Arnold Kling on Patterns of Sustainable Specialization and Trade
Feb 7 2011

Arnold Kling of EconLog talks with EconTalk host Russ Roberts about a new paradigm for thinking about macroeconomics and the labor market. Kling calls it PSST--patterns of sustainable specialization and trade. Kling rejects the Keynesian approach that emphasizes shortfalls in aggregate demand arguing that the aggregate demand approach masks the underlying complexity of the recalculations that periodically take place in a dynamic economy. Instead, Kling invokes the mutual exploration between entrepreneurs and workers for profitable opportunities that pay well using the workers' skills. This exploration takes time, involves trial and error, and can have false starts because businesses sometimes fail or employees are difficult to find or match with employment opportunities. Kling applies these ideas to the current crisis to explain why labor market recovery is so sluggish and what might policies might improve matters.

Arnold Kling on Morality, Culture, and Tribalism
Economist and author Arnold Kling talks about the economic impact of culture and morality with EconTalk host Russ Roberts. Drawing on a recent essay on the importance of social interactions, Kling explores the role of culture and norms and their...
Arnold Kling on Specialization and Trade
Arnold Kling, economist and author, speaks with EconTalk host Russ Roberts about his latest book, Specialization and Trade: A Reintroduction to Economics. Kling argues that macroeconomics ignores the challenges of buyers and sellers working together in the real world of...
Explore audio transcript, further reading that will help you delve deeper into this week’s episode, and vigorous conversations in the form of our comments section below.


impartial spectator
Feb 7 2011 at 9:04am

“Kling rejects the Keynesian approach that emphasizes shortfalls in aggregate demand arguing that the aggregate demand approach masks the underlying complexity of the recalculations that periodically take place in a dynamic economy.”

I do not see how the two paradigms on the causes of downturns in an economy have to be mutually exclusive. Can’t they both be true at the same time?

Is it not possible that after a large bubble has burst a) the economy needs time to recalculate and will not immediately jump back to it’s old growth path, AND b) the increased uncertainty due to the needed recalculation increases the demand for money which, if not met by the central bank through an increased money supply causes the economy to plunge even deeper into recession (because of sticky prices the increased money demand will only be satisfied after a prolonged period of underutilized resources and price depreciation)?

Feb 7 2011 at 9:37am

Seven minutes in, the podcast is not going well. Arnold implied that there was no recession at the end of WWII; the NBER says the US was in recession for most of 1945 (perhaps Arnold meant to say it was less severe than expected).
Then, both Arnold and Russ complain that AS/AD does not use prices. But of course it makes extensive use of the overall price level (which shows up on the simplest AS-AD graph) and the idea of sticky prices. Again, perhaps they meant that it does not give prices for individual goods, but this is not what comes across.
Impartial spectator, both paradigms can be true (or at least useful), which is why Arnold began by telling us to think of paradigms as languages.

impartial spectator
Feb 7 2011 at 11:41am

James, despite Arnold Kling beginning his story as one of wording and paradigms, I understood Mr. Kling to altogether reject the keynesian paradigm as a useful policy guide.

Early in the podcast he and Russ Roberts seem to reject the notion that prices can be sticky downward, so to me it seems that they dismiss any resulting problems resulting from an increased demand for money. If prices are perfectly flexible there just is no keynesian aggregate demand problem.

Later in the podcast (in the 58:59 section of the transcript) Arnold’s use of Cosby’s car analogy also seems to dismiss the notion that any problems can arise from a shortfall in aggregate demand. Klings says that “pouring gas all over the car” does not help entrepreneurs find new patterns of specialization and trade. But if there is an increased demand for money due to higher economic uncertainty that is not what the provision of more money is supposed to accomplish in the first place. Instead its primary purpose would be to avoid a “secondary depression”, so that entrepreneurs can follow their entrepreneurial function without putting the economy under the additional stress of a prolonged deflation.

Feb 7 2011 at 11:45am

This podcast marks a new low IMO.
May I humbly suggest that the host acquaint himself with Donald Davidson’s “Principle of Charity” [1], and that henceforth host and guests alike do their utmost to put it into practice? EconTalk is a real treat when Russ invites someone to speak on a subject with which they exhibit some minimum level of affinity, or who will at least make a concerted effort to represent the subject matter fairly, “as if” they genuinely believed that its authors and/or proponents are in fact sane, generally honest cohabitants of the same universe. Whenever the program involves guests of the other kind (or worst of all, guests of the other kind who share Russ’ openly acknowledged visceral biases), the result is both uninformative and truly unpleasant — as well as profoundly unflattering to the participants themselves.

More-honest-debate… please?


Lee Kelly
Feb 7 2011 at 1:44pm

I can hardly believe it, but I find myself agreeing with Fazzari more than Kling. For me, Kling just doesn’t understand the unique role and peculiar properties of money. He can keep his “paradigm”.

Feb 7 2011 at 3:04pm

Agreed, AS/AD model AND the PSST model simultaneously affect the economy. Which of the two paradigms do you prefer? I lean towards Kling’s explanation. It sits well with me.

I would like to hear suggestions from the hosts regarding how an individual goes about discovering their competitive advantage(s). And/Or real life examples and stories which reaffirm this model in their own marginally-located-lofty-placed-brains.

Feb 7 2011 at 4:31pm

It is not apparent that a webcast or blog is the appropriate forum to establish a new “paradigm” to replace one established 70 years ago by one of the giants of economics (Hicks). This might better be done in professional journals.

Kling’s critique of Hicks and Keynes is perceptive. His comparative advantage approach is interesting, but it should be spelled out in a technical way. Then it could be explained to a general audience.

Feb 7 2011 at 6:07pm

@impartial(?) spectator: “… b) the increased uncertainty due to the needed recalculation increases the demand for money…” What on earth can you mean? Why would anyone pay more for money when they are increasingly uncertain what investment outcomes will be? I haven’t noticed much in the way of results from the Fed pushing on that particular string, btw.

Feb 7 2011 at 11:30pm

What long term comparative advantages does the US have against the rest of the world? The podcast posits that entrepeneurs are not starting new business here because of the cost of labor exceeding the possible returns. However, isn’t it also possible that entrepeneurs are just going to other countries and using the even cheaper labor there? Even if you could make a return using American workers, why not make a greater return using international labor, especially when we are talking about low skilled jobs?

The discussion just seems really America centric, but can you really talk about the isolated economy of single nations anymore, especially one as tied into the global marketplace as the US?

Feb 8 2011 at 8:56am

Rather disappointed in one turn of today’s argument. Both Kling and Roberts implied the booming economy and low unemployment of postwar America owed to some fairly mysterious rush of entrepreneurial insights into comparative advantage. The attribution of mystery is a good sign that, at least here, the paradigm is failing.

After a brief dip immediately following the war, the US economy grew at tremendous rates because of its enormously favorable balance of trade with the rest of the world, a world of hobbled manufacture that had been largely reduced to rubble but which still needed goods. This is a very satisfying aggregate demand story, but one could rephrase it PSST – The US had the comparative advantage of not being rubble in 1946.

Feb 8 2011 at 12:22pm

(A different Aaron)

“My view is a lot of it was just ongoing sectoral shifts that got really accelerated at least in terms of letting workers go during this recession for reasons I just would have to tell stories about–that firms almost like a herd behavior decided now is the time to focus on controlling costs and not a time to expand.”

Was I the only person who immediately thought that “herd behavior” = “animal spirits?”

Feb 8 2011 at 1:29pm

Anyone have a link to Dr. Kling’s paper on price stickiness?

Bob Layson
Feb 9 2011 at 7:13am

Understanding will be helped if the stages of a depression are explained in turn:-

1. There is a severe loss of profitability in some sector of the economy due to an unanticipated supply or demand shock. The cause may be natural disaster or, say, wars abroad leading to a loss of inputs for production or the practical impossibility of trade with the usual partners. A credit expansion induced boom and bust is another route to a rash of losses and redundency as, in the event, malinvestment reveals itself. These redundent assets now have to be found a business in which they can be successfully integrated. Too much was made for present falling sales volume and prices to cover costs even if workers took a pay cut.

2. Redundent assets are earning no income and so sales to firms not yet in trouble must now suffer in turn and lead them to layoff workers or go to short time working. This in turn leads to falling sales and lower income in other firms. This is not a matter of money but the absence of real produce to profitably exchange with other producers. (See W H Hutt on Say’s Law.)

3. As government insists that there is ‘no cause for alarm’and that ‘the fundamentals are sound’ – even as they brag of taking actions unprecedented in postwar peacetime – then people naturally become anxious about their future earnings and rein back on discretionary spending in order to increase their savings. Ironically this leads to even more job losses as firms not involved in the malinvestment boom lose custom.

4. The previous pattern of employment cannot be reinstated by attempts to return to the previous level of monetary demand just as fast as government deficit spending and fiat money creation can manage it.

5. Recovery requires not more money hosed around the economy but productive adjustment and movement of people and prices. After all not all firms were doing the wrong thing but simply too much of it in the face of falling demand. They can recover as the unemployed price themselves employable and return to making their contribution, for a other non-rival firms and perhaps in a different way than previously, to the various exchangable productions that constitute demand. Not all business investment suffers in a depression as some sectors are growing or firms within a sector invest to make profit, or evade loss, by cutting costs. These falling costs allow the consumer to spend on other things and increase employent in other non-rival businesses.

6. Growth returns as being taken on here creates the demand for more workers to be hired here and there.

It is obvious that present day regulation and tax and non-wage legally required labour expenses, when coupled with payments to the unemployed to hold out for a return to normal work and wages, gums up the process of adjustment and leads politicians to conclude that only further intervention will solve matters.

Feb 9 2011 at 8:34am

I’m both bummed and glad at the same time that the critiques I had have already been brought up here. Also, I like/hate that some of them have already been answered in the podcast mere minutes after I formed them in my mind.

Adding my voice to James’ that the WW2 example is an especially bad choice, all that I am left with is to critique Kling’s definition of a paradigm.

It is true that in the humanities like literature, comparative cultural sciences and the like one can pick a paradigm to ones liking (Which, as a humanities major myself, is my greatest beef with it). In the harder sciences paradigms get shifted because the new paradigm explains things better than the old one. Wiki gives a great example: in medicine one could not concievably go back to the miasma paradigm after shifting to germ theory.
Even the example given by Kling is a little off. Even though in both french and english the same sentence could be expressed accurately, what that sentence means and evokes in terms of associations is culturally determined and can therefore be wrong or right. Deconstructivists would even argue that due to the influence of the powerful classes in a society, there are some things one cannot say because the concepts that are needed to say it have been suppressed or warped in a way that make it impossible to formulate the thought. I hate that argument though because Popper would hate it.

To end on a more cheerful note: I still like the idea of PSST.

CA Bob
Feb 9 2011 at 10:40am

RE price stickiness: In the capital goods industries prices are sticky for a several reasons. First, these tend to be highly cyclical and price cuts will do nothing to spur demand if cap ex budgets have been slashed. Second, some contracts stipulate that the buyer has rights to the lowest prices offered by the supplier, as such price cuts can have broad retroactive consequences. Third, it can be difficult to get those prices back to “list,” even though there is often much discounting in the market normally. Selling below cost is a losing proposition in any case. Typically price concessions take the form of e.g. discounts on spare parts and service, discounted upgrades, etc. Too often I hear economists talk of prices as if all goods are commodities, whereas virtually all producers work hard to avoid being commoditized and thus avoid pure price competition — consider brand development. The discussion around employee wages has mirror elements within employers. Having people with the same duties and pay grade but very different pay is a prescription for dissatisfaction and resentment. These are some of the less discussed but insidious aspects of inflation.

Mort Dubois
Feb 9 2011 at 8:55pm

Regarding a couple of statements in the podcast:

“ The neo-Keynesian, new Keynesian claim is that it’s expensive to change prices downward but not upward. I don’t understand that.”

Let’s say the costs of changing your price – reprinting the catalog, training the sales people, whatever – are the same in either direction. When you raise prices, you end up with more money to cover that cost. When you lower them, you end up with less money to cover that cost, along with less money to fund all other operations. This can put you out of business, so you don’t want to do that if it can possibly be avoided. This shouldn’t be hard to understand.

“The idea–which I’ve never understood–that producers are loathe to lower their prices, or that it takes a while, or that it’s costly to adjust prices. All of which are true, but it’s hard to understand how that works in a time when people have all this excess production, output, sitting on their shelf.”

Your concept of a factory is way obsolete. With just in time production, there is very little inventory, so you don’t have a lot of stuff sitting on your shelf. Everything that is in the factory is already sold, although it may be a long time before you get paid for it. If you don’t have an order in hand, you don’t produce anything. You are also presuming that there’s some group of potential customers out there who would buy from you if only you would cut your prices. In the real world most companies make such specialized products that all of the potential buyers are already being served by someone – your competitors would just as quickly cut their price as you would. You are also ignoring the possibility that the input costs of the next round of production (materials, labor) may not have dropped in cost, so selling off the current stuff at reduced price may not leave you enough money to operate. You are also ignoring the fixed costs, like debt and rent payments, that cannot be lowered in the short term.

Also: many sales relationships are based on mutual trust. There’s a real cost to establishing mutually beneficial relationships with a supplier. It takes time and money to fine-tune the relationship. In a just-in-time world an unreliable supplier can put you out of business. So “I’m about to go out of business so I have a bunch of cheap stuff” is not a compelling sales pitch. How do you plan future production around a disappearing vendor? If the failing producer makes substandard components, and you incorporate those in your own products, you are looking at potential costs down the road, in warranty and liability claims.

You should think about interviewing a real factory owner, asking how they deal with these issues. Maybe this guy:

But as always, I enjoyed the podcast, and thank you for it.


Arnold Kling
Feb 10 2011 at 11:34am

I do not believe that there was a recession at the end of World War II. Nominal GDP did fine. At the time, real GDP was measured as doing fine. Since that time, the price adjustment factors have changed to make it appear that there was a decline in real GDP. This is because today we use price indexes optimized to capture inflation in the 21st century, not in the middle of the 20th.

In any case, if you look at the enormous drop in government spending relative to GDP in 1945-1947, it seems incredible from a Keynesian perspective that the economy did not go into a deep, deep dive.

Feb 10 2011 at 11:48am

Mort Dubois,

I often use these comment pages to try to thank Russ Roberts for doing such an educational podcast and letting us listen to it for free (and thx again Russ & Rich), but I also wanted to thank you for the comment above. Your comment helped educate me and I plan to keep it in mind when I listen to this podcast a second time in an attempt to understand the PSST argument better.

To all EconTalk listeners, please keep making these kind of informative comments. Especially the challenging kind, like the one Mort has produced above. And perhaps it goes w/o saying, less like this one that’s a simple “thanks” and adds nothing to the conversation.

Feb 10 2011 at 11:59am

Forgot to say “thanks” to CA Bob as well. His comment was also quite helpful and informative.

Feb 10 2011 at 12:27pm

The podcast has elicited strong reactions, a good sign I suppose.

When I took econ 101 years ago, almost every page of micro sparked “aha” reactions from me. It was logical and explained a lot that had been obscure. Macro was nothing like that. I managed to make IS/LM “work,” for example, when I needed it to on a test, but it seemed like logical links in the chain were missing. And so on throughout macro. In later years I learned that many things Keynes himself said are patently absurd (for example, that by keeping interest rates low an economic boom can be indefinitely sustained). So all I can say is, Go for it Dr. Kling.

John Berg
Feb 10 2011 at 4:01pm

I was struck by the last moments of the podcast and the statements about “certification” as applied to labor. A recent episode of “Little House on the Prarie” noted that in the old West the teacher of the local school was picked from among the local women. I suspect the onerous system of “certification” for specialized labor today may be akin to property regulations like zoning, inspections, and licenses that have the effect of increasing rigidity and raising the price of property.

John Berg

Russ Roberts
Feb 11 2011 at 12:46am


According to the NBER, there was a recession “after” World War II. It began in February 1945 and ended in October 1945. It is difficult to attribute this to cuts in government spending—the war in Europe ended in May and in Japan in August. The massive demobilizations and cuts in government spending presumably took place well after the recession began. And as Arnold points out above, measuring any recession (using real GDP as your measure) is greatly complicated by the difficulty in measuring prices accurately.

What is more interesting is the overwhelming set of predictions made by Keynesians during the war who predicted a Depression and mass unemployment when the end of the war arrived. These predictions were quite inaccurate as even their proponents admitted at the time.

David Henderson uncovered this prediction from Paul Samuelson:

The final conclusion to be drawn from our experience at the end of the last war is inescapable–were the war to end suddenly within the next 6 months, were we again planning to wind up our war effort in the greatest haste, to demobilize our armed forces, to liquidate price controls, to shift from astronomical deficits to even the large deficits of the thirties–then there would be ushered in the greatest period of unemployment and industrial dislocation which any economy has ever faced. [italics in original]

Everybody makes mistakes. But Samuelson had lots of company. I hope to write a brief overview of other predictions and analysis from the time and will post it here. The basic point is that the end of the war was a natural experiment that did not have the dire consequences predicted by Keynesians.

Feb 11 2011 at 4:09pm

Russ that’s a fantastic quote. How many spectacularly wrong Keynesian predictions does it take to spark second thoughts among true believers?

Mike Laursen
Feb 11 2011 at 7:23pm

Very interesting topic. I’m on my third listening.

Feb 12 2011 at 1:47am


If there is no increased demand for money how do you explain the low inflation rate despite massive increases in money supply?

I think you might be a little confused as to the meaning of the sentence you are critiquing.

David Zetland
Feb 12 2011 at 8:16am

Bravo! I am a fan of the Austrian business cycle and Kling’s ideas are compatible (humans getting outa wack with each other). Money surely matters, but it’s not what drives an economy. As Chevron says: People do. This theory scales from Robinson Crusoe on up. It also highlights the fatal weakness of throwing spending (or money supply) at the problem.

More please! (Suggest comparisons of predictions/explanations from different schools for major macro events in past 100 years AND policy recommendations to get out of problems.)

Mike Laursen
Feb 13 2011 at 9:48am

re: “If there is no increased demand for money how do you explain the low inflation rate despite massive increases in money supply?”

I am not an Economist, but isn’t it commonly acknowledged that:

* Banks that have gained from quantitative easing are keeping a lot of their gains on their balance sheets. So, much of the increase in money supply hasn’t spread to the wider economy yet.

* We ARE seeing early signs of inflation in the commodities market.

Mike Laursen
Feb 13 2011 at 9:57am

My background is in Computer Science, not Economics. We are used to analyzing complex networks, much like the networks of PSSTs that make up an economy.

As a computer scientist, it’s very clear to me that quantities such as “aggregate demand” or “aggregate supply” are properly thought of as measures, and imperfect measures at best. That someone would think you could turn an aggregate measurement into a controlling factor for a complex system by operating it in reverse is bizarre thinking.

Feb 13 2011 at 10:42am

Mike Laursen wrote

As a computer scientist, it’s very clear to me that quantities such as “aggregate demand” or “aggregate supply” are properly thought of as measures, and imperfect measures at best.

The economic concepts of aggregate demand and aggregate supply are entirely theoretical. Maybe I misunderstand your use of the word “measures.” They are not measures as I understand the word. I wonder if you mean “proxies” rather than “measures”?

An analogy to either of these aggregates in, say, physics is the amount of matter in the universe. That is clearly a theoretical quantity. It is not a measure, though we could hope to measure it. We can predict that the total amount of matter is the sum of all the individual bits of matter in the universe. Or, we can predict the total amount of matter in various other ways–which may not all agree. Ideally, both the micro (sum of the individual pieces) and macro (total via other prediction methods) should match up. But both micro and macro concepts of the total amount of matter in the universe are theoretical constructs. Measuring them separately or seeing if they match up are entirely different matters (not to pun).

I completely agree with you, though, on this:

That someone would think you could turn an aggregate measurement into a controlling factor for a complex system by operating it in reverse is bizarre thinking.

Feb 13 2011 at 4:33pm

Mike Laursen,

“Banks that have gained from quantitative easing are keeping a lot of their gains on their balance sheets. So, much of the increase in money supply hasn’t spread to the wider economy yet.”

Exactly, the demand for money and highly liquid reserves has increased… that was my point.

“We ARE seeing early signs of inflation in the commodities market.”

It can also be argued that some of this is related to an increase in global demand for many commodities. As large nations such as China and India continue to grow there demand for commodities will only continue to increase driving up the cost of many internationally tradable materials.

Feb 13 2011 at 4:37pm

Mike Laursen,

Regarding the first point, one more thing.

The fact that banks are simply holding onto money and building up reserves seems like a good argument for a greater role for fiscal policy when facing a 0 lower bound situation as the one we are in.

If you want to get money circulating in the economy it makes more sense to give it to people who will actually spend it than those who won’t. While you’re at it you might as well get people to perform some service for this money and create some value for society, unlike when it is basically given to banks.

Feb 13 2011 at 10:03pm

I liked this podcast. A discussion of some points of economics that never made sense to me. I am interested in how to clean out bad investment decisions and yet cope with the pain.

Do more of this.

Feb 14 2011 at 2:21pm

The episode was just so-so for me this time. The explanations were too simple or broad. For example, I would have really liked some more insight than, “the people in the 1950s got lucky on their guesses on what would work and the people in the 2000s didn’t”.

Note: I am not an economist, just an avid listener.

Thanks for the continued good work on the podcasts, Russ!

Feb 14 2011 at 8:37pm

just a thought: is it plausible that with women having filled a large number of jobs at home that the returning soldiers, presumably still being paid for a time though not yet working, amounted to a de facto stimulus, and that the subsequent dropping of a significant number of women from those wartime jobs prevented a short term surplus of labor, easing the passage from shortfall of demand to growth and the reintroduction of women into an expanding economy?

Mike Laursen
Feb 14 2011 at 11:40pm

Lauren, “proxies” sounds like a good word to describe the aggregate thingees.

Mike Laursen
Feb 14 2011 at 11:51pm

re: “If you want to get money circulating in the economy it makes more sense to give it to people who will actually spend it than those who won’t. While you’re at it you might as well get people to perform some service for this money and create some value for society, unlike when it is basically given to banks.”

Except that handing out massive amounts of money has got to be one of the best ways to build up Patterns of Unsustainable Specialization and Trade.

Feb 15 2011 at 12:31am

I may have to listen to this podcast again but I found it very difficult deriving any sort of interest or agreement (other than maybe on seemingly obvious elements) from this podcast.

First of all, Kling wishes to supplant current “paradigms” of economics with his new PSST model. So what exactly does this consist of?

As far as I can tell it basically involves telling vague stories about how things change and adjust and sometimes people have trouble finding ways to utilize workers, etc.

Really? So is a text book on using this model going to basically be a little pamphlet stating that recessions happen because people can’t find comparative advantages so don’t worry about it? Or is it basically going to be a German historical school type endeavor where you just write histories of economic activities without ever really creating models, predictions, or attempting to develop insights?

Fazzari told a much better story for one, and actual had a theoretical basis for it. I think you should have him on as much as Don Boudreaux and other common guests because I am always extremely fascinated with his arguments and positions. And honestly I have seen very little substantive push back on those ideas.

In fact most of the counter arguments are cloaked in the idea that modern Keynesians just don’t understand that aggregates are abstractions, inflation can be a problem, etc. when I have yet to encounter a Keynesian that is not aware of basically all of these things and it seems those debating the Keynesians don’t have a very good grasp on the Keynesian model instead of the other way around.

Feb 15 2011 at 12:15pm

Mike Laursen,

“Except that handing out massive amounts of money has got to be one of the best ways to build up Patterns of Unsustainable Specialization and Trade.”

Handing out money to banks is basically what we have been doing in recent history, I was trying to CONTRAST this with paying people to actually DO something.

In addition, even if one is a monetarist and believes that the mere increase in money supply will be beneficial, I believe this solution works better here as well for the reasons stated before, people will be more likely to actually use this money rather than hoard it.

Mike Laursen
Feb 15 2011 at 2:36pm

re: “As far as I can tell it basically involves telling vague stories about how things change and adjust and sometimes people have trouble finding ways to utilize workers, etc.”

I was under the impression, perhaps an incorrect one, that these are just issues that Kling has been thinking about on his blog. So, there is no Klingian Theory of Macroeconomics.

If you had an expectation that that’s what the podcast was about, then I can see why you’d be disappointed.

Mike Laursen
Feb 15 2011 at 2:40pm

re: “In fact most of the counter arguments are cloaked in the idea that modern Keynesians just don’t understand that aggregates are abstractions…”

I realize that Keynesians get a bad rap because politicians practice pseudo-Keynesian policies, at least when those policies align with what they wanted to do, anyway.

Mike Laursen
Feb 15 2011 at 3:07pm

re: “people will be more likely to actually use this money rather than hoard it.”

Let me give an example of why I am wary of handing out money, even if it’s for something the government has determined is a good cause. It can, and I believe does, bolster unsustainable patterns.

Say, I am entrepeneur who has started a new, still-small residential solar power company in the state of California. My company can outfit a typical residence with a solar electric system for around $7000 less than existing competitors because we have innovated in the areas of racking systems, labor steps, and type of photovoltaic cell. Just as my company gets off the ground, the state of California decides to hand out $15,000 rebates for typical residential solar electric systems, saving a few of my more established, more traditional and more expensive competitors from their marginal financial situations.

Their inferior product survives in the marketplace, my company has trouble competing and goes under. The next year the subsidy program expires, and those companies start having trouble again and start laying off workers.

This is just the kind of dynamic that looking at everything in aggregate form completely misses.

ND Earnshaw
Feb 17 2011 at 9:12am


It would be very interesting if you could get someone from the biology/ecosystem field on the show, to talk along the lines of this article:


Feb 17 2011 at 9:42pm

After listening a second time, I am beginning to understand the point to all the various government stimulus: they are trying to slow the mal-investment destruction rate to match the slow new investment rate. I will have to visit his blog to read up on his recalculation ideas.

This had been a most informative podcast.

Feb 21 2011 at 3:01pm

After a third listening to this podcast along with the Bruce Bueno de Mesquita podcasts, I now realize that political economic intervention is the result of the shortfall and failure of economic theories.

Emerson White
Feb 25 2011 at 2:40am

Hearing ” I want to add a comment that a listener made to the Fazzari post that I thought was very interesting. When I asked: why is it that if you are an unemployed computer person in 2001 you found a job fairly quickly even though there was a lot of unemployment among computer folks at the time, but in 2007 if you are an unemployed carpenter you are in trouble, and his answer was a simple one–in 2001 we still had a use for people who were in computers, not necessarily the ones they are in, consistent with your story; but we’ve got way too many carpenters right now. ” made me very happy, yay recognition!

Alfred Differ
Mar 1 2011 at 5:57pm

I had to smile through much of this interview when the maleability of the Keynesian model was discussed. Not being able to measure accurately while also being able to back fit what is found reminds me of 16th century astronomy. In those days we used a geocentric model with circular deferents, circular epicycles that could be offset slightly, circular epicycles on the epicycles that could also be offset, and so on as needed. When observed planetary motion did not fit the prediction, astronomers could work backward to what the circle sizes, revolution rates, and offsets had to be in order to have produced correct predictions. That version of astronomy didn’t come crashing down because it was wrong. It didn’t topple when Copernicus suggested a heliocentric model. It toppled only after one man (Brahe) took such meticulous data that another (Kepler) found it much too onerous to solve. Even that wasn’t quite enough, though. Kepler eventually tripped across the idea of using ellipses in the heliocentric version instead of circles and suddenly the model crystalized into something simple enough to operate. It was still emperical, but it could be used. I suspect it got adopted by the next generation more for its ease of use than anything else. Newton didn’t come along to provide principles behind it for a few more years.

Another astronomy parallel that has me smiling lately is the comparison with astrologers. Economists are being asked to predict our futures using very malleable models. Astronomers eventually forced an idealogical separation from Astrologers over the use of their models.

Mar 6 2011 at 11:59pm

I find Kling’s claim about tractors in the 30s and new jobs in the post-war economy VERY dubious. From what I heard elsewhere, during the depression lots of people moved from the cities (where people had been going for jobs) to the country (where they could at least eat). I had also heard that the jobs/industry picture looked pretty similar once the economy picked up again. Kling needs to provide some data to back up those claims if he wants people to take his alternate paradigm seriously.

On the other hand, I’m gland Kling aptly replied on attributing unemployment to construction. I’d reference Scott Sumner on the economy weathering the subprime crash until a NGDP shortfall in 2008.

Comments are closed.


About this week's guest:

About ideas and people mentioned in this podcast:Books:


Podcasts and Blogs:



Podcast Episode Highlights
0:36Intro. [Recording date: February 1, 2011.] Posts on EconLog trying to flesh out a new paradigm for macroeconomics. Begin with what you think is wrong with the current paradigm for macroeconomics and what might replace it. A paradigm is kind of like a language. Sometimes people are looking for some decisive experiment to choose between paradigms. I think it doesn't work out that way. You can say just about anything in English and say just about anything in French--but you wouldn't say that English is true and French is false. That's the way a paradigm functions. The traditional paradigm in macroeconomics since Hicks put it down--translated Keynes into this scheme of things--is aggregate demand and aggregate supply. That's the language that people use. I want to switch to a language that uses comparative advantage at its main core. What's wrong with the existing demand-supply story? I think there have been troubling aspects to it since certainly the 1970s, when a lot of ad hoc adjustments had to be made to make the aggregate demand-aggregate supply story work. There are long-standing complaints about micro foundations; there's a real issue of do we need micro foundations or not? Of course, Lucas at Chicago insisted that we do. Someone like Solow at MIT said that we don't. For me, I think the troubling things about aggregate supply and aggregate demand are there are very few instances where aggregate demand policies were successful. In contrast there are many instances of failures. The aggregate demand policies we've tried since 2008 have been failures. I would say that the Japanese aggregate demand policies have been failures. There are other instances. There are some surprising episodes like the period after WWII, massive reduction in government spending but without some sort of recession or depression that would have been predicted--and which in fact some economists predicted. So, many failures of predictions to come true. In retrospect you can always try to fit the data to your theory. It's very easy. I find that troubling. Frankly, a lot of my problem with aggregate supply and aggregate demand is I just don't like the intuition about it. I don't agree that the problem is all of a sudden people have stopped wanting stuff. Intuitively I don't find that a very satisfying description. For non-economists that can appeal to people. If you are not familiar with economics and don't know that supply and demand depend on prices, it's possible to talk about perpetual excess supply or excess demand. If I were to tell a non-economist: China is going to want to consume a lot more oil going forward and the production in many oil fields around the world is going to peak, non-economists would say there's going to be an oil shortage. Economists will say: No, no, they are just going to raise the price of oil. I think the whole aggregate supply-aggregate demand story is a story that mostly gets told without prices and is in line with that kind of thinking.
5:48Let's back up a little bit and go a little deeper into the paradigm as it stands. A little ironic because aggregate demand and aggregate supply try to draw on our microeconomic intuitions about supply and demand--but without the prices. One of the challenges of macroeconomics is that you've got so much stuff going on at the same time--by definition. You've got a market for credit, market for dollars, market for stuff--and all three of those interact. But how they interact is hard to keep in your head at the same time; write down simplified models of these things in mathematical equations so they can try to see and get a hold of these interactions. Normally we'd say: If aggregate demand goes up, or if aggregate supply changes, that's going to change prices and quantities like supply and demand does. That's not the story that the paradigm tells. Go into a little more depth that you don't like the argument or don't find it intuitive about people not wanting to buy stuff. How does that argument work within the aggregate demand-aggregate supply paradigm? Also has implications for labor markets, which are the two things we're going to have to juggle throughout this conversation. We've both been thinking about and writing about these issues. Things going on in the macroeconomy, some sort of snapshot of the economy as a whole; at the same time you have people looking for work, finding work, or losing their jobs. We usually think of those going together. They are not going together so well lately. We've got growth in the overall economy without much growth in employment. Whole separate paradigm we're going to have to think about; related in your mind and mine. Let's go back to the basic idea: What's the explanation of a recession or depression if you are using aggregate demand-aggregate supply? It turns out, as I found out in some sort of back and forth blog posts with various people, that this aggregate demand-aggregate supply paradigm, if you push on it a little too hard you find out that people really disagree what it's all about. So, let me lay out what I think is the textbook version. I think this is the version that if you go back to Hicks's article, "Keynes and the Classics," I think you'll see this. How can it be that people stop demanding stuff? The Keynesian explanation for unemployment is that people stop wanting stuff, and so producers can't sell stuff; and so they have to let workers go. Why would people all of a sudden stop wanting stuff? Keynes talked about people saving; but in a particular form--holding a non-produced good, namely money. So, if you switch your demand from wanting to buy something that's produced, like a car or bread, to wanting something that's not produced, like land or money--but let's focus on money--you no longer ask for production. You no longer want stuff. Aggregate demand. Or you want stuff tomorrow, down the road. Isn't the idea that you are anxious about the future, not sure what's going to happen, so rather than sinking your assets into stuff, you sink it into money as a precautionary measure--because of animal spirits--Keynes's term for mass psychology. I think for this purpose, let's call it hoarding. One of Keynes's views is that the entrepreneur, who is going to invest, is motivated by animal spirits. The saver who is going to supply savings to the entrepreneur is interested in hoarding. And Keynes hates the hoarder--the person who is just trying for safety. In particular, when people hoard money, switching from stuff to wanting to hoard money, that's a problem. If all you wanted to do was consume in the future then in theory you would be putting your money into capital investments. You'd buy stocks and the firms you invest in would be building assets for the future--productive plant and equipment for the future. There would not be an unemployment problem caused by saving. You'd just change from consumption goods to investment goods; might be some reallocation problems. But no unemployment problem caused by saving per se in that sense. It's more when people want to hoard money. I should back up. Keynes, because he believed that the animal spirits motive and the hoarding motive were completely separate, he would suggest that the person who saves doesn't end up changing the production of investment goods, because the production of investment goods comes from animal spirits. Which is somewhat akin to what we have right now. If you put your money in a bank, right now, normally banks lend money out to investors and entrepreneurs--people who want to do something productive with their money. Right now, banks are sitting on those reserves, anxious about the future; as are the possible entrepreneurs. That's sort of the situation we're in now. Would you agree? Could be. Anyway, you can have this phenomenon of aggregate demand falling. When people express it in terms of equations, it's often as if what they want is this non-produced good of money, and because they want this non-produced good, when you switch your demand out of produced goods you get less employment, less produced goods; you get the recession. The sense in which it's an aggregate demand relationship is that if the price level were to fall--the prices of all goods and services were to fall--then people's demand for money would be satisfied by the ratio of the existing money to prices going up, because prices have fallen. That's the sense in which you have a relationship between prices and aggregate demand. The lower the price level, other things equal, the higher aggregate demand. The idea there is the purchasing power in your pocket--the cash you are hoarding--will now buy more stuff if the price level goes down, so now you don't have to keep so much cash. You don't have to hoard as much because what you are hoarding now has more purchasing power. And the stuff you want to buy has gotten cheaper. That's the aggregate demand story.
14:17The harder story is the aggregate supply story, which is why is it that as demand falls, you get something other than a price change. In theory, lower prices will solve all the problems, why don't we just get lower prices and lower wages? So then the aggregate supply story becomes one of sticky prices and sticky wages. The idea--which I've never understood--that producers are loathe to lower their prices, or that it takes a while, or that it's costly to adjust prices. All of which are true, but it's hard to understand how that works in a time when people have all this excess production, output, sitting on their shelf. They just hope for the best? I don't understand the microeconomics, what's going on in the heads of the entrepreneurs in that story. The neo-Keynesian, new Keynesian claim is that it's expensive to change prices downward but not upward. I don't understand that. Actually, my dissertation was an attempt to tell a story about that: that there are search costs and people are more motivated to search when you raise your price. So, you are reluctant to raise your price. But when somebody else is lowering the price, I don't realize unless I'm searching that they've lowered the price. So they have less motivation to cut their price. Could be. What's your assessment of that work today? It continually gets redone. It was not published in a great journal; it was not cited, so people have continued to rediscover that mechanism for price stickiness over the years. There's something to that, but it's hard to understand how, in a time of extreme duress for a seller, that you don't lower your price. And similarly--let's talk about wage stickiness. Certainly a worker is hesitant to take a giant pay cut, for a whole bunch of reasons. One being it would be nice if another job came along that didn't require a giant pay cut--a shame to still be in the original job because it is costly to switch jobs, send bad signals to future employers. May brand you as a low-quality worker if you take a lower wage, is one of the worries. But that's the same argument. Similar argument being made in the labor market, why isn't the labor market more normal, more clearing in the usual sense of the word. In fact, the textbook model I referred to assumes that prices set by firms are fully flexible, and it's only the nominal wage that's the only sticking point in the whole economy. There are a number of problems with that, among which is that the cyclical behavior of real wages is not as pronounced as that model would suggest. What that model would suggest is that all of unemployment is caused by workers not accepting lower nominal wages when demand falls. You would expect to see very countercyclical real wage behavior: when you see low employment y you'll see high real wages and vice versa. That's very hard to tease out of the data. So that's a problem with that textbook story. I liked the image you used of the GDP factory. Talk about what you mean by the GDP factory and why you find it unsatisfactory. The GDP factory is what's called the representative agent model, which is that you choose as your modeling strategy to say we all work in the same industry. So there's just one labor market, one output market, one money market. We're making one good with one kind of worker. And all of a sudden we decide we don't want to buy more of it. Then when we go to work, the plant manager says: It looks like our demand fell, so I can't have you work as much. Then we go home and say we haven't worked as much, so maybe we want to demand even less. Then the manager says: I'm going to have to have you work even less. That'sa troublesome story for many reasons. One is: If the plant manager were all of a sudden to pay workers in terms of output rather than pay them in a money wage--just say: Here, have some of your GDP that you produced as output--all of the recession breaks down. It just disappears completely, which is bizarre. To this day you'll hear people argue that somehow in a barter economy you couldn't have a recession. I think that's absolutely perverse; I think that's 180 degrees wrong. I think in a barter economy it would be much easier to have a recession.
20:21Talk about the aggregation that's involved. So far you've been talking about what I would call the classical Keynesian model, some of the things that are wrong about it or troubling about it. Fazzari podcast--unrepentent Keynesian, good guy but who believes the exact opposite of you, that this is the right paradigm. That there are lots of times when people are anxious about the future, either on the production side or consumption side, and they do hoard or are hesitant to take risks; and that this leads to misallocation of folks in the labor market. Makes it harder to find a job, harder to keep a job, as firms and workers get more anxious. And that this is a very common problem that happens every once in a while in capitalism. What's your answer to that? If you really like the language of aggregate supply and demand, chances are you going to continue to be happy with it and to look at the data in those terms and think of things in those terms. The troubling things I think you will find is if you care about microeconomic foundations, that's a challenge; and if you look at actual predictions of the effects of policy and then the outcomes, I think you'll find those troubling. And the third troubling thing is I think if you go back to the 1970s, I think you'll find that the adjustments you have to make to get that paradigm to fit are troubling. Two other main competitors for paradigm conversations; similar but the proponents hate being described as similar. One could be a monetarist position of the Chicago school that says it's all about money; all about the Federal Reserve making mistakes, tightening when it's a really bad time when we're really doing well. And then there's the Austrian school that says that's part of the problem, but the other part is the distortion of interest rates and malinvestments that take place due to artificially low interest rates when the Fed is trying to change things. Each of those has this kind of GDP factory story to it. You really are talking about sort of aggregate output, though I guess the Austrian story would talk about different types of output, different sectors, different periods of production, different types of output or investment goods that yield results over much longer periods than others. I don't want to say my point of view is unrelated. How about I put the other view out there so we have something to talk about? What I want to do is instead of describing economic activity as spending, I want to describe it in terms of comparative advantage. I use this acronym PSST--Patterns of Sustainable Specialization and Trade. But that really is just another way of saying patterns of comparative advantage. When we have a boom, what I would say is people have found a lot of ways to take advantage of comparative advantage. When people are working, people are doing something that exploits comparative advantage. You'd better explain that, though. I actually like "patterns of sustainable specialization and trade" better because for non-economists in the audience, comparative advantage is a very difficult concept. I think specialization in trade is better. A simple example is if I cook my own dinner tonight and you cook your own dinner tonight, there is no measured economic activity involved in the cooking. There might have been in the buying of the food, but not in the actual cooking. No measured economic activity. If I cook dinner for you and charge you for it, as in a restaurant, and then you do the same for me, so that you eat in my restaurant and I eat in your restaurant, then that becomes measured economic activity. That may seem sort of strange--same activity, cooking and eating--but if you think of it in terms of true specialization in the economy, it actually does make sense. If a surgeon mows her own lawn that doesn't count as economic activity, but if she does another activity and then pays somebody to mow her lawn, that does count as economic activity. I think that's actually right. It would be a tremendous waste if we tried to do everything for ourselves. You couldn't build your own computer, build your own car, find your own gasoline. Even make your own pencil, as we know. All of that economic activity actually involves specialization in trade. We measure it as spending, but what it really is fundamentally is specialization in trade. What's happened is these patterns of specialization and trade have become very complex. Last night my wife was looking at an ad for a company that makes custom shirt sizes--they will send you a shirt that has a combination of neck sizes and sleeve length that isn't standard. If you asked somebody a hundred years ago: What would a shirt company consist of?--it would be tailors and fitters, maybe sales people, people who obtain fabric and process fabric. For some reason, my wife was looking at the website to see if they were offering jobs and what kinds of jobs, and it was web designer, business development manager, customer service representative, social media marketing person. So, it's a really strange economy that somebody's comparative advantage could be social media marketing for a shirt company. Shows how complex the patterns of specialization and trade have become in this economy. If you went back 500 years instead of 50, a shirt was a very different animal; simpler, fewer choices. Even 200 years ago, a lot of people made their own clothes here in this country. We have progressively outsourced more and more tasks; the tasks have become more complex; and the patterns of trade have become more and more complex. Done podcasts on specialization and division of labor, worth going back to the first few chapters of Adam Smith where he talks about it. He basically says that in 1776 when life was vastly different from now, but a little different than a thousand years before; and he made the observation that where these patterns exist, you get a lot of surplus. Because otherwise, you spend all your day making your own clothes and food, you are kind of done. Hard to make enough clothes for yourself and for others. What these patterns of specialization do is to create enough stuff that we have enough left over to interact with each other. And that is what the market is that you are talking about. You are paying people to do it. Not just that it's easier. That whole web of interactions is what allows us to have an unfathomable standard of living compared to Adam Smith's time.
29:41Incredible surplus that gets generated by that process. If you make your own clothes and grow your own food and make your own food and make your own shoes and keep your roof on by thatching it all the time, you are never unemployed. Exactly. A curse, not a blessing, it turns out; there is a negative side to it. In traditional economics, a job is not something that you worry about needing. Work is not a good thing. Work is bad. One of the other peculiarities of macroeconomics is to treat work as a good thing. We've got create jobs. We've got to raise demand in order to create jobs. We've got to spend more. A job in some sense is a bad thing. What you spend most of your time doing is eliminate jobs, eliminate work, make work less onerous, less risky, and so on. In some sense that's what any economy is all about. So, when we get to this patterns of sustainable specialization and trade, we're trying to talk about that activity. In the traditional paradigm of aggregate demand and supply, it's sort of like these equations need to be satisfied. There's an equilibrium there somewhere and the process of full employment is the process of getting to that equilibrium. You just have to get "the" wage. The right wage. The aggregate wage. The right level of spending or whatever. But patterns of specialization and trade--what I want to talk about is the process of finding this comparative advantage. How do you figure out if your comparative advantage is to be a social media marketer for custom size shirts? How does that even happen? It's a case of entrepreneurs coming up with ways to create and exploit comparative advantage and then workers finding those opportunities. That's what job creation is; that's what the process of achieving full employment involves. What I want to argue is that that process can get confused, and that right now there are a lot of people who are in a situation where the comparative advantage relationships in the economy have changed. So, ways that you could have made a living a few years ago, you cannot make a living now. The classic one being extensive home-building. We don't need as many people building houses as in 2006, or selling houses, or securitizing mortgages and trading those. A lot of things cascade from that. In addition, there is just ongoing change: we don't need as many people manufacturing a lot of goods because we keep getting better and better at manufacturing using automated techniques. Just hearing in the news that Borders bookstore is going out of business. Why? Is it because aggregate demand is too low, overall spending is down? I suspect no. I suspect the comparative advantage of having a physical inventory of books in many locations has gone away relative to the Amazon model of having the inventory located in a single place. They probably have a lot less inventory. Instead of putting two extra copies of a book in thousands of locations, maybe they have eight extra copies in one location. Probably tremendous economies there. Plus the explosion with the Kindle reader, not reading physical books at all. That's creative destruction, right? That's the standard story that we think we sort of understand about how progress takes place, that innovation occurs: somebody invents the Kindle, the e-book, the internet comes along that allows somebody to stockpile books in one location and save money on inventory. Great for readers, great for Jeff Bezos and Amazon; tough for Borders and people investing elsewhere who now find themselves with excess higher costs and now going out of business. So that's going on all over the economy. Goes on all the time. If you think of the economy as a set of equations that should be solved in every instance, then that should not cause unemployment. The equations get solved; the people who were stocking, shelving, selling books at Borders are now instantly going to be reallocated into new jobs. Because the aggregate demand that left Borders is now going to Amazon and elsewhere. There's extra money available because Amazon's cheaper. So the aggregate demand shouldn't be a problem.
35:58But the problem with your story is that that's kind of true. I like your story, but I've got some problems with it. In "good times," meaning not a recession, this creative destruction is taking place all the time. It's ongoing. We don't have a lot of problems. Somehow, even though the person who was greeting you at the door of Barnes and Noble and restocking the books physically on the shelf doesn't have the skills to be the social media person at Amazon, which is now the new kind of job that is suddenly more productive than it was before--Amazon going to use their Twitter and Facebook accounts to stay in touch with sales--somehow those reallocations just work okay and fine. So there is something going on that's different today than, say, in 1985 or 1996 or even 2004 when these changes were going on all the time. We've talked about this; and we talked about it two weeks ago with Steve Fazzari: In the JOLTS data [Job Openings Labor Force Turnover Survey] that the Bureau of Labor Statistics collects, there's millions of jobs being destroyed and millions of jobs being created every quarter. But what are called bad times are when the numbers that get created are not quite as many as the numbers being destroyed; and good times are when the numbers created are a little more than are being destroyed. So my challenge to your paradigm is: The PSST idea, the idea that it's hard to find patterns of sustainable specialization and trade, somehow in good times works pretty well. So something is different. That's a fair question, and I end up telling more time-specific stories for it. So, in the 1930s I would tell you that what happened is that tractors really took hold on the farm, and we could basically produce all the food that we needed with way fewer farm workers. So, that just creates this massive dislocation of lots of people whose skill is manual labor. Entrepreneurs just could not figure out--the economy did not figure out what to do with this excess of manual labor until the 1950s, when a lot of these manual laborers had just retired out of the workforce and the people who might have been manual laborers went off, fought WWII; in the process made new friendships, got uprooted from their old locations where they maybe couldn't have found jobs. In the 1950s they come back and they work in sales or do clerical work or in growth industries of fast food and motels and the industries that grew up around the suburbs. The economy that came back in the 1950s was very different from the economy that got left behind in 1930. It didn't get healed. It was different. Had many more women in the labor force; and it wasn't Rosie the Riveter in manufacturing; it was women doing clerical work. Very different ways of organizing and structuring production. And what I suspect is going on now is something like that vis-a-vis the Internet where a lot of skills had just become obsolete. One possibility is that as of a few years ago, 2008, when a lot of companies looked at their balance sheets and said: We don't know if we'll survive; we have to cut back to the bare bones--there was a rather sudden discovery that they had workers they didn't really need. So, this adjustment process sped up all of a sudden. It's an interesting idea and what it suggests is there are times when it's harder to find an opportunity within the patterns than it might otherwise be. But examples I think about a lot: The end of WWII comes along, and all of a sudden there's a few million people looking for work. Some of them go off to college. Not very many. That was maybe a million out of the ten million. And then you've got all these women, on top of that, you are going to have another trend that is going to go for another 30 years and continuing, which is not just that women are going to be in the labor force, but a steady increase in the interest of women in working, or a steady set of improved opportunities that make them more willing to work. We don't know what it is. Whole bunch of things going on at the same time--the birth control pill, smaller family size. Over that time, enormous increase in the number of women wanting to work; enormous inflow, sharp increase in the number of men looking for work. Somehow it works out beautifully. They find opportunities and everything's good. No big unemployment. I would just have to describe that as entrepreneurs got lucky and skillful in figuring out the kinds of businesses to start and their businesses kept rolling. The guesses that they made, a lot of them proved to be good guesses. But in 1999, a lot of entrepreneurs made some guesses about what happened with the internet, in terms of, which failed. In 2005-2006 a lot of real estate developers made some bad guesses as to where they could build houses and shopping centers and so on. Unfortunately, that's about as good an explanation as I can give for the recession: people made some bad guesses about where the patterns of specialization and trade will be, and then on top of that some businesses that had kind of been hanging on to workers or processes kind of without having to think too hard about them--all of a sudden, we're faced with the process of a hanging so to speak, so they said: We're going to cut back.
44:18One of the things we haven't touched on is the nature of the way that workers contribute to businesses. I call that the Garret Jones factor: he says workers no longer produce widgets; they produce organizational capital. A lot of the jobs I can think of are jobs that are not necessary to produce something or to put something in the hands of the consumer. They affect things more indirectly. They affect the firm's ability to connect with other firms, it's ability to manage its processes, track what it's doing. People are building capabilities for firms. Different type of worker. Worker you hire the way you invest in capital--you make a very careful decision, don't necessarily rush in and bring on a worker like that; wait until you are ready to make a big investment in new capabilities. That's a very different hiring process; a lot of fixed costs in that hiring process. I think that somehow is in the mix of how you get a recession and how it can be very difficult to come out of a recession because investment is something you can always postpone. You may be less likely to postpone it when other firms are doing well and you have this fear of missing out if you don't invest quickly; and you may be more likely to postpone it when you look around and see other firms are retrenching at the same time. Let me challenge that story. Want to come back to previous story about post--WWII. Let me make the observation that the worst levels of unemployment--and I think this is true across all recessions--are in the lowest-skill jobs, people with the least education. There are two patterns to this recession that I think--one part is fairly standard and one part not, perhaps. The standard part is that there is a higher unemployment rate if you have less education. We also have a very generation duration of unemployment benefits. The level is not so generous, but when you extend unemployment for 99 weeks, which is two years, that does not sharpen the mind; and it's easy for me to say--I've got a job--not pleasant to be on unemployment compensation because it's not. But it certainly makes some people less willing to take a chance on a new job, a new profession, a new opportunity. So, you have that, laid over the fact that you have a very high unemployment rate over low-skilled workers. The so-called knowledge worker, people with high levels of education, who I think of as the people Garret Jones are talking about--they are not struggling so much. They have a higher unemployment rate than they did two or three years ago--it might be 6% or 5% or 4.5% instead of 1.5%--so it's up, but it's still the case that if you are a knowledge-type worker and you have high skills, I think you can find work fairly quickly. The people who are not finding work quickly I think are people who have two characteristics: they have low education. That's not unusual. Perhaps what is unusual is that it's very geographically varied and different. So the highest unemployment rates are in California, Florida, Michigan, Nevada; perhaps it's a coincidence, but these are the states that had the largest increases in home construction. With the exception of Michigan. Well, but Michigan had a--not so sure--it did have a big housing boom and bust. The parts of the country that did not--Texas cities, North Carolina didn't. I suggest we might go even more micro in your thinking and think of this really as--ironically, we are both heading in a very Austrian direction here. You are moving away from aggregation; I'm suggesting it's sector-specific. You are talking about entrepreneurs finding and searching for opportunities, which is a very Austrian idea. Is the problem perhaps not any more complicated than the following, which is that between 1997 and 2007, we built an enormous number of houses that was not sustainable; hundreds of thousands, closer to a few million workers were out of work because they had been pulled into that sector by the attractive opportunities; but now those opportunities are gone. And they are not coming back for a long while. Couple of problems with that. I want to add a comment that a listener made to the Fazzari post that I thought was very interesting. When I asked: why is it that if you are an unemployed computer person in 2001 you found a job fairly quickly even though there was a lot of unemployment among computer folks at the time, but in 2007 if you are an unemployed carpenter you are in trouble, and his answer was a simple one--in 2001 we still had a use for people who were in computers, not necessarily the ones they are in, consistent with your story; but we've got way too many carpenters right now. There were 200,000 more carpenters a few years ago employed than now, and those jobs are not going to come back for a long time. Only existed because of a really bad set of policies that pulled people into that sector artificially. I think that's certainly part of the story. Some counters to that: It's hard to account numerically for a large chunk of unemployment by simply looking at construction workers. It's about a quarter isn't it, or maybe about an eighth? Maybe even less, depending on how you construct your time period. I'd go from December 2007 to the present; and the fall in employment--I messed this up in the last podcast but I think a quarter of the jobs that have been lost were in manufacturing and in construction. But for manufacturing, it's hard to tell the housing story. Correct. Michael Mandel has recently put up some charts--I think he looks at the total wage bill, where that's been going up--it's clearly in health, education and government, not some of these other sectors. Definitely these sector shifts going on. My view is a lot of it was just ongoing sectoral shifts that got really accelerated at least in terms of letting workers go during this recession for reasons I just would have to tell stories about--that firms almost like a herd behavior decided now is the time to focus on controlling costs and not a time to expand. And at the same time some shortage in some sense of firms in the expansionist mode. The mystery in some sense of what's going on right now is: Why aren't there entrepreneurs looking around saying: There are all these people who could conceivably come to work cheaply. Why don't I take advantage of that, find their source of comparative advantage, exploit them at low wages and bring them back into the labor force? If you look at that question that way, lots of reasonable answers suggest themselves. One is: When you have to pay health care benefits, and the cost of those health care benefits is doubled, that becomes a real issue. I think there are some very important secular changes and the health benefits may be the most prominent, that are going to reduce the proportion of people who can be profitably employed going forward.
54:12But again, we also seemed to manage to overcome those challenges. We could argue they have suddenly gotten so much bigger--no doubt about it. Just to correct the numbers: From December 2007, which is the start of the recession according to the official NBER announcement, to October 2009--roughly 2 years later--employment was down 7.2 million and 3.6 million of that was construction and manufacturing, so half of the job losses. Now, I agree with you it's a combination of stuff. Of that half, 1.5 million in construction and 2.1 million in manufacturing. So it's a little bigger in manufacturing. And as you say: some of that is just an acceleration, maybe of productivity, maybe something else. Maybe there's an aggregate demand story--when you suddenly throw large numbers of carpenters, real estate agents out of work, maybe there is something going on there. Again, we talk about patterns of comparative advantage. If I haven't found something to exploit my comparative advantage, that's going to make it harder for you to exploit your comparative advantage by selling something to me. I'm home cooking and making my home clothes. I do think there's a psychological issue here, and I think it goes back to the 2001 tech burst story. If you were a webpage designer in 2001, which a friend of mine was; and he lost his job; but within a very short period of time found another job. But while he was doing that, looking for work, he decided he had better tool up; he got better at data base management, which is what he does now. That was hard; he had to make some decisions and think about what he wanted to do. He could have left computers and technology completely, but he decided that was still a good idea; and he either got lucky or he was wise. He still has a very good job. If you are a carpenter in Nevada, you've got a tough decision to make. The way I would describe that is that fundamentally the guy who is doing computer work in 2000 is in a growing industry. Exactly. There was kind of a negative blip there, but long term, there have got to be more web developers now than there were at the peak of the dot com boom. Just a matter in some sense of waiting. Maybe supply temporarily got ahead of demand. Your skills are still valuable. Whereas the manufacturing worker and the carpenter--that long-term trend is only going in one direction. At least for a while. It may change in 5 years. Well, you could get a cyclical upturn, but my guess is that productivity in home construction is going to trend up. If you look at home construction 20 years from now you are going to see a lot less labor intensity. The way to think about it is long-term growth in employment depends on how demand grows relative to productivity. For a long time in manufacturing, productivity has been increasing faster than demand. Whereas in things like health care and education, demand has been increasing faster than productivity. So you know over long periods of time we are going to shift workers out of manufacturing into health care and education. And no matter how much Democratic Congressmen and Senators may think that's an awful thing--we're not making things any more, it's terrible--it's just an inevitable long-term trend. One way to interpret the current recession is that that trend accelerated in some sense except that the uptick in education and health care is going to accelerate over the next five years whereas the downtick in manufacturing was accelerated in the last two years. I think that's definitely a part of the challenge.
58:59Let me ask you an economics education, philosophy of science type question. You and I both went to graduate school in the mid-1970s. You went to a salt-water school--meaning near the coast--MIT, which is more prone to be a Keynesian place. I went to what is called a fresh-water school, which is a monetarist place. Our teachers and the people who specialized in becoming scholars of macroeconomics, most of them stayed within their silos. If you were an MIT-trained person you tended to stay in a Keynesian type world view and if you were a Chicago person you tended to stay in a Chicago-type world view. As you matured and went on in your career, you and I have actually moved closer together, somewhere over the Atlantic. As Don Boudreaux used to tease me, I'm a mix of Austrian and Chicago, so my home is somewhere in the Atlantic Ocean. You and I have both sort of migrated together in that direction. We are less comfortable with aggregation, less comfortable with elegant mathematical models that don't quite describe things the way things seem to be. Do you see any evidence that our friends and colleagues who have stayed in more traditional approaches have found the current state of affairs a legitimate challenge to their world view, as opposed to just saying: Well, we just need to fiddle around with the model a little bit, make it better? What are your thoughts on where we are heading as a profession? Do you think this is going to be a watershed experience, or that we'll just recalibrate a little bit? My first thought is that if the 1970s didn't wreck the aggregate supply-aggregate demand paradigm, then what will? In some sense it's easier to use the language of aggregate demand to describe the current story. If you push on it a little harder it's troubling. The nominal interest rate didn't go up; government spending didn't go down; the money supply didn't go down. Always ways to tell stories. Robert Hall will tell you that the risk premiums stayed high--although I don't quite see the evidence for that the way he does--and then that along with the 0-nominal-interest rate bound meant something. The solutions don't seem to work the way they were purported to work. They have a story but it didn't work out the way they had hoped. The story is that actually the economy was in much worse shape than we really thought so what we did wasn't enough. I think it's a language and they can describe what's going on in terms of that language. I think if you do have the ability to shift gears and think about patterns of sustainable specialization and trade and think about the process involved in creating those patterns. Just imagine if you are an entrepreneur today. You've got something like 10 million unemployed workers to choose from. In theory it ought to be really easy to create a business. But put that hat on and say: Well, what business am I going to create that's going to make money, and it's not easy to do. The notion that by undertaking some aggregate spending policy you are going to enable entrepreneurs all of a sudden to discover those comparative advantage opportunities--I think it's somewhat implausible. I use this analogy of this old joke Bill Cosby used to tell: person drives into a gas station, doesn't know where the gas tank is on his car; the gas station attendant can't find it either. And the attendant says: Well, maybe we should just pour some gas all over the car and maybe it will seep in somewhere. To me that's what aggregate demand policy is trying to do relative to finding patterns of sustainable specialization and trade. It's not helping you find those patterns. You use the word "sustainable." That's caused some confusion because a lot of people associate it with an environmental matter. All I mean by that is profitable. The government can certainly create employment opportunities, but if they are not profitable opportunities, ultimately the question is how do you sustain them. I think of a product that works for a few weeks but after a while people don't want it any more. That's not sustainable. Or the housing bubble. Doesn't mean sustainable forever. Ongoing. I don't think economics and environmentalism are that opposed. In some sense you are thinking about using as few resources as possible. The economy is trying to figure out that problem; environmentalists are trying to figure out that problem. Come at it sometimes from a slightly different angle, but there's really a lot of overlap. Sustainable is hardly a dirty word in economics. Another sense in which this paradigm goes back to the classics--and by classics I mean before Keynes--is the emphasis on production rather than consumption, which is I think the right idea and a great way to think about economies at large. One thing you haven't talked about and is powerful about the paradigm is we might imagine what makes patterns of sustainable specialization and trade easier to discover versus harder. You mentioned one--various wedges of some kinds, things that distort or potentially distort the gap between what the worker receives and what the employer has to pay. So, various benefits that are artificially increased via bad policies, such as health care could be an example of that. Two other things: mobility and provision of information. You used the example of post WWII that people who had lived elsewhere found it easier to not just go back to their home town. I suppose the information that's available today should make it easier; but it has not been enough. Another impediment I think we have is a lot of what I call credentialization. A lot of industries that you could reorganize if you had all the degrees of freedom, you cannot because of the laws that protect people based on credentials. Most notably, education and health care. Long term, these are the growth industries in the economy, yet they both are very rigidly tied by credentials. You cannot redesign a health care clinic so that someone other than a doctor performs a function without running into legal problems. You cannot organize a school and get that school accredited--which is very important in this world--without jumping through lots of legal hoops. At public schools, you have to be certified to teach there; requires a lot of hoops, many of which have nothing to do with being a good teacher. So, the sectors with the greatest potential for growth are the sectors most rigidly held by these credentials. I think that's a major factor slowing the adjustment in the current recession.