The rate of profit: the devil in the detail

Andrew Kliman’s (AK) new book (see my post, Andrew Kliman and The Failure of Capitalist Production, 8 December 2011) has provoked a very negative review from Bill Jefferies (BJ).  BJ has posted, as a comment to my post on AK’s book, his review.  And you can read it here  Comments have been coming into my blog from both on some of the key issues of debate.

BJ reckons that AK is a “stagnationist” because he claims that the rate of profit in the US has persistently fallen without significant rises since the second world war, when it clearly has risen since the early 1980s.   This means, according to BJ, that AK shows that the countervailing tendencies to Marx’s law of profitability ‘as such’ have not operated.  The countervailing factors would be: a rising rate of surplus value, the cheapening of constant capital, the expansion into overseas markets for higher profit etc.  BJ says that this cannot be right: wages have declined as a share of national income in the US; there was a technological revolution that cheapened the cost of investment hugely; and overseas profit rose, especially when the ex-Communist states collapsed and globalisation led to several emerging capitalist economies (BJ includes China here) became a powerful force.  Indeed, BJ appears to argue that rates of profit in most capitalist countries, especially the newly emerging ones, were rising from 1982 onwards in contrast to AK’s claims.

But the “nub of the issue” for BJ is that AK measures profits as “property income” which includes all surplus value created by corporations, namely interest, taxes and transfer payments.  BJ reckons that doing so distorts the measure of the rate of profit.  He reckons you should exclude all these parts of surplus value that do not accrue to corporations, particularly taxes that end up in government hands are used to boost the ‘social wage’ of workers.  If you take out this “government income” then the US corporate profit rate, even using AK’s method of valuing fixed capital, is much less dramatic.

Is BJ right to suggest this?  Well, you can measure profit in many ways and each way may be more useful for certain purposes.  In his book, AK discusses the merits of different ways (whole economy, business sector only, corporate sector only, non-financial sector only; or all surplus-value, before tax profit only; after-tax profit only etc).  AK uses what he call ‘property income’ because he wants to show the class-based nature of the capitalist system and that means including all the surplus value created by the workforce, even if much of it is then redistributed to banks (interest); landlords (rent) or to government (taxes).  I agree that this is the best measure of profit to understand the laws of motion of capitalism.  Indeed, I have tried to measure profits in the whole economy, and not just the corporate sector as AK does, in order to capture the whole process.  But if you want to know how profit drives new productive investment in the economy, a better indicator of profit might be after-tax profit.

The after-tax profit may show a much less dramatic fall in the rate of profit – but it still shows a fall.  BJ does not show this in his review, but I worked out the after-tax profitability based on current costs (BJ’s preferred measure) and there is still a trend decline from 1950, although from 1965 at the end of the ‘golden age’ BJ’s after-tax profitability is basically trendless.

After-tax profitability is trendless from 1965 to 2009, but the reason for this is down to how the denominator in the rate of profit is measured.  Remember that the rate of profit is measured as the mass of profit divided by the value of the stock of assets (namely the value of plant, machinery and other technology plus stocks of raw materials and other components).  You can also add the cost of employees to that denominator if you wish.   As any reader of AK’s material knows, he is vehement that, to measure profitability in any meaningful sense, capital stock must be valued in historic terms.  That measures what capitalists paid for it before production starts and not what it could pay for at the end of production.

BJ denies this is right and quotes Marx to justify his view that Marx would have valued capital stock at its current cost.  BJ argues that “Kliman is wrong to assert that the mass of the fixed capital stock must always be valued at its purchase price  .. the effect of depreciation shows that technological progress reduces the value of the fixed capital stock from its purchase price to its current price.  If Kliman were correct, capitalist crises would be impossible as the wholesale devaluation of the fixed capital stock could not take place”.

