first people to recognize the problem.

Banks are flush with reserves, but there is not much loan demand and there are few profitable lending opportunities. Interest rates are near record lows. Access to credit is NOT one of the problems we face right now.

One concern worth accepting is that banks are holding reserves as a precautionary motive or shoring up for higher capital requirements, and therefore there isn't as much credit available as we think. I still think there is sufficient credit available for worthwhile projects.

Willi, No one is disputing housing is structural, as well as real estate. The author was over inclusive.

No, the author wasn't. A key point of THIS business cycle is that those houses were financed with extraordinary leverage, and the mortgages were repackaged, sold, and insured. That occupied a lot of jobs in the financial sector. With government revenues pouring in, rather than cutting deficits and paying off debt, state and local governments hired more workers and expanded spending. So government rode the housing bubble too.

The author has the benefit of hindsight, but that doesn't make his choice of sectors arbitrary. These sectors are "impaired" by current conditions - excess inventory of homes, high government budget deficits, and household deleveraging. This is all quite clear in the posted quote.

Structural unemployment is when the skills of unemployed workers don't match the skills desired by employers who are seeing workers. The chart and the implications upon which it rests show this quite clearly.

Not really. Home equity drove a lot of large purchases during the boom. For example, the record years of auto sales earlier in the decade were driven in part by home equity.

Willitits, If I were to use your definitions, all prior recessions, and including the Great Depression, would meet the definition of structural imbalance, which is to say nothing.

Well, Bill, this recession was a lot like the Great Depression in some ways. First, start by reading Reinhart and Rogoff. Then read Ed Leamer. Then realize that this is not really saying nothing at all.

There is a difference between cyclical and structural unemployment Bill. You're looking at structural unemployment here. This doesn't always happen in every recession.

I didn't define anything. I was explaining why some sectors were impaired. "Impaired" already has a perfectly good definition and it's obvious which sectors are impaired and why. The Government sector didn't have to be impaired. It CHOSE to be impaired by spending every dollar of revenue it received in tax revenues during the boom. States didn't have to have high, progressive taxes. States didn't need to be borrowed to the hilt. They didn't need to have underfunded pension obligations. They didn't need to commit to higher entitlement spending, higher public sector.wages, and higher public employment. These choice negate inevitability.

Sectors that create long-lived assets are more cyclical. Governments with progressive taxes have more volatile revenues. Sectors that produce luxury goods and services are cyclical. Many of them recognized the bubble and rode it to the top. Who cares if the workers get laid off and resources are idled when you get your fat bonus or election?

Nobody who knew cared, and nobody who cared knew. Many people didn't know and didn't care.

This has all happened before. It will happen again.

This isn’t tautological at all, but it does involve a good amount of hindsight.

Well said. The job-holders themselves may not realize how dependent on the real-estate bubble they are. A restaurant surrounded by mortgage broker offices may not even realize the giant lull in customers it will face when the offices go unoccupied for a year or two.

While it may be hard to preicsely identify the structurally-impaired jobs, a few heuristics will give you reasonable numbers, and it looks like that's what they did. It will overestimate in some areas (construction firms primarily associated with sustainable businesses) and underestimate in others (the restaurant jobs mentioned above).


Finance people are still doing great, I don't know what you're talking about. As for the others, there's really nothing else they can be trained to do - unless the government undertakes large infrastructure projects there's little else that's going to absorb them.

You don't know what you're talking about. Millions of "finance people" lost their job and didn't get them back. Of course the people who kept their jobs are doing OK, but they aren't getting the bonuses they used to get. Lots of contruction workers who kept their jobs are doing fine too. That's not the issue.

Why do you think government needs to train people? How about they train themselves? Government isn't the expert in what jobs our economy needs.

No I think the government needs to accept that a huge (and growing) chunk of the population will be permanently unemployed or employed in very low productivity industries and adjustments are going to have to be made to that reality.

You might be right about that. What's your solution?

I tend to think, though, that hunger and need will provoke some people to lower their reservation wages and do jobs they didn't think they would do.

If you were a construction worker, it's about five years too late to consider a new line of work, but you might as well start now.

We could really use a good baby boom with a lot of stay at home moms doing home schooling. Or perhaps dads, considering they got hit harder on the jobs.

Well I'm not even talking about construction workers - which is the big fallacy that it's all uneducated construction workers unemployed out there. Many university graduates - even those who did the science programs that Tyler praises are in the same boat basically.
Like I said in another post - we need consumers more then we need producers but because of the moral stigma surrounding production and consumption people don't want to accept this.
Why is it necessary to force people to work to do things that we don't even need done?

