We are no longer in the short run

Here is more from Eli Dourado, excerpt:

NGDP is almost 10 percent higher now than it was at the pre-crash peak. The number of people employed, even with population growth, is still below the pre-crash peak. Even assuming that insider nominal wages are totally inflexible, nominal output per worker has grown fast enough that insider real wages have probably adjusted. Furthermore, in five years, a non-trivial fraction of insiders retire or change jobs.

There is much more at the link.


No, no. The short run is excessive pessimism, while the long run is excessive optimism. Judging by the markets, we're not yet back to that long run of American economics, excessive optimism.

So, if we're in the long run, and corporations have the money to hire people, they obviously don't see a profit in doing so. If we rule out insufficient AD as the cause (because NGDP is 10% higher now), that leaves ... ZMP (or negative MP) workers? Anything else?

Actually NGDP is up 8.2% while real GDP is up 1.9% from the peak. But the cost to hire new workers is uncertain, reflecting the possibility of Obamas' reelection and the costs of Obamacare and increased taxes.

You're using Q2 data. Q3 NGDP should be in the 9+% range above peak.

But insider wage expectations should be ~20% higher than they were at the pre-crash peak if they were *completely* inflexible. w/NGDP is still higher than at the peak, so the Sumner contention cannot be easily dismissed.

Also, why does nobody ever mention Truman Bewley in these debates? We construct models and graphs, but when someone actually attempted empirical research, we're not even cognizant.

"But the cost to hire new workers is uncertain, reflecting the possibility of Obamas’ reelection and the costs of Obamacare and increased taxes."

How did you work that one out? Oh. You didn't work it out. You just made some stuff up that fits with your prejudices.

If we are playing that game:

It reflects the possibility of Romney's re-election and the effect of a political party that will not be able to control the deficit because it won't raise taxes.

Uncertainty is demonstrably and measurably increased.

And TC expends considerable energy explaining that Romney will raise taxes. However, that only rescues the government from it's self-f-----ed balance sheet, it's not a good thing overall.

"Uncertainty is demonstrably and measurably increased. "

So is the number of people on the planet. Your Point?

"TC expends considerable energy explaining that Romney will raise taxes"

Sorry. I wasn't being clear. I thought Rich Berger was asserting with a great deal of certainty something for which there is no solid empirical support. I meant to mimic his approach. Hence, the "If we are playing that game.".

I doubt Romney would do a terrible job. Claims that Obama's management of the economy is incompetent (relative to other governments or presidents not relative to some Platonic ideal) are plain daft.

Obamacare lowers total health costs, but dont let reality get in the way of right wing stupidity

Total health spending will go up in real terms every year after obamacare takes effects. Per capita health spending will also go up. Please provide evidence or citations if you can counter this claim. No credible health economists is going to project lower health care costs overall.


Its the difference between the two programs that matters. I assume by lowering total health costs fredjames means that the costs will grow at a slower rate than would have occurred under the previous system.

Re: that leaves … ZMP (or negative MP) workers?

How does a business know someone is ZMP without actually employing them? Assuming that anyone currently without a job must perforce be ZMP (except in the purely trivial sense that anyone not working is not producing anything) is a wildly inaccurate way to assess people's potential.

A more accurate term would be ZMP positions.

Zero Marginal Product Workers? There is no such problem. Are you telling me that if for some reason people starting going to McDonald's more McDonald's couldn't open more stores or pump out more fries and burgers? There's simply no one on the unemployment lines who have even the bare min skills to hire from?

If by Zero Marginal Product you mean a case where if a business where to hire another worker, it's profits would not increase or might possibly decrease then that's fine. But that says nothing about the workers themselves as business's abiliity to sell more output...in other words a demand problem.

Always reason to a demand problem.

If you refuse to define your terms, you leave me with no choice but try to define them for you. A zero marginal product worker could mean:

Output based: Worker adds nothing to actual output. "We put him on the hamburger line and yet our hamburgers produced per hour didn't increase". This is likely a supply problem. The worker is bad or possibly the capital is not able to accomodate the added worker to makemore output.