But is BJ right?  Using historic costs as the measure of fixed capital does not exclude a devaluation of these costs in a crisis – usually that would happen when capital is liquidated in bankruptcies or even physically in war.  At each new period of production, the historic cost of fixed assets would incorporate that cheapening of capital or depreciation just as it does using a current cost measure.  What the historic cost measure does is provide a more realistic measure of the value that must be reproduced in any new production period to make a profit.  Capitalists measure their profit against what the value of their fixed assets cost when they start production, not on what they might cost to replace them in the future.  As AK says in his book (p112), “what makes the current cost rate of profit bogus is ….that it is not a measure of profit as a percentage of past investment”.  Using historic costs, after-tax profitability still shows a trend decline after 1965 as well as after 1947.

Now it could be argued that the depreciation of fixed assets (not the value of the stock of assets) during the production period should be measured at current cost.  That means the fixed assets are measured in historic costs, but the depreciation is measured at current cost.  AK does this measure of profitability in his book, Figure 6.3 on p111.  That figure shows that from 1982 there was a rise in the rate of profit to 1997 of 12%.  That’s much less than the current cost measure which rose 45%, while AK’s favoured measure (historic cost assets and historic cost depreciation) fell 2% between 1982 and 1997.  I ‘cherry pick’ 1997 because any reader of my blog knows well that it is my thesis that the period of 1982 to 1997 (the so-called period of neoliberalism) does show a rise in profitability, however you measure it.  And this historic cost measure confirms that.

I also measured the US rate of profit measuring fixed assets in historic costs, but with depreciation in current costs.  I find that the rate rose 12.1% from 1947-65, fell 25.8% 1965-82, rose 11.9% 1982-97 and then fell 17.4% 1997-09.   The rate of profit was in trend decline 1947 to 2009, or from 1965 if you prefer, but there was a cyclical feature to US profitability.

BJ also argues that the rate of profit measure should include wages (variable capital) in the denominator as Marx did and also inventories in the measure of constant capital – in other words, circulating capital has been excluded by AK unreasonably.  AK explains why in his book that he does that – basically data on the turnover of circulating capital are unavailable and/or unreliable.  AK does try out a measure including inventories in his book (Figure 5.3).  BJ argues that this measure shows again that the rate of profit does not fall so dramatically – but it still falls.

I have done a measure of after-tax profits that includes employee compensation and inventories, in other words, circulating capital.  I’ve done it using the current costs measure to fixed assets.  In other words, I have used all the categories that BJ wants altered or added from AK’s – and ignored all AK’s caveats about using these categories.  It still shows that there was a trend decline in the rate of profit since 1950.  Moreover, the rise in the after-tax rate of profit, using BJ’s categories, peaks in 1997 and subsequent peaks do not surpass that year.

Phew! So after all this to and froing with measures of the US rate of profit, what can we conclude?  Since 1947 did the rate of profit in the US rise or fall or do nothing?  On nearly all measures of the numerator and denominator, whether by AK, myself or all the other attempts by Marxist economists in the last decade, there was trend decline to 2009.

Was there a rise in the rate of profit from 1982 to 2009, suggesting a different era for US capitalism and suggesting that the rate of profit is not a key cause of the recessions of 1990-1, 2001 or the Great Recession?  Many say yes, presumably including BJ.  AK says no (or at least he says there was not a significant rise).  I say there was a significant rise from 1982 to 1997, but since then US profitability has been in a down phase right up to now.  Both AK and I conclude that Marx’s law of profitability is the underlying (not proximate) cause of capitalist crises and I think the empirical evidence for the US economy goes a long way to confirm that.

But I’ll come back again to interpretations of the US and other countries’ profitability data in another post.