With no access to the Credit Suisse report, this is one person's opinion that can't be verified, particularly if you haven't seen how they defined terms and on what basis. If you DEFINE structurally impaired sectors to include "manufacturing and government", aside from real estate and finance, then you are pretty much tracking overall unemployment. Also, if you look at the last cycle, government employment was steady, and rapidly increased during the 2001 recession.

I have some charts too that I can make if you let me define the terms and don't ask too much about the source materials.

The structurally impaired positions had lower growth rates from 04-07, a time that they were supposed to be riding the boom. Maybe today's non growth is a return to the pre-crisis, and pre-boom, levels.

These are the sectors that at least in part rode the “bubble” economy wave. Many of these jobs were credit dependent, with growth beyond what the economy could sustain naturally.

Why don't we say instead that these are the jobs that took the hardest hit in the recession, plunging to unsustainably low levels?

Because there is no such thing as unsustainable LOW employment. Employment can stay low in a sector forever.

There are very specific reasons why these sectors got hit the hardest. They are the same reasons why these sectors had the strongest growth. The growth was based on artificial price signals, drawing in too many resources. When the true price (and quantity) signals were sent, the resources were dumped onto an economy that couldn't absorb them. A hospital has very little use for more than a few carpenters. The education sector didn't need real estate agents. The federal government didn't need ski instructors. Food and drinking establishment don't need stock brokers.

"Could it be that 2011 is what macroeconomic recovery looks like, minus the remaining structural problems?"

Yes, it could be. It also couldn't be. You keep asking these open-ended questions and I'll keep answering.

You could rephrase Tyler's post as : "We could aim lower"--The poster is basically arguing that the levels of past unemployment and past economic activity were artificially high, and that therefore we shouldn't stimulate either monetarily or fiscally to close the output gap between, say, current unemployment and labor resources, or current business capacity utilization and full capacity.

If you aim low, you are bound to shoot your foot.

So, you don't think there was a boom?

People asked for the data. There is the data. The non-structural appears to never equal the structural. Now we need the chart of how much the (false) profits from the structural fed the non-structural and we'll have the actual non-structural animal spirits type portion.

Er, housing starts are well below historical trend now.

Andrew, see my earlier comment above re categories the author characterizes as structural.

"animal spirits" I really love that phrase.

Could it be that I choose not to rephrase it because the original poster is in fact purposely evasive?

I don't think he's evasive. I think he's being somewhat Socratic.
And I don't have a problem with somebody using that technique as long as their not using it in a passive-aggressive manner. Which isn't the case here.

"Manufacturing" is structurally impaired? Manufacturing was consistently losing jobs during the credit bubble.

No, no, please. Just accept that people who can't find a job are not trained for the millions and millions of high tech jobs that are open in Silicon Valley or the technician jobs in manufacturing.

There is nothing we can do.

There is nothing we can do.

Seems to me that CS is telling a story. And we know what TC thinks about stories and which ones are being told.

Came here to say what Mark said.

The analysis is bogus. The author pinpoints real estate and finance as housing-bubble-related... and then throws in manufacturing? And for that matter, Government?


I agree. "State and local government" had nothing to do with either the boom or the bust!

So the 20,000 or so messengers hired recently aren't "structurally impaired."

“The question is,” said Alice, “whether you can make words mean so many different things."

Also what Bill above said.


The second graph confuses growth rates with levels. Just because the growth rate of the non-impaired sector employment is the same now as it was pre-recession it does not follow that everything is normal. The growth actually needs to be much higher to make up for large collapse in the growth rate of the non-impaired during the crash. That hasn't happened. In other words, employment in the non-impaired sector has not made up for past losses. More importantly, it has not kept up with population growth. Thus, this figure actually speaks to there being an insufficient aggregate demand problem too.

Recall that from the onset of the housing recession in April, 2006 until about mid-2008 employment in the non-construction and non-real estate sector actually grew. That is about two years of employment growth despite the recession in housing-related sectors. This screams aggregate demand. See the first figure in this post:

"employment in the non-impaired sector has not made up for past losses. More importantly, it has not kept up with population growth."

I agree, this is exactly why the employment-population ratio is important in assessing job creation

Shouldn't the growth rate be higher in a recovery than it was at the peak?

"should" doesn't come into this. We don't get to vote on what "should" be.

But you are making a good observation. The growth RATE has returned to a high level, but it must be sustained for a long time to recover the lost jobs. And because the growth rate isn't higher than it was, it will take a very long time.