Profit based: "We added another hamburger cooker but profit didn't go up because people aren't buying the extra bergers so we had to cut prices to keep them from spoiling!" The worker isn't the problem here, the business can't sell the output he is making as a result he adds nothing to the bottom line. Demand problem.

Now demand problems are common because almost any business would, if faced with more customers, hire more workers and expand output. Sometimes this is impossible, though. There's only one Cher, for example. If she does one concert per year and people want to see her live, you can't just 'hire more Chers'. You have a supply limit so you charge more per ticket.

But few people demand that Cher make their hamburgers at McDonalds or stock their shelves at Wal-Mart or build them a car. They are happy to let anyone do this for them and business is set up so that whenever possible jobs are interchangeable enough so that you rarely get in a situation where one single employee can hold the company hostage.

In the second quarter of 2013 real final sales of domestic product was $13,508.2 B (2005 $).

At the fourth quarter 2007 peak it was $13,506.2 B (2005 $).. That means that real AD is only 1.4% above the pre-recession peak.

This measure is the correct measure of output he should use, not nominal GDP.

He is comparing nominal output to real employment -- any first year student in intro economics should be taught not to do this.

Never let things like mathematics get in the way of dogma.

What about retail sales? Why is that not the proper measure?

Because you want to include exports and sales to other businesses and government in AD.

Moreover, a significant share of retail sales is foreign products.

But I thought we were measuring demand, not supply? So why does it matter if it is domestic or imported?

Over this same period hours worked in private business fell from 102.3 ( 2005 = 100) to
96.6, or -5.6%.

Isn't the comparison that was made nominal GDP to nominal employment?

People, we are testing the continuing applicability of *nominal* models here in explaining real employment.

A clarifying question:

Is this a) a refutation of AD explanations for high unemployment OR b) a claim about the limitation of monetary policy as a solution to current high levels of unemployment?

I don't think these are the same. Am I wrong? It seems to me a refutation of monetary policy is a refutation of sticky prices as the current source of high unemployment. But sticky prices are not the only reason why AD may be low?

The nominal level of the stock market is still lower than its pre-crash peak. Maybe that says something about nominal earnings expectations, with consequences for investments in labor.

from the end of Eli's post: "There is enough ambiguity in our current situation that reasonable people can disagree about what is going on. But I don’t think that reasonable people can be totally certain that all we need is more nominal stimulus."

Actually reasonable people agree with Eli that monetary policy is no cure all. Just the latest example from today's speech by SF Fed President John Williams: "Monetary policy cannot solve all problems that affect our economy. But it can help speed the pace of recovery and get our country back on track sooner." http://www.frbsf.org/news/speeches/2012/john-williams-0924.html

I would only add that there's an important distinction between arguing something can't solve a problem and arguing that it can't be part of the solution. Measured success is still success (depending on your alternatives, of course). Also to embrace micro foundations I would like to see Eli's empirical arguments for the structural increase in unemployment / decline in productive capacity, like job mismatch or technological regress. All that said, I very much agreed with Eli's last paragraph (even did a double take to make sure I was still reading a blog).

Since the start of the crisis, labor participation has been skewed downward by millions of workers simply due to the aging of baby boomers. Any analysis that references static employment levels is bound to miss the point.

You won't solve this by using an aggregated labor model. It isn't that there's no recovery in the job market. It's that a certain pay range of jobs isn't recovering.

Jobs paying $14 and under (the bottom third) have recovered more than 100% of their loss. Jobs in the middle third have recovered around 20%.

Robert Cringely posted this graph that tells the story


He has 3 recent articles on his site relating to this issue.

The graph itself is from an article


and is definitely worth reading. This chart of the occupations with the biggest unrecovered losses and their average salaries is from the article:

Driver, sales workers and truck drivers -495,539 15.34
Secretaries and administrative assistants -345,101 15.80
First-line supervisors/managers of office & administrative workers -327,559 18.50
Carpenters -211,954 16.56
Real estate brokers and sales agents -181,078 21.06
Maintenance and repair workers -145,139 17.96
Computer, automated teller, and office machine repairers -118,822 19.86
Data entry keyers -115,455 14.13
Billing and posting clerks and machine operators -114,610 15.23
Insurance claims and policy processing clerks -108,842 15.26
First-line supervisors, managers of production & operating workers -107,059 19.90
Painting workers -99,319 14.24
Electricians -97,714 20.94
Pipelayers, plumbers, pipefitters, and steamfitters -90,995 18.53
Carpet, floor, and tile installers and finishers -88,774 14.11

Some of these jobs will recover to the extent construction does. Some are probably gone for good.