20 thoughts on “The rate of profit: the devil in the detail

  1. I think this is a pretty fair summary of the differences. AKs theory says that the original cost of the investment “can’t be revalued”. That the value of that investment does not exist in the assets its purchases, but in the “book value”, their ideal representation on the books of the capitalist firm. That’s basically completely wrong.
    Once money has been spent then its value exists in the assets it has purchased.There’s a simple reason for that, if the capitalists wanted to realise/terminate their investment, their only way of doing so would be by selling the assets they had purchased at the current value, not the price they had originally paid for them.
    Although Kliman’s theory is hardly clear.
    After asserting that the book value “can’t be revalued” he then says that the assets should be measured net of depreciation – that is after they have been revalued.
    But we’ve been over this at length.
    On the issue of “property income”, this confuses/conflates capitalist profits with the tax income of the capitalist state. In so doing it removes the key motive or neo-liberalism, the shrinkage of the state to boost capitalist profits. Much of those state expenditures in fact form part of the social wage of the workers.
    Finally on the issue as to whether the rate of profit is “trendless”. Well that really depends on how you draw your trend doesn’t it? Looking at both your graphs I would suggest that they show a generally falling trend in profitability to the early 1980s and a generally rising trend after, including the aftermath of the recent recession. Kliman doesn’t include the most recent years in his book, but its clear that the sharp rise in profitability up to 2007 has been more or less restored over the last two years. So if you want to take the trend up to now – that is literally now – it is not possible to argue that the trend is down, for the simple reason it is up.

  2. One other small comment. I do not think there’s anything wrong with using the “historic” value of the fixed capital stock as part of the denominator for the rop estimate.
    As depreciation takes place at current rates, it necessarily reflects current rates of moral and physical changes in the socially necessary labour time required for commodity production. In other words the “historic” value of the fixed capital stock, is not a historic in Kliman’s unchanging sense at all. In fact it is a modified form of current value. IMO the actual rate of profit is likely to be somewhere between the historic and current rates.
    Kliman claims that the current value of the fixed capital stock is lower than the historic value and estimates of the rate of profit based on it are consequently, higher.
    That’s wrong too.
    If you take the BEAs estimates for the current value of fixed capital stock, it is higher not lower than the historic value due to the effect of inflation. This generally outweighs the effect of depreciation. Consequently, estimates of the rate of profit using the current measure are lower not higher than historic measures.
    The opposite of Kliman’s assertion.

  3. There are several problems here I think.

    1. I don’t think there is any general disagreement over the idea that the Rate of Profit declines over the longer term in line with a rising Organic Composition of Capital. That is what we see in the trend from 1950. But, surely the issue is over changes in the Rate of Profit over shorter time periods i.e. over the cycle of the Long Wave. A look at that shows pretty much what I would expect. During the early part of a Long Wave Boom (1949-65, 2000-2010)it tends to rise, because Capital can exploit abundant supplies of Labour, it has access to cheap raw materials. But, from about 10-12 years in, the Rate of Profit begins to decline as these things reverse (1965-83). A similar trend rise can be seen from 1983, for the reasons set out above, abundant labour, cheap raw materials, increases in productivity.

    2. It is not the fact that AK includes taxes in the calculation of profits per se that is the problem. As marx sets out Surplus Value is indeed divided into Profits, Rent, Interest and Taxes. The problem is two fold. Firstly, “Property Income” in the way used is a return to Adam Smith’s “Trinity Formula” criticsed by Marx. It does not account for the Value of Constant Capital in Total National Product. It thereby considerably overstates profits. Secondly, in the Critique of the Gotha Programme, marx sets out what he means by taxes, when he says that they are the basis of the financing of the State’s operation and nothing more. Including as Taxes, things which go to other functions than simply maintaininig the Capitalist State’s administrative function is therefore false. “Taxes”, which go to pay for other things such as the provision of health, education, social services, pensions, benefits and so on are not really taxes at all in Marx’s terms. They are collective payments by workers to cover these things, just as if they bought them from wages in the marketplace. They are in effect no different from the Truck System used by 19th Century employers, whereby they deducted money from wages, providing workers with tokens which could only be redeemed in the Company Store.

    In other words, they are deductions not out of profits, but out of the Wage Fund. Again that exaggerates the level of profits.

    3. As Marx showed in his analysis of The Rate Of Turnover and its effect on the Rate of Profit, simply taking the figures presented in Company P&L Accounts, or in National Accounts at face value can be highly misleading.