"Could it be that 2011 is what macroeconomic recovery looks like, minus the remaining structural problems?"

I don't think that's the case at all. One should look at the employment-population ratio to draw conclusions on the pace and strength of the recovery. And so far, this indicator is showing no signs of recovery in the field of job creation, i.e. we still have a jobless growth

I generally agree with you, but it depends. If the labor force participation rate is declining because more people are entering retirement and students are staying in school longer to realign their skills, it might not be all bad. Many people may have become stay-at-home parents, recognizing that their expected wages are less than their child care costs.

If people who left the labor force are staying home watching The View, we've got a serious problem.

You're right, but even if students decide to stay at school longer (or people enter retirement earlier), wouldn't this imply that their decision is influenced by the fact they are unable to find decent work?
I believe that these factors (retirement and student population) are constant across the observation (i.e. always have been present) and are thus not significant enough to explain why the situation isn't improving

Some argued that people able to retire had delayed their retirement because their savings portfolios were impaired. If all of it was in S&P Index funds, it makes sense that the stock market recovery has now allowed them to retire.

Some students are staying in school, as you say, because there are no jobs and they want to defer their student loans. I would hope that they are getting a Masters degree in something that matters to the current economy, but somehow I doubt a sociology major is going to medical school. We've created moral hazard with our student loan programs.

I don't doubt your observation, Vuk, I'm just saying that it might not be as bad as it looks. It can only be worse than it looks if someone reports they are in the labor force but they are really not seeking work. The anchoring effect of home ownership might be a substantial barrier to job seeking.

true, it might not be as bad as it seems..but the fact remains that the situation is not good, not yet..I tend to be a bit pessimistic sometimes

btw, "somehow I doubt a sociology major is going to medical school. We’ve created moral hazard with our student loan programs." This is absolutely true and disturbing at the same time

Not to mention that even many of the STEM grads that Tyler praises end up, at best, in unproductive and unrelated fields (minimum wage service jobs)

"somehow I doubt a sociology major is going to medical school"

Most PRE-Med majors aren't even going to make it into Medical School - it doesn't have much to do with student loan programs when getting into medical school is fucking ridiculously hard

@Vuk: you have good reason to be pessimistic.

@CBBB: We have had gluts of STEM majors before and will have them again. Architects aren't doing too well right now. The market might not be sending a strong signal when you start your STEM degree, but it sends a clear message to people considering their degree path when you are graduating.

The difference with the housing and financial crisis is that the market was sending artificial signals for more construction workers, more financial analysts, and more realtors even when the party was over. People in cyclical careers should know they are in cyclical careers.

Regardless of the number of STEM degrees needed, those people are smart.enough to make something useful with their degree. Soft social sciences, liberal arts, fine arts, not so much. Take a look at the unemployment rates by major. Those were posted here not too long ago. If you're a nuclear engineer, things might be tough, but I'm sure they will find work soon. I'd rather be a nuclear engineer on the job market than a sociology major.

Unemployment rates without a breakdown in occupation mean nothing - NOTHING

"Regardless of the number of STEM degrees needed, those people are smart.enough to make something useful with their degree. Soft social sciences, liberal arts, fine arts, not so much. Take a look at the unemployment rates by major. Those were posted here not too long ago. If you’re a nuclear engineer, things might be tough, but I’m sure they will find work soon. I’d rather be a nuclear engineer on the job market than a sociology major."

The WSJ data showed no such thing. There's certainly meaningful variation in salary - not in unemployment rates, though. If you want to be employed the best bets are nursing and education.

"If people who left the labor force are staying home watching The View, we’ve got a serious problem"

Or sitting around reading Marginal Revolution...

"If the labor force participation rate is declining because more people are entering retirement and students are staying in school longer to realign their skills, it might not be all bad."

I would disagree with regard to the retirement aspect. With the current employment to retiree ratio, the US can't afford to pay for their retirement. SS/Medicare are in very bad shape with the current ratio, particularly with the FICA tax cut subtracted from the revenue picture.

Structurally we'll either have to substantially cut benefits, raise taxes, or the retirement age or some combination. Since everyone born in 1960 or later is already facing retirement at 67 as normal, I can't imagine there is a lot of sentiment to support additional taxes so the baby boomers can retire at 65.

The SS Trust Fund report treats the issue with kid gloves, but still spells it out clearly.

Why is everyone being pedantic about this? The article is quite clear.

