The evolution of the rate of surplus value
Kliman’s discussion of the evolution of the rate of surplus value over the last 40 years is, in my opinion, the weakest part of his book. Most Marxists—and non-Marxists, including the great bulk of U.S. workers—would agree that the portion of income going to the rich—the capitalist class—has risen considerably in the U.S. since the early 1970s. This widespread popular belief is clearly reflected in the rise of the Occupy movement.
Kliman strongly disagrees with this. Using U.S. government statistics, he attempts to demonstrate that the share of the U.S. national income going to the workers has risen at the expense of the share going to the capitalists. Or in Marxist terms, the rate of surplus value has actually fallen. A falling rate of surplus value, even if the organic composition of capital remains unchanged, implies a fall in the rate of profit. If a fall in the rate of surplus value is accompanied by a rise in the organic composition of capital, the result will be a marked fall in the general rate of profit.
Which is right: the general popular perception and the view of the Occupy movement that American capitalism and world capitalism is growing more exploitative, or Kliman’s contrary view?
Kliman quotes John Bellamy Foster and Fred Magdoff—leaders of the Monthly Review school: “…wages of private non-agricultural workers in the United States (in 1982 dollars) peaked in 1972 at $8.99 per hour, and by 2006 had fallen to $8.24 (equivalent to the real hourly wage rate in 1967), despite the enormous growth in productivity and profits over the past few decades.” (p. 155)
These figures would seem to clinch the case for a considerable rise in the rate of surplus value in the decades preceding the “Great Recession.” It would seem that on the eve of the Great Recession in 2006, a typical U.S. worker got less in use value terms for each hour of labor power she sold to the capitalists than her mother earned for similar work 34 years earlier. Furthermore, the productivity of human labor has hardly stood still over the last 34 years. This means that the commodities that a worker consumed in 2006 embodied a considerably smaller amount of human labor value than was the case in 1972.
This is true for two reasons. First, the worker in 2006 received less use value for every hour of labor power she sold to the capitalists. Second, each unit of use value she did receive in exchange for her sold labor power represented less embodied abstract human labor—value—than it did in 1972.
This would mean that there has been a marked growth in what Marx called relative surplus value when if the total work day remains unchanged workers will be working a smaller amount of time for themselves and a greater amount of time for the capitalists. This can be the case even if the standard of living of the workers actually increases, if the increased number or quantity of commodities the workers get to consume in exchange for their sold labor power represents a smaller quantity of value.
Kliman disagrees. He thinks that if anything the rate of surplus value, at least in the U.S., has fallen over the last 40 years. In attempting to prove this, he quotes economist Martin Feldstein as an authority. Feldstein wrote that it is a “measurement mistake” to “focus on wages rather than total compensation.” Feldstein complains that this has “led some analysts to conclude that the rise in labor income has not kept up with the growth in productivity.” (p. 153)
Kliman doesn’t inform his readers that Martin Feldstein is an extremely reactionary economist who has dedicated his life to defending and prettifying U.S. capitalism, though he does mention that he was the head of the National Bureau for Economic Research.
Marxists, beginning with Marx, have often quoted bourgeois economists when these economists’ research exposes some of the truths about capitalism and its exploitation of the workers. When the hired apologists for capitalism are obliged to admit a portion of the truth about the exploitative nature of capitalism, it is especially telling. The more reactionary the particular apologetic economist is the better.
But for a Marxist to quote reactionary economists when they use statistical data in a way that actually strengthens their apologetic views of capitalism is rather unusual, to say the least. While we can’t prove that American capitalism has grown more exploitative simply because Feldstein claims it hasn’t, Kliman’s conclusion is strongly in line with Feldstein’s natural ideological bias.
A Critique of Crisis Theory 
1. The foremost problem with this critique of my discussions of income distribution is that it operates with too few distinctions and concepts. It therefore mixes and matches different things, and ends up attributing clams to me that I don’t make.
There’s no single thing called “income.” A whole lot depends on what a particular income measure includes and excludes. There’s no unambiguously correct definition; it all depends on the question being addressed.
2. “Most Marxists—and non-Marxists, including the great bulk of U.S. workers—would agree that the portion of income going to the rich—the capitalist class—has risen considerably in the U.S. since the early 1970s. … Kliman strongly disagrees with this.”