    ” The s’n in the formula p’ = s’n (v/C) stands, as has been said, for the thing called in Book II [English edition: Vol. II, p. 295. — Ed.] the annual rate of surplus-value. In the above case it is 153 11/13% multiplied by 8½ or in exact figures, 1,307 9/18%. Thus, if a certain Biedermann [Biedermann — Philistine. A pun, being also the name of the editor of the Deutsche Allgemeine Zeitung. — Ed.] was shocked by the abnormity of an annual rate of surplus-value of 1,000% used as an illustration in Book II, he will now perhaps be pacified by this annual rate of surplus-value of more than 1,300% taken from the living experience of Manchester. In times of greatest prosperity, such as we have not indeed seen for a long time, such a rate is by no means a rarity.

    For that matter we have here an illustration of the actual composition of capital in modern large-scale industry. The total capital is broken up into £12,182 constant and £318 variable capital, a sum of £12,500. In terms of percent this is 97½c + 2½v = 100 C. Only one-fortieth of the total, but in more than an eight-fold annual turnover, serves for the payment of wages.

    Since very few capitalists ever think of making calculations of this sort with reference to their own business, statistics is almost completely silent about the relation of the constant portion of the total social capital to its variable portion. Only the American census gives what is possible under modern conditions, namely the sum of wages paid in each line of business and the profits realised. Questionable as they may be, being based on the capitalist’s own uncontrolled statements, they are nevertheless very valuable and the only records available to us on this subject. [In Europe we are far too delicate to expect such revelations from our major capitalists. — F.E.]”

    As Marx shows in this section there are numerous things, which influence the Rate of Turnover of Capital. If we consider all of those things, such as the icnreases in productivity, the revolutionisiing in communications and transport that the Internet, and other forms of ICT have brought about, if we look to the shift in the locus of where production now takes place i.e. the huge decline in large scale, heavy industry and manufacturing (much of which have very long turnover periods) towards, service production, light industry, and so on – as well as the factors I have set out elsewhere such as the shift towards industries which rely on very high value, complex labour-power rather than large amounts of Constant Capital (media, computer games production, software production, financial services, entertainment and leisure and so on) it is not difficult to understand that raw data on profit volumes against Capital employed can be very misleading, as Marx demonstrates above.

    On the one hand, then AK’s method overstates the volume of profit in certain ways, and on the other it understates the changes in the Rate of turnover of Capital, shifts if the Organic Composition of Capital etc.

  4. One point where I disagree with Bill is his comment “In so doing it removes the key motive or neo-liberalism, the shrinkage of the state to boost capitalist profits. ”

    I don’t think the objective of Neo-Liberalism ever was the shrinkage of the State. Neo-Liberalism simply reflects the fact that Money Capital in the West became more powerful than Industrial Capital, along with the inability of Keynesian Fiscal Stimulus to deal with the end of the Long Wave Boom. But, in both the US and UK, the rise of Neo-Liberalism was not accompanied with a shrinkage of the State. In both countries the State continued to rise under both Reagan and Thatcher. Under Reagan it led to a ballooning of the deficits, which led to the crash of 1987. Under Thatcher and Major, between 1979 and 1997 borrowing accounted for 3.4% of GDP, between 1997 and and 2005 it averaged just 1.2%.

    In line with the fact that Neo-Liberalism simply means that the Money Capitalists have the upper hand, what changed was that State intervnetion shifted from Fiscal Policy to Monetary Policy. Who benefits from low interest rates maintained by money printing? Money Capitalists, because it affects the price they pay to borrow in the Money Markets, far more than it affects the commercial rates charged to businesses and consumers.

    Of course, Capital in general seeks to minimise the amount that the State pays out for things such as education and healthcare, provided the same level of quality of provision can be sustained to meet its needs in the reproduction of Labour Power, just as Capital sought to reduce the price of food through abolition of the Corn Laws. Its only in this sense that by raising efficiency in the production of State provided wage goods, thereby reducing the Value of Labour Power, that Capital seeks to reduce State Expenditure.