"Some of these jobs will recover to the extent construction does. Some are probably gone for good."

Umm, why are they gone? That's a pretty wide range of occupations to my eye. What is the supply shock in the recession that rendered them all obsolete? Why wouldn't there be these jobs if there were more overall demand? I am not saying I know the answer here, but drilling down to the occupation level is not enough.

Also the hollowing out of the work force is not a new idea. Here's a summary of some of debate: http://campaignstops.blogs.nytimes.com/2012/07/08/the-future-of-joblessness/

Because computers and increases productivity is replacing secreteraries and clerks

And we discovered computers in the Grear Recession? I agree that recessions can be a time of structural realignment (ongoing trends are important too), but recessions are also usually a time of adverse demand shocks (including heightened uncertainty). I am asking for the empirical evidence that weighs out these explanations.

I'll give you an anecdotal example from the field I work with. Car dealers have been able to stock their lots with used vehicles for years, but most hesitated to do so. Then the recession created a lack of supply and decreased staff. This forced many of them to adapt. Now, many buy 100 percent of their inventory online. Often in business, people only change when they have no choice.

"Because computers and increases productivity is replacing secreteraries and clerks"

Computers, email and automated phone directories replaced most secretaries and clerks 10 years ago. And in any case those positions would be the low wage positions that the graph indicates have actually grown.

You'd think so, but the detailed BLS figures indicate otherwise. For instance, there clearly were 345,101 Secretaries and administrative assistant jobs lost at $15.80 from 2007 to 2011 and 327,559 First-line supervisors/managers of office & administrative workers at $18.50 who were likewise expendable.

It takes a long time for these trends to play out. Secretarial jobs have been in decline since the 1970s. There just happened to be a huge number of them and a lot of inertia in the process..

The two trends are - replacing human labor with machinery and changing work flow and product design to eliminate the need for certain kinds of jobs. You don't need repair people for things that rarely break and are discarded when they do. For instance, dehumidifiers aren't repaired anymore. If one dies under warranty, they send you a replacement.

Nice link!

The construction related jobs may come back, but once a company discovers it can survive with fewer executive secretaries, and first line supervisors, it won't hire them back.

I remember years ago when visiting a company meant parting a sea of receptionists, secretaries, typists and assistants. The trend to eliminate clerical and supervisory jobs goes back to before Hollerith. It won't reverse.

Economists like to cite two big mechanizations that didn't have long term negative results for labor - the agricultural and industrial revolutions. This isn't all that persuasive.

There were 3 great revolutions. The first was from hunter gathering to agriculture. It may have been positive in the long term, but it was certainly negative for most of the population involved. We can estimate levels of nutrition from skeletal remains. The results are clear. Agricultural society increased population, but decreased individual welfare. That we wouldn't be here otherwise doesn't change this.

You can make a similar case for the industrial revolution. The period from 1815-1870 was not a paradise for labor. This was the time of the Poor law and the workhouse. Again, the positive long term result doesn't change the 50+ years of misery.

So you can argue for a long painful period of adjustment that we won't live to see.

Or you can observe that it appears that automation is removing jobs faster than we can create the unmet needs to motivate new ones. The force creating new jobs is the need to recycle profits through labor so that labor can act as consumers for consumer goods. It is possible to envision changes to the economy that would to some extent, at least, obviate that need. I don't think this is desirable, but such considerations rarely stop social changes.

That gives the traditional 3 possibilities - the good, the bad and the ugly. Pick one.

One thing I've noticed is the fact that there are plenty of myths regarding the financial institutions intentions while talking about foreclosed. One fairy tale in particular is the bank wants your house. The lender wants your hard earned cash, not your home. They want the bucks they gave you having interest. Keeping away from the bank will only draw the foreclosed conclusion. Thanks for your article.

I am totally agree with your thoughts. Keep doing these type of work.


Comments for this post are closed