The STRUCTURALLY impaired jobs are those that were credit dependent and linked to malinvestment in residential and commercial construction, its financing, and sales. The credit and housing bubble had ripple effects throughout the rest of the economy, spreading into the non-impaired sectors. These sectors are not impaired because they don't require an expansion of credit to recover. They are also not impaired by an excess inventory of homes, slow loan growth, and large state and local budget deficits and obligations.

This is related to the idea of structural unemployment which is crystal clear in economics. Workers in construction, FIRE, and government need to acquire or apply skills that are in demand in the current economy. One would hope that the decline in the labor force participation rate is at least partially related to people realigning their human capital through training and education.

Was the connection between housing and manufacturing so strong as to taint the whole manufacturing sector as impaired? Also if you ignore "real estate related industries, finance, manufacturing, and the state and local government" is there an ex ante strong argument that the remaining sectors were any less connected to the bubble?

Yes, as others pointed out a lot of manufacturing was affected by the low unemployment and high wages of the bubble, not to mention consumer credit.

You bring up a good point about the "whole" manufacturing sector. It would be easy enough to separate out durable from nondurable manufacturing. I'm certain that cars, electronics, and airplanes were affected by the bubble and burst. Durable good employment is rising, but nondurable employment isn't. It's clear the bubble and bust had an effect on manufacturing, but both durable and nondurable have been falling for more than a decade. The author probably has to take trend into account.

But even within sectors there is diversity of jobs. If you are a carpenter at Disneyland, you are in the Leisure sector, not construction. If you are an accountant for Ford, you are in the manufacturing sector, not financial services. All this data is going to have some distortion.

Residential investment and durable manufactures usually pull us out of recession. That's not the case here, at least with RI. And durables are on shaky ground. Manufacturing production is up, but employment is (and has been for some time) dropping.

Can we agree that there is no such thing as a bubble and that assets are priced based on expectations of the discount rate?

Also, when discussnig structurally impaired jobs, isn't this a function of technology more than anything else. Improvements in technology and overall efficiency lead to less work - with the ends state being very few people actually have to work to provide the same goods/services we have now. We just have to start finding better ways to spend our free time.

It's obvious that bubbles do exist. Hyperbolic discounting, incentives, embedded options, externalities, and subsidies distorted price information in the market.

You are correct about technology. After this recession, many firms (particularly manufacturing) found ways to replace workers with capital. Low interest rates and low returns made it optimal to invest. Either that or they were operating with extraordinarily high and inefficient amounts of labor during the boom. There is evidence of that in financial services. Efficiency wasn't as important when the market was heating up. Firms became revenue maximizers instead of profit maximizers.

I don't think that spending free time is the problem here.

I have all the free time in the world - but what the hell can I do with little to no money - not to mention student loans to pay off from my useless STEM degree. Society can easily afford to better distribute income - the reality society needs consumers more then it needs produces but for moral reasons consumption is shunned.

Credit Suisse -- doing the economic science that guild incentives give tenured professors & grad students no competence or incentive to tackle.

+1 on the Econ professor bashing

You have to look beyond real estate to see what effect the bubble had. First, most local governments are funded in large part by property taxes. The higher the value of property, the more they can either add staff or cut taxes. Both become unsustainable when values drop by 40 percent. As for manufacturing, I can tell you at least one large area of manufacturing - automotive - was driven by the real estate bubble as buyers took out home equity loans to buy cars.

Ted, Government employment was fairly constant during the period and proportional to population of the local area.

Fairly consistent. And compensation? And revenue?

Also, you can put off a lot of structural changes when the money is flowing in.

Ted, If you are talking about tax cuts that are unfinanced by future growth, then I don't know how you can hang that on the government worker, who performed a service, and not on the politician who gave the tax cut. In any event, that definition of a bubble--negative revenues caused by tax cuts--is clearly a unique way of describing a bubble where there isn't a change in output.

I'm not hanging anything on anybody and saying I am, implying an assault on workers, is a weak trick. What I'm saying is the property bubble created other bubbles. Municipalities often either cut taxes or slowed price increases. At the same time, they took on debt, added positions and ignored structural problems. Their output per worker probably stayed fairly low, as well. A lot of companies did this, but governments are challenged in their response.

Bill and Ted's Excellent Adventure into government finances.

Recoveries are not what they used to be.

Neither are depressions......

But politicians, lawyers, and economists never change.

Excellent post, creating much cognitive dissonance.

Has the relationship between the loss of structurally & non-structurally impaired job losses been examined? The data appears to highlight a trend of increasing job losses which followed the initial loss of jobs in the non-impaired market. This relationship could help forecast future effects of rapid de-leveraging of impaired markets, on the remainder of the job market.

Comments for this post are closed