    A look at the experience of healthcare in the US demonstrates this from the other side. The inefficient nature of private health insurance in the US means that US Big capital is faced with a hurdle compared to European Capital. It has to pay out large amounts in Health Insurance payments for its workers, whereas European Capitalists benefit from a more efficient socialised insurance scheme. That is why Big capital has sought to have introduced some kind of similar system – small capital has opposed it, because in general they do not contribute, making their workers pay themselves, or rely on Medicare. It is also why George Bush took €750 billion of prescription costs off the books of private companies and socialised it on to the books of the State!

  5. On the issue of the impact of changes in turnover. As the NIPA are aggregates of incomes then they will reflect a change in turnover, as wages, profits etc. will change with the turnover of capital.
    In other words, if as under globalisation, turnover has increased, then the same fixed capital stock will be able to generate a higher proportion of profits in the same period.
    As indeed, has in fact happened, these higher profits will be reflected in the annual aggregate mass of different incomes in the statistics.

  6. The point is that the amount of Capital laid out is reduced. As marx demonstrates this is not reflected in Capitalist Accounting. Suppose, a firm employs 100 workers at £10,000 per worker. In the accounts this will appear as £1 million paid out in wages i.e. Variable Capital. But, as Marx points out, if the product of the workers is realised quickly, then the amount laid out in Capital is in reality much less than this.

    A shipyard or aircraft factory may indeed have to keep paying these 100 workers each week for a year until the ship or plane is finished, and the Value created by the workers is realised with the sale of the product. The capitalist will have laid out £1 million. Suppose the Constant Capital used up is also £1 million , and the Surplus Value is £1 million. The rate of profit will be 33.3%.

    However, suppose we are talking instead about a high class restaurant. Now, the product of the restaurant is completed and sold on an hourly basis, but let us take it as being on a weekly basis for ease of calculation. Now, the capitalist lays out £20,000 a week in wages (in fact, because they are likely to be paid in arrears the Capitalist will realise the product of their labour before they have been paid for it), and also lays out £20,000 for Constant Capital. But, by the end of the week, the restaurant will have taken in £60,000, £40,000 of which goes to repalce the Variable and Constant Capital used up. So, now as Marx sets out the Capital advanced is not £2 million, but just £40,000, because the cost of reproducing labour and Constant Capital is met out of sales.

    But, the total profit earned is still £1 million. Now the Rate of Profit is £1 million/40,000 x 100 = 2500%. This is not reflected in the official data.

    But, it is precisely this kind of shift from manufacturing to services, and towards high value production, using complex labour, and a far more rapid turnover due to the ability to transmit the end product more quickly due to the nature of the product, the introduction of the Internet etc. that we have seen over the last 30 years. We have also seen the introduction of other things such as Just In Time, and Flexible Specialisation in production that is likely to have resulted in a significant increase in the turnover of Capital, and a subsequent rise in the Rate of profit.

    1. Yes I agree with your example, but this is still reflected in the aggregates measured in the BEA stats. Here’s how;

      Let’s say that one cycle of a capital looks like this
      Fixed capital stock = 400
      Moral depreciation = 25
      Capital consumption = 25 (i.e. cost of repairs if required etc.)
      Raw materials = 50

      (Total constant = 500)

      Variable = 100

      (total capital = 600)

      Surplus = 100

      Rate of profit = 100/600=16.7%

      Turnover doubles so this cycle

      Fixed capital stock still 400 – doesn’t count already paid for = 0
      Moral depreciation – doesn’t count as the period is still the same = 0
      Capital consumption = 25
      Raw materials = 50

      (Total constant = 75)

      Variable = 100

      (Total capital = 175)

      Surplus = 100

      Rate of profit = 100/175 = 57%


      Fixed capital stock = 400
      Moral depreciation = 25
      Capital consumption = 50
      Raw materials = 50

      (Total constant = 525)

      Variable = 200

      (Total capital = 725)

      Surplus = 200

      Rate of profit = 200/725 = 26%

      The extra cycle of production reduces the organic composition of capital and so raises the rate of profit. Capitalist now earns super profit, assuming no fall in the price of the good,as the prevailing socially necessary labour time applies until new capitalists come in attracted by super profits, the increased turnover of capital becomes the norm and competition brings the price of the good down accordingly.