I do *not* disagree that the portion of income going to the rich has risen considerably. On pp. 159-60, I report three different measures of the change in inflation-adjusted income between 1979 and 2007, and I note that “All three definitions of income lead to the conclusion that income inequality increased” (e.g., the income of the top 20% grew more than any other group).
3. “Using U.S. government statistics, he attempts to demonstrate that the share of the U.S. national income going to the workers has risen at the expense of the share going to the capitalists. Or in Marxist terms, the rate of surplus value has actually fallen.”
I do *not* attempt to demonstrate that. This misunderstanding is a result of the lack of distinctions and concepts I noted above.
[What I do demonstrate is that these statistics indicate that
(1) the share of corporations’ output (net value added) that paid out as compensation of employees was trendless after 1970, and it was lower in the early postwar period.
(2) The available data suggest that the hourly compensation received by people in “management, business, and financial operations” occupations did not increase much faster than hourly compensation of other employees between the end of 1985 (when the data set begins) and the end of 2007. This suggests that these other employees’ share of corporate output declined only slightly.
(3) Expressed as a percentage of national income, the income of the working class–measured as the sum of (a) wages and salaries, (b) nonwage compensation (retirement and health benefits) paid by employers, and (c) and receipt of government social benefits (minus tax contributions that partly pay for them) such as Social Security, Medicare, Medicaid, and unemployent insurance– rose substantially between 1960 and 1970 and was basically flat from 1970 to 2007.
(4) Inflation-adjusted hourly compensation of all workers, and of “regular” (production and nonsupervisory) workers, both increased substantially between 1980 and 2009, by anywhere from 25% to 37%, depending on the group of workers and the inflation measure.
(5) Inflation-adjusted hourly wages and salaries rose between 1981, when Reagan took office, and 2009.
Note that in (1), I wrote “corporations’ output,” not “national income,” and that I said nothing about “the portion of income going to the rich”; and that in (2), I said nothing about “the share going to the capitalists.”
Also note that these findings are based on my analysis of the government data. Martin Feldstein didn’t make the data up, and I drew my conclusions from the data, not from him. ]
4. “Which is right: the general popular perception and the view of the Occupy movement that American capitalism and world capitalism is growing more exploitative, or Kliman’s contrary view?”
I draw *no* conclusions about whether “world capitalism is growing more exploitative.” I don’t even discuss it.
5. Sam reports the fact that wages and salaries, adjusted for inflation using the CPI-W price index, fell between 1972 and 2006, He then writes, “These figures would seem to clinch the case for a considerable rise in the rate of surplus value in the decades preceding the ‘Great Recession.’ It would seem that on the eve of the Great Recession in 2006, a typical U.S. worker got less in use value terms for each hour of labor power she sold to the capitalists than her mother earned for similar work 34 years earlier.”
In my book, I avoided discussion of Marxian categories like “rate of surplus-value” because I was using data based on different categories, those of U.S. government. But Sam’s interpretation of this category is really something. He’s suggesting (while hedging slightly) that only wages and salaries count as value received by workers. Employers’ (and governments’) payments of retirement benefits and health benefits to retired workers––“deferred compensation”—are instead supposedly SURPLUS-VALUE, merely because the workers didn’t receive them on the spot.
Excuse me? Retired workers are now part of the capitalist class? The typical U.S. worker is now more “exploited” because her mother is living longer and better? This must be in the missing chapter of Capital in which Marx talks about capital’s werewolf hunger for Social Security and Medicare benefits.
What he writes in chapter 6 of volume 1 is rather different: “Whether money serves as a means of purchase or as a means of [deferred] payment, this does not alter the nature of the exchange of commodities. The price of labour-power is fixed by the contract, although it is not realized till later …. The labour-power is sold, although it is paid for only at a later period” (p. 279, Penguin ed.).
Thus the price of labor-power is the amount received, irrespective of when it’s received. As I note in my book (note 2 to chap. 8), regarding Richard Wolff’s notion that that recipients of retirement benefits and Medicare benefits are not workers, but *former* workers, “This is like saying that recipients of cash wages are not workers because they receive their paychecks after the workweek is over. (In both cases, the recipients receive income in exchange for going to work, but only after they are finished working.) It is also like saying that the unemployed are not workers, but former workers, and that people who work eight hours a day, five days a week are only part-time workers because they don’t work 24/7.”
In 1959, 35% of people 65 and older in the U.S. lived below the official poverty line. In 1970, 19% were still below the line. In 2009 and 2010, the percentage was less than 8%. And whereas their poverty rate was far higher than the national average in 1959 and 1970 (22%, 13%), it was well below average in 2009 and 2010 (14%, 15%). And while inflation-adjusted median household income rose by an average of 21% between 1967 and 2010, it rose almost 5 times as fast—by 99%—among households headed by people 65 and older.
These facts, which have everything to do with rising government and private retirement and health benefits—plus the fact that the share of the population that’s 65 and older rose by 45% between 1859 and 2010—are the secret behind the decline in the wage-and-salary share in the narrow sense. The reason why the average worker isn’t receiving much more *current, take-home* pay than his/her parents received is that the parents are much less likely to be dead or poor than their parents were.
How can Sam possibly suggest (while hedging slightly) that retired workers are capitalists, or at least that retired workers’ income is capitalist income (surplus-value)? The answer, once again, is that he operates with too few distinctions and concepts. Workers receive wages and salaries; capitalists receive surplus-value. Certain kinds of income aren’t wages and salaries. “Therefore” they are surplus-value.
It’s now past 2 a.m. These comments haven’t even gotten through his 5th paragraph. And this is becoming too long for a blog comment. So I’ll have to stop here and resume with additional comments where I left off.
The “failure of capitalist production.” Kliman has it upside down. The failure is in capitalist distribution, or as Marx said, the mode of production is being outstripped by the means of production. Although capitalism can produce billions of commodities, there are fewer and fewer people who can buy them, thus the production has to be shut down. GM, for instance, could produce 5 times as many cars as it does, but there is no “market” for the cars. Not only that, but GM would not even exist now as a car manufacturer except for the massive transfer of wealth from the working class to GM in the bailout.
Kliman should stop trying to tinker with government statistics and demand the revolutionary takeover of GM production by the auto workers.
I wanted to add that, in terms of Marx’s dialectical analysis of capitalist production, then Kliman may be more correct. Marx pointed out in the Grundrisse that production is a unity made up of different parts or “moments;” that production, distribution, exchange and consumption are all contained within the same whole, production, and all these moments have a cause and effect determination on each other. How a society consumes is determined by how a society produces, and so on.
In those terms, I think, Kliman is correct: capitalism is a failure.
See my comment of April 18, 2012 at 6:48 am for my first 5 points in response to Part 3. Continuing:
6. “… the worker in 2006 received less use value for every hour of labor power she sold to the capitalists.”
This conclusion depends on Foster and Magdoff’s exclusion of large and growing chunks of the compensation workers receive (see my point 5 above). It also depends on Foster and Magdoff’s use of an inconsistent series, the CPI-W price index, when adjusting for inflation. When the personal consumption expeneditures price index, a consistent series, is used instead, the hourly real wages and salaries of production and nonsupervisory workers in 2009 were 12% higher than they were in 1972, and about 18% higher than they were in 1970.
7. “Kliman … thinks that if anything the rate of surplus value, at least in the U.S., has fallen over the last 40 years. In attempting to prove this, he quotes economist Martin Feldstein as an authority. Feldstein wrote that it is a ‘measurement mistake’ to ‘focus on wages rather than total compensation.” Feldstein complains that this has ‘led some analysts to conclude that the rise in labor income has not kept up with the growth in productivity.’ (p. 153)
“Kliman doesn’t inform his readers that Martin Feldstein is an extremely reactionary economist who has dedicated his life to defending and prettifying U.S. capitalism, though he does mention that he was the head of the National Bureau for Economic Research.”
As I said above (point 5), I tried to avoid linking U.S. government categories to Marx’s categories. But it is indeed the case that the share of corporations’ net value added received as compensation by their employees has been trendless since 1970. And the available data suggest that it fell only slightly if we exclude people working in management, business and financial occupations (point 3).
That the compensation share and the profit share have been basically constant is a well known and accepted fact. For instance, in The Crisis of Neoliberalism, Dumenil and Levy have a graph very similar to mine. And Piketty and Saez, who have become famous for their studies arguing that income inequality has increased markedly, also have a similar graph in a 2003 paper, and they write, “As is well-known, factor shares in the corporate sector have been fairly flat in the long run with the labor share around 70–75 percent, and the capital share around 25–30 percent.” I assume that Sam won’t try to dismiss the Dumenil, Levy, Piketty, and Saez as extremely reactionary economists who have dedicated their lives to defending and prettifying U.S. capitalism.
As for Feldstein’s contention that the rise in labor income *has* kept up with the growth in labor productivity, this follows directly and necessarily (at least for the corporate sector) from the finding that the compensation share of corporate net value added has been basically constant. It’s simple math. Productivity is output (net value added) per labor-hr. “Labor income” is compensation per labor-hr. Since output and compensation have grown at the same rate, so have output per labor-hr and compensation per labor-hr. (BTW, Dumenil and Levy explicitly accept this, too. I’m sure that Piketty and Saez do as well.)
As for the point that it’s a measurement mistake to focus on wages and salaries, while ignoring nonwage compensation, this is *obvious*. I didn’t need Feldstein to tell me this. (When I first made this point, I hadn’t read his paper, so it’s an independent conclusion, not an appeal to authority; see my article at http://www.marxisthumanistinitiative.org/economic-crisis/lies-damned-lies-and-underconsumptionist-statistics.html.)
As I note in my book (and the article), the nonwage compensation data come from the very same table that Foster and Magdoff used to get their wage and salary data. Their graph and discussion certainly give us the impression that when they write “wages and salaries,” they mean total income from working or total compensation, but the table that their data come from doesn’t use “wages and salaries” to mean that. The main category is “Compensation of employees, received,” and it consists of two subcategories, “Wage and salary Disbursements” and “Supplements to wages and salaries,” the latter of which is broken down into “Employer contributions for employee pension and insurance funds” and “Employer contributions for government social insurance.” (Note the term “Compensation of employees, received”; the U.S. government agrees with Marx (see point 5) that nonwage compensation is compensation that’s been received, even though it was received with a delay.) When I looked at Foster and Magdoff’s graph, and then saw this table that they cited as a source, I was frankly shocked at what they had done.
8. “Marxists, beginning with Marx, have often quoted bourgeois economists when these economists’ research exposes some of the truths about capitalism and its exploitation of the workers.”
That’s what I did.
9. “But for a Marxist to quote reactionary economists when they use statistical data in a way that actually strengthens their apologetic views of capitalism is rather unusual, to say the least.”
I think that knowledge and expertise count for a lot, and that Foster and Magdoff are nowhere in the same league as Feldstein in that regard. I also think they have a rather apologetic view of capitalism––underconsumptionism––with which their exclusion of nonwage compensation is consonant. I’m surprised that Sam seems not to see this, given that he has quoted Sweezy’s claim that “[t]he second indispensable change needed to make the private-enterprise economy work better is a redistribution of wealth and income toward greater equality” (see his Part 1). And whatever Feldstein’s intentions may have been, my use of the compensation data certainly does not strengthen any apologetics for capitalism. It strengthens the case that the Great Recession was rooted in problems in the capitalist mode of production, not in the realm of distribution.
10. “Even if it were true that ‘non-wage compensation’—such as health insurance, for example—has increased so much since 1972 that real income—hourly wages plus non-wage compensation—has risen for each hour of labor power that U.S. workers sold to the capitalists between 1972 and 2006, it would still remain to be shown that U.S. workers are receiving more value for each hour of labor they perform. Remember, as long as the productivity of labor is growing it is quite possible for the standard of living of workers to rise while they are more exploited than ever. This is what Marx’s concept of relative surplus value is all about.”
The phrase “even if it were true” makes it seem as if my conclusions about trends in total compensation are based on Feldstein’s say-so. That’s not the case. They come straight from the data.
Yes, in principle “it is quite possible for the standard of living of workers to rise while they are more exploited than ever.” But this has NOT been the case in the U.S. corporate sector, not if “more exploited” means that the value received as compensation has fallen as a share of value added. This does not “remain to be shown.” What shows it is the finding that the compensation share of net value added has basically been constant (see points 5 and 7). The compensation share is compensation divided by net value added. If that’s constant, then the ratio of compensation per hour to net value added per hour is also constant—i.e., compensation per hour worked is a constant share of value added per hour worked.
Now, the compensation and net value added figures are in dollar terms. But this makes NO difference. If you turn them into figures in terms of gold, or in terms of labor-hours, you multiply the numerator (compensation) and the denominator (net value added) by the same number, and you end up with the exact same ratio.
11. “Casting further doubt on Kliman’s claim that the rate of surplus value of U.S. workers did not rise in the decades preceding the Great Recession is the fact that the last 40 years have seen a tremendous weakening of the U.S. union movement. ‘Union membership as of 2010 in the private sector,’ states Wikipedia, ‘has fallen under 7%—levels not seen since 1932’—that is, since the days of Herbert Hoover when union membership was greatly depressed during the super-crisis phase of the Great Depression. It would indeed be remarkable if the rate of surplus value extracted from American workers had actually declined despite this huge weakening of the union movement, combined with the increase in real unemployment, only partially reflected in the official jobless numbers, that has occurred since the postwar economic prosperity ended 40 years ago.”
In the private sector (my figure for the compensation share refers to corporations, and thus to the private sector), the relative decline in union membership began long before the last 40 years. The percentage of private-sector nonagricultural workers who were in unions fell from 38% between 1952 to 31% in 1970 (see Richard B. Freeman, “Contraction and Expansion: The Divergence of Private Sector and Public Sector Unionism in the United States,” Journal of Economic Perspectives 2:2, Spring 1988, Exhibit 1).
This was a period of rapid wage growth in the private sector, the “golden age of capitalism.” So trends in union membership rate are not good evidence of trends in wages, much less trends in the rate of surplus-value. No doubt has been cast on my findings. (And again, I don’t claim that the rate of surplus-value has fallen during the last 40 years (see point 3 above).)
12. “Even if we were to accept Kliman’s attempt to demonstrate a fall in the rate of surplus value in the United States since 1972, his problems wouldn’t end there. Kliman leaves out … the shift of capitalist production from … countries …where wages are relatively high, to countries …where wages are dramatically lower. …
“When this is taken into account, it is hard to draw any other conclusion than that on a global basis—and this is what counts for determining the average rate of profit, especially in today’s globalized world—the rate of surplus value has risen dramatically.”
This complaint seems to be based on the false belief that I said anything about the rate of surplus-value in the world as a whole (see point 4 above).
13. “Why does a radical socialist economist like Andrew Kliman attempt to prove the opposite, relying on U.S. government statistics—hardly an unbiased source—and the reactionary Martin Feldstein to make his case?”
The stuff about U.S. government statistics is rather strange. It doesn’t help Sam’s case in the least. If he wants to challenge these statistics on grounds of bias, he also—to be consistent––has to challenge the statistics that Foster and Magdoff cited, which come from the same biased U.S. government. And he has to challenge Piketty and Saez’s and everyone else’s findings that income inequality in the U.S. has increased; all of the data on this come from the U.S. government.
Why does Sam Williams rely on statistics reported in the underconsumptionist Monthly Review?
Why does he say that statistics published by the biased U.S. government “seem to clinch the case for a considerable rise in the rate of surplus value”?
In point 11, I meant to write: “The percentage of private-sector nonagricultural workers who were in unions fell from 38% between 1952 AND 1955 to 31% in 1970
Sam,
I hope to see a response to Klinman here. I think he hits you on two fronts
1. because it a blog post, AK faults you for playing too fast and loose with terms, categories etc which may have lead to misattributions and so forth.
2. BIG DEAL. He is still beating the falling rate of profit drums and addresses NONE of the haymakers you’ve leveled at him thus far. Hit him some more buddy.
I even wonder if the question of the rate of surplus value might be a sidetrack. I don’t think you need it to decimate the “failure of capitalist production” and i think its cheapens your overall case if the research/data is so thin or scanty that you have to appeal to popular perception
@KOBH: “He is still beating the falling rate of profit drums and addresses NONE of the haymakers you’ve leveled at him thus far.”
If you point them out to me, and if I understand them, I’ll certainly do so.
Isn’t it time to replace the phrase “surplus value” with profit?
@allanrharris: I didn’t say ““… the worker in 2006 received less use value for every hour of labor power she sold to the capitalists.” I quoted it. Note the quotation marks.
In fact, I think that the conflation and mixing-and-matching of Marx’s concepts with concepts employed in the statistics and elsewhere has detracted from the clarity and intelligibility of what’s been written.
You’re right professor. I apologize.
No problem. I was just trying to clarify things, not looking for an apology.
Kliman says, “… the worker in 2006 received less use value for every hour of labor power she sold to the capitalists.”
Why can’t this sentence be put in plain English? “The employee in 2006 received less in wages than what she produced,” which is the fundamental basis for capitalism.
Besides, in Marxist terms, a worker “receives” the exchange value for the sale of labor power. The capitalist buys and receives the use-value of labor and the worker then produces, for free, surplus-value for the capitalist. The worker then takes the value received in wages to Walmart and buys a commodity. The value, or price, of the commodity received is equal to the value of the wages. The trick for the capitalist is that she did not pay the full price for the production of the commodity. The problem for the capitalist is that after a while workers don’t have enough in wages to purchase what the capitalist has to sell, then there is another crash.
Phrases like exchange-value, use-value, surplus-value, labor-power, and even the word value, have long lost any real connection to the concepts they were designed to describe.
Price, utility, profit and labor now adequately express the same ideas and can be understood by a layperson. Even the word labor is being replaced by “human capital,” a particularly ugly phrase. However, if they want to use that phrase then socialists should appropriate it and demand that human capitalists have the same power and rights over their own capital as the inhuman capitalists have over theirs.
Marx said in Grundrisse that he assumed and accepted the categories of bourgeois economics and then took them to their logical conclusion, as in the equal exchange of commodities. And from Smith and Ricardo he fully developed the labor theory of value.
For instance. Keynes discussed the diminishing marginal utility of profit. Keynes did his best to ignore Marx, but he accepted Marshall’s (Marshall managed to completely ignore Marx, however, he did use the phrase “the German writers”) idea of marginal utility. Now, the diminishing marginal utility of profit is nothing more than the tendency of the rate of profit to decline.
Why can’t Marxist economists accept at face “value” the categories of the Wall Street (“bourgeois” is also a word that is outdated) economists and apply them to the real world of the workers’ fight (not struggle, you struggle with your conscience, homework, etc.) against capital?
We’re not really discussing the key theoretical argument whether there is a realisation problem at the heart of the capitalist system. A realisation problem based upon commodity money & where credit money allows overproduction of commodities relative to the money commodity to occur. In otherwords, exchange-value diverges from value. The law of value doesn’t permit this divergence to continue indefinitely, credit crunch leads to recession & we have a theoretical explanation of cyclical crises.
Hi dr,
Credit-driven booms and busts, and increases in nominal values relative to real values, occur even when there is gold money.
Marx, who lived when there was gold money (sort of), called the credit system “the principal lever of overproduction and excessive speculation in commerce” and noted that “the reproduction process, which is elastic by nature, is now forced to its extreme limits” by the credit system (Capital, vol. 3, p. 572 of Penguin ed.). I discuss his view of the role of credit in crises, and the relevance of it to the financial crisis of 2007-08, on p. 14 and pp. 19-22 of my book.
So gold money is a fifth wheel. It isn’t needed in the explanation and doesn’t add anything to it. That’s because gold money doesn’t prevent increases in nominal values relative to real values.
This doesn’t mean that inconvertibility is unimportant. It may influence the length of time that credit-driven booms and bubbles can last, and the extent to which nominal values can diverge from real values, and thus the magnitude of the “corrections” that follow.
What you call a “realisation problem,” which is not the way the term is usually used, is really just a problem of uncollectable debt. Again, gold money is a fifth wheel here.
Note to Sam: yes, I quoted the word “overproduction” again. But I’m happy to use it myself: overproduction, overproduction, overproduction. It’s just that I then have to explain that overproduction is a phenomenon, not an explanation of anything, and that there’s no chronic overproduction problem resulting from the fact that investment, and thus productive capacity, grow faster than personal consumption. It’s easier to avoid using terminology that Marx used that is typically misunderstood and harnessed to alien theories.
Yes, the gold standard still has credit money & recessions.
Under a fiat money regime the overproduction can be much greater.
Big crash to come along with currency debasement?