CEPR - Center for Economic and Policy Research


En Español

Em Português

Other Languages

Home Publications Blogs Beat the Press Capital Biased Technological Progress: It Doesn't Just Happen

Capital Biased Technological Progress: It Doesn't Just Happen

Wednesday, 26 December 2012 11:12

Paul Krugman has been rightly troubled by the continuing shift of income shares from labor to capital. However the explanation he considers in the form of capital-biased technological progress requires a little more careful examination.

Krugman discusses the case where there is an exogenous change in the nature of technology that makes capital relatively more productive than labor. This leads to more capital being used, driving up its price, and less labor being used, driving down its price (i.e. wages).

This is a relatively straightforward story, but there is a serious problem. Capital is not a well-defined item. Back in the good old days we could have one good models where capital was corn that we had chosen to use as seed rather than eat. However, once we move into the real world, we have to recognize that what counts as a "capital" is a diverse array of items that includes not only physical goods, but also things likes patents.

There is a long literature on the problem of measuring capital. (The Cambridge capital controversy gives some of the flavor.) But, just to make a simple point, we might end up with considerably less "capital" if we shortened, weakened, or eliminated patent protection, especially in areas where it arguably is impeding technological progress (e.g. software and prescription drugs).

For this reason, the fact that we may appear to be seeing capital-biased technological progress should not be viewed as just some unfortunate event in the world that we have to learn to cope with. If we are in fact seeing capital-biased technological progress it is almost certainly the case that it is at least in part the result of policy decisions that could be handled differently.

Comments (20)Add Comment
written by Wayne Harris, December 26, 2012 10:58
Didn't Keynes theorize that increases in productivity would lead naturally to more leisure time for everyone? Especially in the U.S., it seems to have led instead to resource-depleting uber-consumption, concentration of wealth, corruption and debt peonage for almost everyone.
Further work in this area?
written by David M, December 26, 2012 11:04
Thanks for raising this critique. It seems like a fundamentally important point that is too often ignored--cf this quote from the Wikipedia page you linked to:
"In the United States, on the other hand, mainstream economics goes on as if the controversy had never occurred. Macroeconomics textbooks discuss 'capital' as if it were a well-defined concept — which it is not, except in a very special one-capital-good world (or under other unrealistically restrictive conditions). The problems of heterogeneous capital goods have also been ignored in the 'rational expectations revolution' and in virtually all econometric work." (Burmeister 2000)

Clearly, there are all sorts of ways that capital's value gets inflated at the expense of labor, as I've learned from your blog.

I'd love to know where to learn more about problems with the definition of capital and theories that try to account for the indeterminacy. Any recommendations for a well-informed- but lay-reader?
Patents are only a fraction.
written by LSTB, December 26, 2012 11:16
Nick Rowe was right that excluding land from the model clouds the discussion of capital-biased productivity. (http://worthwhile.typepad.com/...obots.html)

Rents from patents pale in comparison to rents from land ownership, and given that productivity increases inure to land values, we can tax the land rents without inhibiting progress by taxing capital, however one chooses to define it.
capitol based progress
written by Jennifer, December 26, 2012 11:30
This is an excellent companion piece to Krugman's post. People recognize that technology has made rapid and significant changes and generally see this is a good thing but also recognize the negative effects. It is great that so much information is on the web, not so great that many jobs--both high-paying and low paying--in the newspaper and magazine industry are gone. It is great that I can get a song for 99 cents in a minute but not so great that the record companies are losing tons of money and Tower records is gone. This is not to suggest that these were terrific examples of industry whose loss should be mourned but they did supply a lot of jobs and there is not much to take their place. I really think that when people are asked silly questions about the economy--deficit, "fiscal cliff", etc.--this is the issue actually causing people angst and concern for the future but it is hard to articulate. Whenever the issue is broached at all in the media it is ALWAYS presented as a choice between the future and the past, i.e. are you a Luddite or can you move on. The idea that you can have increases in productivy that are distributed fairly, among everybody and not just those directly connected to the "knowlege economy" is completely absent. Until you can recognize the bars you cannot understand you are in prison. People are catching on that there are limits to what education and creativity can get you, they recognize the system is rigged but I don't think they can see that the "leaders" of the "knowledge economy" and their enablers--the media, the congress, the educational-industrial complex--are a large part of the problem. Dean I know this is not new topic for you but I feel even in your writings there is a large emphasis on what the problem is and how it is harmful but not so much on a path to how to change it and what this might look like (this is not really meant as a criticism I tend to be very concrete in my thinking). Is anybody else looking at these issues in a serious and holistic kind of way? As a slight aside do you have any thoughts on the RSC copyright memo? It did not make it out of committee and I think the writer was fired but it was a sign of life on that front.
David Noble's work, is it of any relevance to this debate?
written by Alex, December 26, 2012 12:05
Where does the old MIT technological historian David Noble’s work on automation being used to enhance managerial control and de-skill workers fit into all of this stuff? Here is a sample of his work, taken from the detailed online footnotes to Noam Chomsky’s Understanding Power (http://www.understandingpower.com/chap7.htm; notes 51 & 52):

51. On the design of automation to enhance managerial control and de-skill workers, see for example, David F. Noble, Forces of Production: A Social History of Industrial Automation, New York: Knopf, 1984, especially chs. 5 to 9. This detailed study concludes (pp. 334, 338):

nvestigation of the actual design and use of capital-intensive, labor-saving, skill-reducing technology has begun to indicate that cost reduction was not a prime motivation, nor was it achieved. Rather than any such economic stimulus, the overriding impulse behind the development of the American system of manufacture was military; the principal promoter of the new methods was not the self-adjusting market but the extra-market U.S. Army Ordnance Department. . . . The drive to automate has been from its inception the drive to reduce dependence upon skilled labor, to deskill necessary labor and reduce rather than raise wages.

. . .

52. On managerial selection of automation not being dependent on its profitability, see for example, David F. Noble, Forces of Production: A Social History of Industrial Automation, New York: Knopf, 1984, chs. 7 and 9. See also, John A. Simpson [Director of Manufacturing Engineering at the National Bureau of Standards], "The Factory of the Future," talk at Rensselaer Polytechnic, June 3, 1982 [cited in David F. Noble, Progress Without People: In Defense of Luddism, Chicago: Charles H. Kerr, 1993, pp. 89, 97 n.38]. An excerpt:

In metalworking manufacture, direct labor amounts to roughly 10 percent of total cost, as compared to materials at 55 percent and overhead another 35 percent. Yet, as of 1982, management was expending roughly 75 percent of managerial and engineering effort on labor costs reduction and 10 percent on overhead cost reduction. This is a striking disparity.

. . .

Is the above stuff of any relevance to Krugman’s and Dean’s recent technology blog posts?
jennifer hit the nail on the head
written by mel in oregon, December 26, 2012 4:12
right-on, because you have the media, corporate america, the military, the wa-dc political power mad officials, & the education establishment, especially ivey league malarkey peddlers that are hugely responsible for the economic mess that to most people seems unsolvable. we are going down as a nation because you cannot bankrupt the majority of your citizens & expect demand to stay high. the power structure in this country is bankrupt morally, they have no character or any sense of decency, very similar to rome before it fell.
Capital effect historically disproven
written by Lrellok, December 26, 2012 8:33
Can we all please call him Prof. Luddite until Krugman gets over whatever impulse has seized him? In the 1950's through the 1970's, wages reflected around 51% of GDP. At this time, there where HUGE investments in infrastructure by the government, in addition to huge investments in capital equipment by most major industries. Is Prof Luddite attempting the assertion that public infrastructure investment is not labor saving capital investment? That the expansion in manufacturing and distribution in the 50's and 60' was not efficient?
In addition, if you take the data back to 1930, the end of his 2nd industrial age and the start of the great depression, wages are...wait for it...51% OF GDP! In fact, though out the whole of the great depression, wages never sank below 48.7% of GDP. Is Prof Luddite seriously going to attempt the assertion that between 1929 and 1969, NO aggregate laborsaving advancements of any form or kind occurred? Hydro electric power did nothing? Rural electrification zilch? Nothing from the interstate highways? Nada for automatic switchboards? Is he seriously, seriously going to assert that the advent of automated manufacturing between 1957 and 1964 produced NO increase in capital efficiency AT ALL? What is he smoking and where do I get some of it?

Even more strange is the 1930's themselves. If Gordon's hythoisis is credible, then in 1930 we should have been coming off of a massive reduction in wages. Yet wages are at 51% of GDP? Sadly, the data I can get does not go further back then 1929 (Grrrr) but I am fairly certain that had there been a wide scale collapse in the value of labor, wages would not be anywhere near half the economy.

Counter-pose that to today, where wages functionally reflect 44% of GDP. SO what happened to wages? From 1969 to 1993, wages as a portion of GDP slump from 52% to 46%. However, employment to population scream's from 60% to 66%. The idea that this is related to technology is wrong for two reasons. First, if labor is being replaced with capital, why is LF increasing in strong (p=-.837) inverse correlation with wages decline? If aggregate capital is displacing aggregate labor, LF should go down, not up. Second, most of this period is documented here http://www.cepr.net/images/sto...1-2012.jpg

As we can clearly see, the increase in equipment does not remotely justify the decrease in wages. The only credible increase is from '78 to '80, with a steady decline from '81 to '94. Wait, wait, look there, from '95 to '00 it INCREASES, at precisely the time wages are ALSO INCREASING.
The preponderance of evidence thus seems to show that not only does technological capital NOT crowd out wages, it in fact shifts wages from low skill to high skills sectors, and can have a net increase on wages as the cost of increasing numbers of higher paid workers do not necessarily equal the cost of firing lower paid workers.
Prof Luddite (Krugman) needs to stop making excuses for his refusal to acknowledge that in certain markets quantity determines price, not the other way around. Once this is accepted, the cuase for the decline in wages becomes clear, increasing downward pressure on wages either in the form of unemployment or in the form of increased labor supply, both resulting in an artificial surplus of labor quantity crippling the negotiating ability of workers for pay.
No capital-biased technological change in Europe or US from 1945 to 1975?
written by Indiana Way, December 27, 2012 12:26
What puzzles me about these discussions is that it strikes me as rather obvious that there was a considerable amount of capital biased technological development during the thirty glorious years: the Western Golden Age from 1945 to the early 1970's. Why did wages rise in this period?

Another thing: can't capital biased technological change be thought of as making labor relatively more scarce? In what sense is making a labor saving widget that saves twice as much labor not roughly equivalent to acquiring two labor saving widgets? Capital accumulation in East Asia from the 1960's to today (the Eastern Golden Age) has been accompanied by dramatic wage increases? If 4 million Singaporeans make on average 50% more than Americans, why doesn't everyone just acknowledge that there is considerable room for US wages to grow before we converge on the world's economic leaders?
written by skeptonomist, December 27, 2012 8:30
I previously scrounged up wage data (mostly from the Census) going back to 1860 - here is a log plot comparing real wages with GDP:


This is for the BLS category "non-supervisory production workers" or equivalent. As Lrellok points out, there were many stages of technological change from the Civil War to the present. There was clearly a huge change in wage growth around 1970 which is not obviously related to any technological change. What happened at that time? Economists need to face up to this transformational stage in economic history.

The current attempt to blame recent technological change for unemployment and low wages seems like a replay of the "structural unemployment" argument. It is very strange that Krugman appears to lend credibility to this argument.
Is Prodictivity Growth Over? I think not.
written by A Populist, December 27, 2012 9:18

Paul Krugman asks the question: Is growth over?

Of course, the important question is: What will be our future Productivity growth?

It seems to me, that having low wages is likely holding back Productivity growth, as it reduces incentives to save Labor cost.

Has this been investigated in terms of predictions for future Productivity growth?
Progress on Copyright/Patents
written by Dean, December 27, 2012 9:20

I was delighted to see the RSC memo suggesting much more limited copyright. It was great to see some serious thought on this issue. (It would be nice to see some on the Democratic side.) Of course the Republican leadership quickly censored the piece and fired the author.

In terms of progress, our system is so awful, almost anything else would be better. My hope is that some developing countries adopt alternatives. They will quickly shoot past us both in the pace of innovation and in the savings on the price of copyright/patent protected items.

U.S. trade policy is aimed largely at preventing this possibility, ostensibly while promoting "free trade." There is much at issue in blocking or delaying the new round of trade deals. If countries have the opportunity to experiment with superior paths, our system will be killed.
The Rise in CEO Pay
written by J, December 27, 2012 10:00
one clue as to what is going on is the rise in CEO pay from 30x the average worker's salary in 1970 to about 250x today. If technological progress is hurting labor, it's very clear that it is not hurting those at the top. Corporate profits as a percent of GDP are 70% higher than their postwar average and the biggest contributor is lower wages but not for everyone. The best explanation can be found in Dean Baker's book An End to Loser Liberalism - there was a redistribution of wealth upward, by design.

Baby with bath water
written by A Populist, December 27, 2012 10:32

Regarding patents and copyrights:

I would concede that there are some anti-competitive effects from patents, and I think that having copyrights last for generations is just bizarre - and totally unnecessary.

However, I would make a couple points:

First, extrapolations of increased medical costs due to patents on drugs and medical devices could be overblown, as drugs and devices come off patent. If future advances are significant enough to be profitable, then the inventors deserve to be rewarded.

Second, it seems to me that you may be under-weighting the value of rewarding inventors, and the companies who cultivate and promote innovation. If everyone just waits for some sucker to do the innovation, and then just pounces with a cheaper me-too product - then innovation simply may not happen, as those saddled with the costs of innovating, lose money. I would argue that his could actually exacerbate the downward spiral of wages, as cost and price dominate in a world-wide race to the bottom.

Also, politics is not limited to government. The fact that an inventor must be explicitly listed on a patent, provides some measure of protection for inventors within companies. The pressure for cost-reduction within firms is very strong, and without the (relatively) objective metric of having your name on a patent, it is quite easy for inventors to lose their job - and the ability to invent.

Invention requires more than just an inventor. Such inventors need to have an environment which is conducive, such as being aware of customer needs, knowledge of manufacturing processes, etc. it is easy to assume that if an inventor is fired, that he will just invent somewhere else. Don't discount the possibility that a fired inventor ends up working at Starbucks, unable to use his talent.

While I think that there is always room for improvement, patents are a valuable incentive, and serve to protect and reward inventors, and the companies that foster innovation.

Patent myths
written by David, December 27, 2012 7:33
A Populist seems to have forgotten:

Ben Franklin filed no patents and died a wealthy man.

Thomas Jefferson, one if the first US patent examiners, concluded that the patent system, cloned from the royalists, is a political embarassment to the US, re freedoms.

Patent trolls and large corporate patent libraries undoubtably stifle and block promising innovations, not promote innovation or adoption of efficient/efficient innovations.

Government granted/protected monopolies cause distortions in the markets
patents and innovation
written by pjm, December 27, 2012 11:51
2 cents.
1) It is pretty obvious that intellectual property slows down research that happens in an academic environment (though I am not a huge fan of TED talks, there is a good one on "open-source" drug development).
2) The academic psychological literature on the relation between incentives and creativity, I have heard, *strongly* indicates that incentives have at best a neutral and possibly counter-productive effect. You can incentivize brain-dead labor but not creative. Supposedly the literature on this is pretty large.
So yeah, dismantle the patent system.
written by A Populist, December 28, 2012 3:21
.Re: "You can incentivize brain-dead labor but not creative."