  7. Bill,

    No, this is still wrong. If we assume here that the circulating capital now turns over twice in the period instead of once then:

    1) Fixed capital laid out remains 400;
    2) Capital laid out for Capital consumption remains for the whole period 25, because that used in the second turnover is financed out of revenue i.e. out of the sales at the end of the first turnover, not out of Capital;
    3) Variable laid out for the whole period remains 100, for the same reason;
    4 Total Surplus is 200, because 100 is made in each turnover.

    The Rate of Profit is then 200 / 400 + 25 + 100 = 38.1%.

    To go back to my original example think of it this way. Suppose the two Capitalists begin with the same Capital of £2 million. The first Capitalist lays out there Capital as specified above. They have to keep paying £20,000 per week, and £20,000 per week for materials out of this Capital, because until the commodity is sold at the end of the year, they have no revenue to cover the payment of these things.

    Now look at things from the perspective of the restaurant Capitalist. They have to lay out £40,000 of Capital per week. They conclude that with their £2 million they can open not one but 50 restaurants. So, now over these 50 resturants they lay out each week £1 million in Constant Capital and the same in Variable Capital.

    C 1 million + V 1 million + S 1 million = 3 million.

    But, because the capital is turned over each week, they can continue to operate at this level without resort to any addiitonal Capital. Out of their £3 miilion of weekly takings they are able to pay for next week’s labour and materials. So, whereas, Capitalist 1 will have started with a capital of £2 million, their total profit during the year will be £1 million, giving them a rate of profit of 50%.

    Capitalist 2 with exactly the same amount of Capital, with the same Rate of Surplus Value will have made a total profit of £50 million during the year, giving them a rate of profit of 50/2 = 2500%.

    But, as Marx points out if you looked at the books of these two companies, and this applies to national accounts too, you would see that Company 1 would have laid out £1 million in wages, and £1 million in materials, whereas Capitalist 2 would be seen to have laid out £50 million in wages, and £50 million in materials. Both would appear on this basis to have made a rate of profit of 50%. But, as marx says, Capitalist 2 has NOT laid out £50 million for wages, and for materials. It has only laid out the Capital it began with. It has paid for wages and materials in the other 49 cycles not out of Capital, but out of revenue, Constant and variable Capital reproduced itself in the commodity, and the commodity was sold in a week rather than in a year.

    As I’ve pointed out before this is highly important for Capital Accumulation and development theory. Bukharin pointed to it in his Economics of the Transition period – Ken Tarbuck did a good account of it back in the 1970’s – where he pointed out that an economy like the USSR that invested too heavily in large scale projects with prolonged turnover periods could actually experience negative Capital Accumulation and growth, precisely because Value was being absorbed in the production that was not being reproduced in the sale of the commodity, and thereby acted as a drag on the rest of the economy.

  8. Errata:

    Forgot to include the 50 for raw materials, which again remains at 50 for the same reason as Variable capital, and repairs. I left out the 25 for moral depreciation also, but I beleive that this amounts to a capital Loss not a cost of production. In which case it would be:

    200 / (400-25) + 25 + 50 + 100 = 36.36%.

    1. Don’t agree Boffy. Engels discusses it (as I’m sure you know) here, as far as I can tell this confirms my thinking;

      For this purpose we select the cotton spinnery of 10,000 mule spindles described in Book I (S. 209/201) [English edition: p. 219. — Ed.] and assume that the data given there for one week of April 1871, are in force during the whole year. The fixed capital incorporated in the machinery was £10,000. The circulating capital was not given. We assume it to have been £2,500. This is a rather high estimate, but justified by the assumption, which we must always make here, that no credit operations were effected, hence no permanent or temporary employment of other people’s capital. The value of the weekly product was composed of £20 for depreciation of machinery, £358 circulating constant advanced capital (rent £6; cotton £342; coal, gas, oil, £10), £52 variable capital paid out for wages, and £80 surplus-value. Therefore,

      20c (depreciation) + 358c + 52v + 80s = 510.

