Arnold Kling  

Keynes vs. Bohm-Bawerk

Recalculation Story Watch... How FDA Regulation Stifles Dru...

This morning, we will find out whether the recovery remains jobless. Meanwhile, I have an essay, Labor is Capital, that offers an explanation for jobless recoveries.

Much of today's American workforce is engaged in roundabout production, which Böhm-Bawerk equated with capital. There is no longer a meaningful distinction between labor and capital. Labor is capital.

If labor is capital, then we have lost the automatic tight connection between spending and employment. Firms can vary their output with little or no variation in employment. This explains how we can have a "jobless recovery," meaning a large percentage increase in output without a comparable percentage increase in employment. For firms in today's economy, labor represents an investment. Firms hire workers in order to develop capabilities that will eventually produce output more efficiently. The return on an investment in workers may take as long or longer to realize as the return on investment in a machine. The return on investing in workers may be at least as uncertain as the return on investing in equipment.

Read the whole essay. It really spells out differences between the Austrian-style Recalculation story and the Keynesian spending story.

[UPDATE: Last month's increase in private-sector employment was only 71,000. The recovery remains pretty jobless. As I remember it, key Administration officials get a "sneak peek" at these figures several hours before they are released. However, my guess is that the timing of Christy Romer's resignation as CEA chair is just coincidental.]

Comments and Sharing

COMMENTS (11 to date)
david writes:

Keynes demand-side policy works just as well for unemployed capital as well as labor, so there's no problem if human capital is, well, capital. And your argument about sustainable patterns fails to hold for (say) Keynesian temporary tax cuts.

I'm sure you know that nothing in Keynesian theory necessitates industry-specific spending, even if the readers of AEI editorials might be more ignorant. Don't be disingenuous. Temporary payroll-tax cuts and other broad such measures are wholly Keynesian.

The stereotypical Keynesian story of digging a hole and filling it back up has obviously no intent in perpetuating the practice - the idea is just to put the remaining unemployed labor and capital back to work by increasing current consumption.

Rebecca Burlingame writes:

In the coming recalculation, I have a number of concerns, and your essay gave me a way to voice a few. For one, the questionable efficacy of money in that it moves to the newer perimeters of economic activity after a period of time, as you explained, which leaves the velocity in primary economic activities (the harvest) almost too low for economic flow in local areas. Hence government money creates rural economic activity to an unbelievable degree.

For some time, I have not felt human labor to be a part of capital (in monetary terms), but your essay reminded me that much of the capital used in roundabout production is now threatened in much the same way as human capital itself. Only look at the ongoing auctions of so much technical equipment that often can scarcely be resold. This suggests that the deflation of primary capital assets is not unlike the deflation of human capital and they may therefore be thought of in the same light.

fundamentalist writes:

David, I think you're making Keynesian economics too broad to be of any use. Like the theory of evolution, your theory of Keynesian economics can explain anything and its opposite. Such theories aren't much use.

Nice job, Dr. Kling on the essay!

bob writes:

It would be impolite to fully convey my reaction to "read the whole essay" when you wrote the essay.

Ted writes:

Your column has quite a few problems, I hate to say =/

Firstly, you are talking about Keynesian theory that died 40 years ago. Why must you insist on using a dead straw man? Modern theory doesn't work by just putting people to work like you basically claim. Modern New Keynesian macro says that when the zero lower bound becomes binding (I refuse to use the term liquidity "trap" since it's not really a trap at all), then an increase in government consumption acts as an autonomous spending shock which pushes up inflation expectations since, unlike in normal times when the monetary authority would just neutralize the government expansion by raising interest rates, supposedly now when nominal interest rates are zero they welcome the increase in the price level and don't move the rate (there is another way fiscal stimulus can increase output, but it doesn't work in countries with independent central banks). Incidentally, this also means all the "implementation lag" discussion was nonsense in the spring / summer 2009 since the expectations of the government spending should cause the price level to rise right after it is passed - it doesn't matter when the projects actually start. Of course, what I just described it something that monetary policy can do on its own. What always confused me about the argument for fiscal stimulus is that even as a theoretical proposition it is suppose to work by pushing up inflation expectations, but the central bank can do just that up committing to a higher price level or nominal income target (I don't buy the deflationary bias / time-inconsistency arguments against this policy) - so why even bother with fiscal stimulus when monetary stimulus should have a more certain result without the massive future tax liability? It's a mystery to me. Also, I believe in some cases that the necessary fiscal expansion, in absence of good monetary policy (since you have to assume crappy monetary policy for stimulus to be even be on the table), would need to be so extremely large that government would begin to consume goods and services that are approximate substitutes for private consumption, and so just crowd out and have no effect.

But also let me correct something you state, wrongly. You argued that "in spite of the application of stimulus treatment ...". You have a problem here. There was no net stimulus! We got approximately $0 worth of "Keynesian" stimulus. All the ARRA did was offset state fiscal contractions with federal fiscal expansion (see this paper by Aizenman and Paricha: What matters in the NK model is aggregate government spending, not just federal expenditure. Since it appears that we got $0 worth of net stimulus, you wouldn't expect it to do much good at all. Now I would never recommend fiscal stimulus as a policy option since I think it's unnecessary (and we are uncertain whether the NK models theoretical case for stimulus is even correct), but I would have loved to see a massive fiscal stimulus in the United States for purely intellectual reasons. I would have loved to see whether the NK models theoretical predictions played out as one would expect.

Secondly, I don't understand your labor is capital story at all. It seems entirely incoherent and your explanation is just so muddled I can't even refute or agree with it. It just doesn't make sense.

Thirdly, a recalculation story is plausible in a few industries. It is not plausible as a wide-ranging phenomena across multiple industries as it is now. What was everything just "overproduced"? I don't think so. A more likely explanation is that the public is unsure what policies the Fed is going to pursue in the future and so do not want to make sunk cost investment. Banks are especially unlikely to extend loans if they believe the Federal Reserve will allow deflation (which is looking more plausible by the day), since they then know their loan won't be repaid (and then you get the fun debt-deflation phenomena!). In terms of the job market, there are three points. Firstly, employers don't want to sign long-term contracts with growth prospects uncertain, due to tight money. Secondly, AD was never restored to its proper level due to bad monetary policy so hiring won't commence until AD is restored. Thirdly, I have a countercylical labor hoarding theory which I freely admit has no empirical evidence besides just my casual observations. Every firm attempts to maintain the optimal firm. However, that point is difficult to find and requires a bit of trial-and-error. What I posit is that during boom times, producers and business owners become more lax with maintaining their firm because it requires a lot of cognitive effort. During the boom years it's simply not worth the producers time to put in the effort to maintain the correct firm size, the mental cost of improving the firm is simply not worth it to people when they are extremely profitable. So, what happens is you let some things go. Alice and Bob, who contribute nothing to the firm, are kept on the payroll because the producer doesn't take the time to investigate who is doing their job. All of inefficiencies go unnoticed and technology depreciates, as well as cost-saving technology adoption is foregone because of the cognitive cost of investigating it. But, during a downturn the monetary cost of letting these things go outweighs the cognitive price the producer has to pay. Not only do you have the drop in demand for the firms product, which naturally leads to layoffs - but you also have a sort of unit root reduction in the firms labor force. All of a sudden, the owner realizes Alice and Bob are worthless to the company - they never did anything anyway, so they are fired. He realizes that if he upgrades to this new technology he can save himself money by firing some people who are no longer needed since it utilizes labor more efficiently. Etc. Etc. Furthermore, as the "boom" is longer and longer more and more inefficiencies build up and so the reduction in firm size is more substantial once the recession hits. This gives you a counter-cylical labor hoarding. This is why I think we have jobless recoveries now. Good monetary policy has extending the boom period for a much longer time period than over the last century, which means more and more inefficiencies build up. So when the recession does come, more of the reduction in the labor force is semi-permanent. Returning the economy to proper NGDP growth isn't enough now because a lot of the labor utilized wasn't needed and less of it is needed in the aftermath because of efficiency improvements. So, to return to "full employment" one needs to wait longer for growth to "catch up" and for the lax producers to produce inefficiencies again.

Finally, let me also discuss something else you recommend - payroll tax cuts. That's a terrible idea given our current situation. Yes, payroll tax cuts are usually expansionary, but they also impose deflationary pressure. Payroll tax cuts cause real wages to fall and so producers provide more goods and services at lower prices. Normally, the Federal Reserve follows a so-called "Taylor Principle" and so the nominal interest rate falls by more than one-for-one and so the real rate falls and inflation expectations remained anchored, while output expands. But with our idiotic Federal Reserve they would just watch the deflationary pressure from the payroll tax cut come along and probably just sit there since they can't move nominal interest rates down anymore - which means deflationary expectations would set in, the real rate would rise and output would contract.

fundamentalist writes:

Ted, I'll have to take your word that NK is different from Old Keynesian (OK), but your description doesn't seem much different to me, except for the expectations part.

the "implementation lag" discussion was nonsense in the spring / summer 2009 since the expectations of the government spending should cause the price level to rise right after it is passed - it doesn't matter when the projects actually start.

Didn't happen though, did it? That's because consumers don't base their expectation of inflation on government policy. They base it on past inflation rates.

There was no net stimulus!

Is that supposed to be one of the great NK innovations? Seems to me more like an excuse for the past failure of OK. Keynes understood that local government revenues contract during a crisis. He assumed that federal spending would make up for it. The whole point is that federal spending makes up for all of the shortfalls in spending, private as well as local government spending. So even though there wasn't "net" stimulus, there was stimulus and very large. It should have had some effect. Except for Blind and Blinder's paper no one can find such an effect.

A more likely explanation is that the public is unsure what policies the Fed is going to pursue

Uncertainty fits will with recalc macro. And recalc doesn't apply within industries, it's a cross-industry analysis. Some industries got too big, bigger than what consumers wanted from them, and workers in those industries need to find jobs in other industries. Of course, there are some industries in which demand merely declined because workers in industries that got too big have lost their jobs.

Yes, payroll tax cuts are usually expansionary, but they also impose deflationary pressure.

Seems to me you have it backwards. How will payroll tax cuts cause inflation? Tax cuts might cause businesses to hire more workers, who will then spend their earnings and cause price inflation.

ThomasL writes:


I think you'll find that outside economics journals New Keynesianism has little purchase. When the New Keynesians find themselves explaining to the politicians how the pieces fit together and what they policy should be, they seem to reach for a [simplified] Old Keynesian explanation, not the version from their latest paper.

Lord writes:

Do businesses invest when they can or when they must? A bit of both, I would say. Do they wait for good times or bad? Neither, I would say. Sustaining demand may be a prerequisite though. What is really needed are opportunities. Business will say there are always opportunities, but they do not always act like it, or their opportunities vary widely in quality. Innovation is rarely done to plan, at will, in time, or on demand. We can only make things favorable for it, and deflation is not one of these. A German work sharing arrangement would be more equitable, do much to limit the damage being done to human capital, and reduce demands for stimulus. It is really our inadequate economic system that is at fault.

Hyena writes:

I really don't see the mismatch between "sustainable patterns of trade and specialization" and Keynesian stimulus programs.

If some pattern is unsustainable, it won't outlast the stimulus, sure. But if everyone is aware of that fact, then people will invest the profits of current activity into new patterns rather than try to hoard enough cash to survive the long night.

The point isn't to preserve the old patterns but to give people the confidence they need to create new ones.

Carl The EconGuy writes:

Arnold, a quibble on your paper: working for a private non-profit is just as welfare-enhancing as working for a for-profit. An NGO has to satisfy its donors, who get consumption value for their contributions, along with tax benefits. Every NGO has to compete for resources with for-profits. What's not to like? Your two kids deserve the same respect for their contributions as your buddy Alan's, because on the margin they earn every dollar they make, just as for-profits do. So, get off their case, will you? At least, they're not doing wasteful government work, right?

On your labor-capital theory, a point. Your essential argument is that in a complex economy the cost of producing specific human capital increases for every firm. That makes the recalculation problem more difficult for the producer, you say. But there is an equally problematic issue on the labor supply side. The more specific a person's human capital, the greater the differential (the rent) between the current job and the opportunity cost in the next best job is likely to be. The greater the rent, the more difficult a worker's recalculation problem becomes, in case of unemployment. So, the more specific HC becomes, the longer expected spells of unemployment will be, and the greater the average windfall loss of HC wealth for the worker. Of course, UI benefits will only delay the required recalculation on the labor side, as we all know.

James J. writes:

The entire Labor = Capital concept seems problematic to me. That argument would seem to warrant for Producers to be more hesitant to shed workers in whom they have invested in. Instead, we have seen the exact opposite, the diminishment of the worker as an investment, and his treatment as a objectivified input into the Productive process, whom one wants as little investment in as possible, in order to easily divest of excess capacity as need arrives.

The cause of this may perhaps be multifaceted: the (government imposed) increased costs of long term hiring, and specialization and the accelerated need to "recalibarate" your process to fit changing End-Product requirements spring to mind.

Comments for this entry have been closed
Return to top