I agree with that 100%. Well, maybe 90%. But, yes - The highest performing innovators are motivated primarily by non-monetary motives.

But invention does not happen in a vacuum.

It requires companies to retain innovative people, and give them the resources necessary to innovate. If there is no payback to companies for these expenditures, the companies will not bother. Have you ever asked the finance department for money to spend on tooling, prototypes, man-hours, months of testing - which may or may not pay off? The first thing they want to know is: How is this going to be paid back? Why should those in charge take a chance on losing money, if a competitor can quickly copy it? And if those knock-offs are cheaper (race to the bottom), even a fabulous idea could easily lose money.

Rewarding innovative individuals and companies allows those who are the best innovators, to accrue more resources - and innovate even more.

This is the same principle with capitalism in general - those who excel are rewarded and thrive, while those who are inept or corrupt - fail or die.

What mechanism do you have to cull non-productive innovators in your scheme? What objective criteria do you have to prevent cronyism?

Apparently you guys think the free market and capitalism are total BS, and pure socialism is the way to go. If it works for innovation - Hey, why not the whole economy?

I think capitalism and socialism are both flawed, but that capitalism with some band-aids (safety net, minimum wage, etc), is probably the optimum system.

I think patents are another useful band-aid - which allows those who innovate to recoup some of their investment.

I agree that there are problems - and I think the transition from "first-to-invent" to "first-to-file" (America Invents Act) was really a bad thing - both for small inventors, and for the US in particular (We had the advantage that we did not have to "race to the patent office"). So, yes, the largest companies have sold out some big advantages. And the lawyers win again.

But if your only position is "patents need to be eliminated", I guess I have to disagree.

I have seen groupthink before.

Go ahead and hit the "minus" button. Enjoy yourselves.
written by David, December 28, 2012 7:41
The groupthink is found in the massive thoughtless support of the patent system. Jefferson makes the argument that it's unnecessary, based on evidence. If "Populist" wishes to abandon rationalism and mathematical systems analysis for ad hoc arguments, then he has no place to accuse others of groupthink.
written by A Populist, December 28, 2012 10:08
Well, the Wall Street banks hired a bunch of PhD's with "mathematical analysis" proving that their derivative concoctions would all work out just fine, while ad hoc, concrete examples of analogous systems made it obvious that they were either wrong or dishonest - and made key assumptions which were fatal. So, pardon me, if I am not impressed.

Is your reference to "rationalism" meant to imply that I am not rational?

That is not an argument.

Look, I respect Dean more than any economist out there - period.

So, I am actually inclined to listen to his arguments, and even consider changing my position. If you can't sell me on your argument, you are certainly not going to sell it to the people who are inclined to dismiss you outright.

Dismissing actual real-world examples which illustrate flaws in theoretical models as "ad hoc", is not the best way to avoid groupthink - or just bad ideas that may not work, or need improvement.

If you have an argument to make, and a detailed alternative proposal - make it. Or give links, references, or whatever.

Nothing wrong with both of us being provocative - but I would actually like to know more about your arguments. What I have seen so far, doesn't seem persuasive to me. And like I said - If you can't convince someone who is actually willing to listen - good luck with everyone else.
The case against patents
written by David, December 29, 2012 7:47
The case against patents can be summarized briefly: there is no empirical evidence that they serve to increase innovation and productivity. There is strong evidence, instead, that patents have many negative consequences.

[url= http://papers.ssrn.com/sol3/pa...id=2148738

Me, I'm on the (snowy) road.
Boldrin and Levine link
written by David, December 30, 2012 12:52

Empirically, the patent system does not foster innovation nor protects the independent inventor.

Write comment

(Only one link allowed per comment)

This content has been locked. You can no longer post any comments.


Support this blog, donate
Combined Federal Campaign #79613

About Beat the Press

Dean Baker is co-director of the Center for Economic and Policy Research in Washington, D.C. He is the author of several books, his latest being The End of Loser Liberalism: Making Markets Progressive. Read more about Dean.