      The weekly advance of circulating capital therefore was 358c + 52v = 410. In terms of per cent this was 87.3c + 12.7v. For the entire circulating capital of £2,500 this would be £2,182 constant and £318 variable capital. Since the total expenditure for wages in one year was 52 times £52, or £2,704, it follows that in a year the variable capital of £318 was turned over almost exactly 8½ times. The rate of surplus-value was 80/52 = 153 11/13. We calculate the rate of profit on the basis of these elements by inserting the above values in the formula p’ = s’n (v/C) : s’ = 153 11/13, n = 8½, v = 318, C = 12,500; hence: p’ = 153 11/13 × 8½ × 318/12,500 = 33.27%.


  9. Bill,

    That was the quote I was referring to, but it does not agree with your calculation. The whole point is that in this calculation, Engels multiplies the average rate of profit calculated on your basis by the number of times that Capital turns over in the period. That is exactly what I was arguing.

    In the example above, the official calculation of profit would be 50%, but because the capital turns over 50 times during the year, you have to multiply that figure by 50 i.e. by “n” in the formula Engels sets out above.

    Just think about it. If I invest an amount of money and make a 50% return on it, in a week, is this the same as if I make a 50% return on it, in a year. Clearly it is not, even on a simple interest basis it is 50 times as much.

    In fact, in this section of capital, he sets out various formula for calculating the rate of profit based on different rates of turnover for different elements of Capital.

  10. The relevant narrative to go with this formula where Engels multiplies the Rate of profit by the number of times the Capital turns over is this:

    “To single out the effect of the turnover of total capital on the rate of profit we must assume all other conditions of the capitals to be compared as equal. Aside from the rate of surplus-value and the working-day it is also notably the per cent composition which we must assume to be the same. Now let us take a capital A composed of 80c + 20v = 100 C, which makes two turnovers yearly at a rate of surplus-value of 100%. The annual product is then:

    160c + 40v + 40s. However, to determine the rate of profit we do not calculate the 40s on the turned-over capital-value of 200, but on the advanced capital of 100, and thus obtain p’ = 40%.

    Now let us compare this with a capital B = 160c + 40v = 200 C, which has the same rate of surplus-value of 100%, but which is turned over only once a year. The annual product of this capital is, therefore, the same as that of A:

    160c + 40v + 40s. But this time the 40s are to be calculated on an advance of capital amounting to 200, which yields a rate of profit of only 20%, or one-half that of A.

    We find, then, that for capitals with an equal per cent composition, with equal rates of surplus-value and equal working-days, the rates of profit of the two capitals are related inversely as their periods of turnover.”

    1. In which case we’re not disagreeing, just arguing about the maths, not to say that’s unimportant but its not that important.
      The whole Kliman thing is an attempt to “solve” the transformation problem by maths – the fact that in Marx’s example the total of value is not equal to the total of the prices of producation after value has been transformed into prices.
      And there is no mathematical solution to that. Marx’s example is both historical and current. The transformation of values into prices of production took place over several centuries. But it also goes on in the current on a daily basis now. Values form the centre around which prices of production fluctuate. That is inherently disproportional. A non-equilibrium transformation.
      The other point is of course that Marx assumes that rates of profit are equal in his example, in reality they are never equal, which is why capital flows in and out of different sectors.

      1. Well, I think the maths are pretty important in this respect, but that’s a separate discussion to the substantive point about Kliman’s argument. There have been numerous mathematical solutions to the transformation problem from Von Bortkiewizc, to Francis Seton (whose solution as far as I understand it, I think is the best) to a range of solutions provided by economists writing in Capital and Class during the 1980’s.

        But, I agree with you that most of these mathematical solutions miss the point. A few years ago, I used an Excel Spredsheet’s Solve function to show that its possible to arrive at solutions that meet all the necessary conditions, the only thing being that you end up with a different Rate of profit based on prices than with Values. But, that is not important, because the Rate of Profit is itself a derived figure, not a constant.

        But, the point is that these mathematical models are removed from the actual process by which Values are transformed into prices, which as you say takes palce over a period. The essential process is this. In areas of production with high organic compositions of Capital, and low rates of profit, Capital does not accumulate so fast, or even moves to other areas of production, where the opposite is the case. As a result, Supply falls in the former and rises in the latter. Consequently, market prices rise in the former and fall in the latter, with a consequent effect on profits.

        The sharing out of the total profit is not done by some mathematical process, which determines what prices Capitalists will charge, but is itself a simple consequence of Competition, and Supply and Demand!

        But, you are quite right to say that there is no such thing as an “Average Rate of Profit” other than as an abstraction. It is bourgeois theory, which is based on the idea of equilibrium. Marx’s theory by contrast is based on the fact that there is constant and dynamic disequilibium, and as a result any average rate of profit is and has to be momentary, no sooner achieved than disappeared.

      2. As an addition, and example, look at the experience of many products. The Harvard Business School Product Cycle Model is quite good in terms of explaining this.

        Many new products begin by requiring smaller amounts of Constant Capital, but large amounts of complex Variable Capital. The latter is employed in R&D, for example, and skilled workers are needed for production. In general, these products as a result are highly priced, and have high profit margins. Think of calculators when they were first introduced in the early 1970’s. Seeing these high profit rates, other producers enter the market, supply increases and prices begin to fall. Competition between suppliers causes them to look for ways to mechanise production, so as to replace the high skilled labour. Machines are developed to do that, which in turn allows production to be ramped up.

        The consequence is that over a period the organic composition of Capital tends more to the average, prices and profits fall as output rises. As the PCM shows, at this stage, production of many mature products are then relocated to low wage economies.

      3. To see the importance of the maths in relation to the Rate of Turnover consider Michael’s data provided above. If we take a figure for the average turnover of Capital in 1950 as being 10, and today as being 11, then we would have using his first set of data, an actual Rate of Profit, using Marx and Engels formula of 7.2 x 10 = 72%, and 6 x 11 = 66%. If we took the figure for 2007 prior to the crash we would have 7 x 11 = 77%. Moreover, as we are still in the Spring phase of the Long Wave Boom, it is likely that the Rate of Profit has not yet hit its peak.

        Additionally, because the rate of turnover would have been increasing continuously between 1950 and today, the figures for all the intermediate data points would need to be revised upwards, so that instead of a declining trend line, we would be likely to see that the trend over the period for the real rate of profit has been rising.

        In fact, whatever the actual starting point for the average rate of turnover in 1950, the increase in its rate over the last 60 years is likely to be far higher than the 10% assumed here, for all the reasons I have set out elsewhere in relation to the shift in the nature of production, and the introduction of much improved means of communication, and payment, whther it be the introduction of container ports and shipping, to the massive increase in bulk shipping capacity for tankers, to the introduction of large scale jet air freight, to the introduction of motorways, and now the introduction of the Internet.

    2. Also came across another amusing point. Kliman’s quote “you can’t build the house of today with the bricks of tomorrow” actually originates with Boem Bawerk! it was used by Bukharin in arguing against Preobrazenski in the 1920s.
      Well there’s a thing huh?!

  11. Re: the two graphs.

    You cannot just fit a regression line to any old set of data. The underlying assumption is that the data lie in a straight line with normally distributed errors. If the raw data and do not fit the assumptions (and these do not), you need to try and find a suitable transformation rather than just ignore it.

    Drawing the best straight line through a U-shaped set of data-points gives completely meaningless results. You have in no way shown that there is no trend because you have not made a serious attempt to estimate what the trend actually is.

    If you’re not sure how to deal with transformations, find a statistician to help you. If you really can’t find one, you could play around with this app until you have some kind of idea of what you are doing:

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.

%d bloggers like this: