1967. Ernest Mandel: A critique of the theory of economic growth, in the spirit of Marx’s Capital

This text was Ernest Mandel’s contribution to a colloquium on the occasion of the centenary of Marx’s Capital, organized by the institute of political science of the Goethe University Frankfurt and publisher Europäische Verlagsanstalt. The contributions and discussion were originally published in Kritik der politischen Ökonomie heute,100 Jahre Kapital (Frankfurt am Main, 1968). 

In the contemporary theory of economic growth, growth is mainly a function of two elements: the savings ratio and the capital coefficient.ⁱ If a country saves 12 per cent of its national income, and the capital coefficient is 4:1 (meaning four billion Marks must be invested to increase the national income by one billion), the rate of growth of the national income will be 3 per cent per year. 

In this contribution we will leave aside the specific problems of economic growth in underdeveloped countries and explore current views only in regard to the industrialized capitalist countries.
 

1.
 

The Keynesian theory of income, which posited the identity of aggregate net savings and aggregate investments, emerged in the context of prolonged economic stagnation in Britain between the world wars and the depression of 1929-32. We do not want to deny the functional character of this theory, from the point of view of effectively intervening in the capitalist economy and limiting economic fluctuations, but we have the impression that in reality it raises more theoretical problems than it solves.

For one thing, the equation of net social savings with investments leaves out the important difference in the purpose of saving by different social classes. If real incomes rise, then without doubt wage and salary earners will also increasingly participate in saving. But we

cannot overlook that this saving has a very different function, depending on whether we are dealing with the working class in Marx’s sense (meaning the social class which, whether its income be high or low, is forced to sell its labour power while it lacks sufficient means to organize an income ‘on its own account’), or with the petite or haute bourgeoisie.

In the first case, savings represent only deferred consumption, which is most obvious in the so-called ‘institutional saving’ of the social security system. These savings do not lead to the purchase of means of production or accumulation of wealth and only among a part of the working class does such saving lead to home ownership. In any case, in the industrialized countries such homes should for the most part be regarded as durable consumer goods, not as value-producing capital assets.

Only among those social classes who are either already owners of capital, or who are in the process of primitive accumulation of capital, does saving lead to the acquisition of wealth, meaning new capital that is not used up wholly in consumption in over a period of, let’s say, 20 years.

This distinction is already by itself significant because it gives insight into the long-term reproduction of the basic conditions of a capitalist economy: its division into owners and non-owners of means of production, wealth and capital. The data show without doubt that this remains the case today. To give just one example: in Britain, 78 per cent of the population during the 1950s owned less than 10 per cent of net private wealth, i.e. on average less than 1,000 Marks per capita, while 2 percent of the population owned 60 per cent of this wealth.ⁱⁱ And Professor Titmuss is of the opinion that income inequality in the exemplary welfare state that is called Britain has considerably increased over the last 30 years, instead of decreasing.ⁱⁱⁱ

And moreover, the equation of aggregate net savings with net investments gives no insight into the parallel phenomena of the creation of inflationary credits by the banks and the increasing wastage of the social surplus product from the point of view of the social reproduction process. From the point of view of society, in the long run it is impossible to consider irrelevant whether additional goods and services created by new investments enlarge the material base of production, generate a quantitatively and or qualitatively improved labour force and growing constant capital, or whether they do not enter economic reproduction at all, such as objects for speculation or weapons. Economic growth can in the long term only occur in the form of expanded reproduction, meaning on the basis of a larger workforce and better working equipment. Therefore, it seems to us relevant to regard not so much the absolute savings ratio but rather the rate of productive investment as definitive for the problem of economic growth.

Finally, there is, as several authors have pointed out, a necessary relationship between capital stock and the economic growth induced by productive investments. If the turnover of fixed capital remains constant, the part of total investment which only maintains the stock of total social capital, rather than enlarging it, will increase as the stock increases. The growth rate of incomes and net new investments must then also increase, in order to enlarge this stock with at a constant growth rate. Via this detour, several modern economists have rediscovered Marx’s law of the increasing organic composition of capital.

This law in no way contradicts the tendency towards a declining fixed capital outlay per manufactured product unit – as long as the wage costs involved fall even more quickly. This is a statistic fact. In US industry, between 1880 and 1922, expenditure on fixed capital for production of one unit increased from 100 to 275; in the same period, wage costs per production unit fell from 100 to 50. Between 1922 and 1950, expenditure on fixed capital per of unit fell significantly, for US production as a whole (manufacturing and agriculture) to around 125, and for industry alone to 150. But at the same time, wage costs per unit fell to 25, a mere quarter of the 1880 level.^iv
 

2

The formula of the capital coefficient, meaning the relationship between the newly created income and the capital stock enlarged for the creation of this income, is given little attention in Marxist economic analysis. As is known, academic economists assume that this capital coefficient remains stable in the long term, at least in Western capitalist countries.^v In the Marxian framework, this would mean that, if the mass of constant capital and wages grows, all kinds of fluctuations in surplus value are still possible. The main forecasts of Marxist economic analysis – increasing organic composition of capital, a rising rate of surplus-value, and a falling rate of profit – would in this framework be mathematically possible.

However we might assess the importance of capital coefficients in econometrics (their uses in planning techniques are especially obvious) they are hardly relevant from the point of view of individual enterprises. Entrepreneurs could not care less exactly how the sum total of wages and profits relate to total capital outlays (including the wage bill); what interests them is the relationship of profits to capital invested, the profit-rate. If the capital coefficient falls, for example due to a boost in labour productivity as a result of relatively low-cost technological innovations while at the same time the rate of profit increases, then investment activity will increase and so will the growth rate of industrial production and national income, as shown by the Harrod-Domar model.^vi If, on the other hand, a falling capital coefficient is accompanied by a falling rate of profit, which always remains a possibility and has happened historically during economic slumps, for example, then the outcome will not be a rising but a falling growth rate in production and national income.

In other words, in a capitalist economy, macroeconomic objectives can only be achieved if they line up with the corporate striving for profit. Rising productive investment activity and a falling profit rate or falling capacity utilization are in the long run incompatible. This is one of the innate contradictions of the capitalist mode of production, against which late capitalist post-Keynesian techniques of money and credit creation and general ‘planning’ of the national economy are in the long run powerless. If one can regard recent credit constraints in West Germany and Britain as the immediate trigger of the current generalized economic downturn in Europe, its deeper cause goes back to the early 1960s and the uninterrupted increase of excess productive capacity since then. Otto Schlecht, the West German director-general for economic affairs, has estimated this at 25 per cent of total capacity in West Germany, and even 40 per cent in agriculture.^vii

But this in no way means that business circles who blame ‘excessive wage increases’, meaning decreasing profit rates, for the downturn in investment and the consequent economic recession are right. Undoubtedly high employment levels enabled a rise in real wages during 1961-65. But when we consider that during the same period capacity utilization rates were already gradually declining, we can only conclude that, higher real wages notwithstanding, effective demand of the ‘final consumers’ already trailed behind the expansion of productive capacity. If real wages had increased more slowly or not at all, then capacity utilization rates would have been even lower – in turn paralyzing employers’ investment activity.

It is indeed a characteristic of the capitalist cycle that only a coincidence of rising profits and expanding markets boosts productive investment activity. This coincidence, however, can only be temporary. That is why investment activity in late capitalism retains its wave-like movement. This will impose a cyclical character on production, despite all the post-Keynesian anti-cyclical techniques.

3

Marxist and post-Keynesian economists might well agree on the point that increased investment activity, determined in the last instance by profit expectations, is the prime mover of economic growth in industrialized capitalist countries. But the question can be rightly asked: What kind of investment activity, and for what purpose?

Even in a mode of production in which insufficiency and excess of capacity periodically alternate, corporations able to control their markets should also be able to calculate maximum growth rates of productive capacity in the long term, and thereby largely avoid ‘overheating’ and growing over-capacity. In such cases, recessions (but also booms) would be reduced to a minimum. In addition, business associations implement economic ‘programming’ in particular branches of the economy, while bourgeois states do the same in the entire economy. Through organs of the European Economic Community such programming is even coordinated internationally. The question then arises why a sudden upsurge of investment activity can occur at all in the late capitalist economy. This phenomenon cannot be explained solely or even primarily by the fact that in non-monopoly sectors of the economy, the investment activity of small and medium-sized business retains its classically anarchistic character, as this type of business is clearly of steadily declining importance in the capitalist mode of production.

As is known, Schumpeter explained the periodic upsurge of investment activity as a result of innovation^viii, and many economists today agree that the main cause of economic growth in industrial capitalism in the last instance is technological progress.^ix At the highest level of abstraction, Marx would not have disagreed. But, to borrow a well-known phrase from Engels, where others saw an answer, Marx saw a question. As to the question of why a periodic upsurge in business investment activity happens, Marx’s Capital gives a dual answer: competition among capitalists, and competition between capitalists and the working class.

The uneven development of technological progress between different economic and industrial branches explains why production increases do not spread gradually and evenly over all industrial sectors but occur by leaps, initially in individual corporations and sectors. Now, such a sudden burst of technological innovation conditions a major increase in profit margins. Or, in Marxian language, it allows a corporation to sell its product at current market prices while simultaneously realizing surplus-profit.^x But because the total mass of surplus-value is predetermined, meaning it is already created in the sphere of production and in the circulation process cannot be expanded, the surplus-profits of a corporation or industrial sector occur at the expense of other corporations and sectors, whose profits thus fall below the average. Here we have an initial explanation of the question why business owners are periodically forced to increase their investment activity, which as we know then quickly spreads and culminates in overheating and a runaway boom.^xi Whoever lags behind in the race is doomed to perish. Competition and private ownership of the means of production condition this unforgiving law.

In a situation of increasing market monopolization by a small number of giant corporations which largely eliminate price competition, the quest for surplus-profits shifts more and more towards the tendency to radically lower production costs through technological innovations, and the hunt for ‘new’ products through technical and scientific research. Just how much this uneven development of technical progress and technological innovation asserts itself today under the rule of monopoly capital, is shown by a brief look at the development of big business in the United States over the last 30 years, from 1935 to 1965. Numerous corporations that 30 years ago ranked among the biggest have fallen down the list, or even disappeared; think of the copper industry, the aluminium industry, the meat industry, railways, mining, agricultural machinery and even several steel trusts. On the other side, numerous corporations which for 30 years were of little importance or even hardly existed, today occupy leading positions, for example the airplane industry, electronics, computers and foodstuffs as well as the petrochemical industry. The only corporations whose position remains more or less the same are in car-manufacturing and the oil industry, and the major chemical trusts.

Even without this mutual competition among capitalists, which under conditions of private ownership of the means of production is inevitable, there exists another force imposing technological progress in the capitalist mode of production. This is competition between capitalists and workers. Under conditions of neo-capitalism, which in the current global context cannot afford any big economic crisis and consequently generalizes anti-recessionary and anti-cyclical policies of inflationary money and credit creation, the acute danger for capitalism emerges of long-term full employment. Given free trade unions, this could cause wage rises large enough to shift the total distribution of income between social classes, and rapidly lower the rate of surplus-value and profit.

Certainly, in theory there is a way out of this predicament: a voluntary or imposed restriction of trade-union power, meaning foregoing any attempt to utilize favourable market conditions to raise the share of wages in added value. But history shows us that this policy can in the long term only be imposed by a harsh dictatorship, and even then only for a limited time, since acute labour shortages cause conflicts of interest between individual employers and the employing class as a whole.^xii The only possibility for capitalists to force down wage costs consists of the accelerated replacement of ‘living’ by ‘dead’ labour, meaning the reproduction of the ‘normal’ wage regulator built into the capitalist mode of production: the reserve army of labour.

A clear conjunctural example of this, in West Germany in the last months of 1967, was how the end of full employment quickly increased ‘labour discipline’ within the enterprise, and the intensity and productivity of labour increased, meaning that profit rates increased as a result of an increase in absolute and relative surplus-value. And structurally the uninterrupted increase of semi-automation and full automation in industrialized countries signals nothing other than the forcible expulsion of millions of unskilled and semi-skilled workers from the production process, which in North America and Western Europe after the honeymoon of post-war business cycles, reintroduced mass unemployment. In the United States, the number of unskilled workers is said to have declined from 13 million in 1950 to less than 4 million in 1962.^xiii

Technical progress and technological innovation are not exogenous factors in the growth process of the capitalist economy. They are an inescapable result of the internal logic and contradictions of this mode of production.

4
 

‘Organized’ late capitalism, operating with economic programming and anti-cyclical techniques, can prevent deep economic crisis only at a double cost: increasing excess capacity, and permanent ‘creeping’ currency devaluation. Here, we do not want to explore the ramifications of this ‘secular inflation’, as Professor Hofmann calls it^xiv, for the pattern of economic growth. But the question arises, How is this growth process related to the phenomenon of permanent under-utilization of productive capacity?

We have already looked at one side of the process: the discontinuous character of technological innovation, which periodically causes a sudden spike of capital investment in so-called ‘emerging industries’ or ‘growth industries’. This rise triggers a general acceleration of the growth process, extending from the sector supplying equipment goods necessary for the ‘new’ industries to the sector creating means of production as a whole and from there extending to the economy as a whole. If in a particular sector, let’s say the steel industry where there is already 10 per cent excess capacity, there is suddenly demand to raise output by 20 per cent, there will be a need for new investment. At the end of the cycle, this new investment could mean excess capacity of 15 or 20 per cent, instead of 10 per cent.

The other side of the process should not be overlooked: the growing number of major corporations that can dominate their markets, meaning the growing number of areas characterized by the disappearance of price competition and ‘organized’ turnover, as well as the tendency towards general excess capacity in capitalist industry, must in the longer term lower the rhythm of growth. This tendency towards ‘secular stagnation’ in late capitalism has been recognized and analysed by numerous authors.^xv

If despite this, in the last 15 years in Western Europe and Japan, and in the last six years in the United States, a strong increase rather than a decline in growth rates has occurred as compared with averages for the first half of the century, this must be explained with reference to the operation of major countervailing tendencies. Among these are:

(a) the reduced turnover-time of fixed capital, or, as Marx might say, the accelerated ‘moral depreciation’ of fixed assets, resulting from an acceleration of technological innovation, which in turn is bound up with the permanent arms race between capitalist and non-capitalist states; (b) unusually high state expenditure compared with averages in previous longer periods in the history of capitalism – this applies especially to expenditure on armaments but also on infrastructural investments (in Western Europe and Japan, associated with the reconstruction of cities destroyed in the war); and (c) the increasing flow of society’s capital to areas other than domestic industry, to industries abroad, and to a much greater extent to the so-called ‘services sector’.

The flow of American capital from the United States to Canada, Western Europe and Japan (which is much larger than the export of capital to the so-called ‘developing countries’) stimulates a process of international competition and concentration of unprecedented proportions. But ultimately the only outcome is that the average level of labour productivity and technique is equalized between industrialized countries. The export of capital, then, does not offer a long-term solution to the problem of economic growth.

The same applies to the emergence of ‘new’ industries. These can temporarily stimulate the rate of growth but, since the economy of late capitalism is characterized by the presence of a great mass of surplus capital, this capital must flow with extraordinary speed in the ‘new’ sectors and there over a few years lower the profit rates to the social average profitrate, creating excess capacity. In recent times, we have seen this happen not only in the synthetics industry and petrochemicals in general but even in the electrical equipment industry.

There is thus a mass of surplus-value whose ‘normal’ capitalization increasingly threatens the valorization of capital as a whole. There are only two long-term possibilities to turn this mass into capital: the weapons industry and the services sector. The inherent danger of a ‘permanent arms economy’, that it will, to borrow Marx’s phrase, ‘transform forces of production into forces of destruction’ and, given the current state of military technology, destroy not just human culture but wipe out the human species altogether, needs no further comment here. However, the social-economic dynamic of the so-called ‘services sector’ deserves to be examined more closely.
 

5

We are confronted here with one of the most complex issues raised by Marx’s economic theory. There is a certain contradiction between the first three volumes of Capital, which seem to attribute the capacity to create new value and surplus-value only to material labour producing commodities that have a use-value, and the so-called fourth volume of Capital (the Theories of Surplus-Value) in which, by contrast, all wage labour enabling a capitalist entrepreneur to valorize capital, i.e. which is exchanged against capital and not just against money, is considered to create surplus-value.^xvi For the purpose of my inquiry, it is irrelevant how this contradiction is solved. Because in any case it is incontestable that the capitalist entrepreneur whose capital is invested in the services sector and there employs wage labour has, just like a merchant working with assistants, the opportunity to appropriate a fraction of total social surplus-value. The discussion centres only on the definition of the total volume of this part of surplus-value, not on whether capitalist service enterprises participate in its distribution.

So, what are the effects of an increasing flow of capital into this sphere? In the first instance, obviously increased expenditure of society’s capital on wages in the services sector. Since the real dilemma is not ‘productive or non-productive allocation of capital’, but rather ‘sterilization of capital or investment in the services sector’, the allocation of capital in this sector undoubtedly increases aggregate buying power and therefore the growth rate.^xvii

Second, because capital invested in the service sector participates in the general distribution of total surplus-value, pressure emerges towards increased rationalization of services. This, however, promotes the flow of even more capital into the sector. Mechanization, semi-automation and full automation advance by leaps and bounds, even if with considerable delay as compared to industry. Throughout this phase, the ‘industrializing’ service sector, just like ‘industrializing’ agriculture, provides expanding demands for industries producing means of production, and is since recent times a motor of increasing growth. Undoubtedly, this has played a significant role in the United States in the last 15 years.

Third, as soon as the process of rationalization and automation has reached its endpoint^xviii, or passes its highest point, the services sector loses its stimulating effect on the growth rate. This is because of the initially relative, later absolute, fall in the mass of wages, and because of falling demand for machinery and fixed equipment. On the whole, one can say that, with respect to the level of technological innovation and labour productivity, a belated process of equalization took place between the services sector and industry, thereby removing an additional refuge for surplus capital. A similar process occurred earlier in agriculture.

I have considered the services sector here only insofar as capitalist or potentially capitalist businesses are concerned, distribution, banking and finance, transport and entertainment. Obviously, a further deployment of capital in part of the so-called service sector that in more and more capitalist countries has been withdrawn from private capitalist enterprises, such as education, health, science and art, offers surplus capital no way out. It does not even do so indirectly when, by raising the qualifications of the industrial proletariat or skimming off the productive results of scientific research, it helps raise the profits of already invested capital.
 

6
 

We arrive at the following conclusions: the Keynesian theory of economic growth set itself the task of discovering pragmatic solutions for the problem of major economic crises, meaning avoiding a sudden sharp increase in unemployment of workers and means of production. In this way it strongly reinforced the tendency towards an autonomous and inflationary creation of money and credit that is inherent to late capitalism. This enabled a temporary acceleration of economic growth in industrialized late capitalism, growth that today is conditioned by power politics. But a growing proportion of the new economic wealth is diverted into areas which are sterile from the point of view of social reproduction (arms spending, real estate speculation, luxury assets, etc.).

This renewed economic growth occurs under conditions of monopoly capitalism, in which a growing number of economic spheres are dominated by a declining number of corporations that are able to control markets. The disappearance of price competition leads to investment financing through pricing, meaning to growing self-financing and overcapitalization. This leads in turn to the problem of excess capacity and permanent capital surpluses in the highly industrialized countries, a problem which is generally underestimated by the post-Keynesian school.

Excess capacity and full employment of the workforce can temporarily co-exist, above all if the balance of power compels the bourgeois state and capitalist class to enable this. Part of the working class, no longer employable in actual production, flows into the service, administration and military sectors. The temporarily burgeoning services sector represents, as Gillman argues^xix, above all an attempt to overcome the growing difficulties in the realization of surplus-value. Because the post-Keynesian school focuses more and more on the problems of inflationary equilibrium disturbances, without paying sufficient or indeed any attention to the long-term problems resulting from the combination of underemployed machinery with full employment and a growing drain to non-productive areas, their theory of economic growth increasingly acquires an abstract, unreal character.

This becomes quite clear in the third phase of late capitalism when, because of rapidly increasing automation, and after a previous shift in the balance of forces between social classes, excess capacity once again combines with increasing long-term unemployment.^xx Vacillating in their pragmatic method of approach between the temptation of deficit spending and the struggle against inflation through credit restrictions, the post-Keynesian school shows itself to be ever less capable of recognizing the structural problems of economic growth in late capitalism, never mind of actually solving them.

Inasmuch as it operates with monetary aggregates of national income, consumption, savings and investment ratios, the post-Keynesian theory of growth is able to provide pragmatic insights about how to mitigate or neutralize short-term equilibrium disturbances. But it cannot give insight into the long-term distortions caused by the contradictory tendencies of overcapitalization and automation which are intrinsic to late capitalism. Because it abstracts from the historically delineated structure of bourgeois society and its internal contradictions, it cannot recognize the development of these contradictions in society’s production process. Above all, it fails to grasp how social wealth, vastly expanded with the aid of modern technology, clashes with the requirements of private property, and how much the potential abundance of goods clashes with the laws of market economy.

Marxist economic theory is better equipped for this. For this theory, the six strategic relationships that determine the rhythm and scope of economic growth in capitalism are the rate of surplus-value, the rate of profit, the rate of accumulation, the organic composition of capital, the turnover-time of fixed capital and the volume of non-accumulated surplus-value spend as revenue.^xxi With these factors, one can establish that the greater the degree of monopolization and market control in individual spheres, the more powerfully surplus capital makes its appearance. Under conditions of ‘organized’ late capitalism this leads to the phenomenon of deferred overproduction, to excess capacity, which must ultimately cause long-term decline in the economic growth rate.

To be sure, even under these conditions, the wishful dream of continuing ‘frictionless’ functioning of the social system can appear, only under one condition: that a growing proportion of non-valorizable surplus-value is spent on social security. In other words, a growing portion of the population without jobs that provide an adequate living standard would simply be sustained by society. This would then produce a tendency towards the regression of the modern working class into a ‘proletariat’ in the sense it has for Classical antiquity.

Such a ‘solution’ would possibly conform to the economic logic of a modern slave society, but is really not compatible with the capitalist mode of production. A guaranteed minimum living standard for all members of capitalist society would lead to a sharp rise in the minimum wage. This is so because without a significant difference between this guaranteed minimum income and the minimum wage, the mass of workers would simply stop working as the economic compulsion to sell one’s labour power, necessary for the existence of capitalism, would have disappeared. But a rise in the minimum wage sufficient to keep people working would cause a sudden decline in profit rates and hence a rapid decline in investment activity.

In addition, such a society would be fundamentally a stagnant one, and long-term stagnation of the market would lead to the gravest overproduction crises in the sector producing means of production; modern economics has already shown how much full employment in this sector depends on the overall economic growth rate.

In either case, the system will not be consolidated, but rather carried on ad absurdum. More and more machines and equipment will be produced at gigantic expense, to be utilized less and less. An ever-growing proportion of the gigantic mass of commodities pumped out by automated production would scarcely correspond to any real need. Through direct or indirect manipulation, and at constantly growing expense, such commodities would have to be pushed onto a bewildered mass of consumers. And even then, a part would remain unsellable. The valorization of capital will run up against insoluble difficulties. Long before this point, however, social contradictions will have become acute to the point of exploding.

From the point of view of Marx’s theory of value, this is hardly surprising. In this theory, generalized automation – including of services –— means a rapid decline in total surplus-value produced, and growing difficulties for the valorization of capital. Production which constantly generates less and less income, because it employs fewer and fewer people, cannot in the long run function on the basis of commodity production. An abundance of goods is ultimately incompatible with a market economy.

A commodity and money economy has the inherent tendency to reify human activity and commercialize leisure activities, thereby turning such activities into new sources of alienation. But the socialization of the means of production under conditions of workers’ control and their planned organization would allow a move away from the goal of maximizing individual income towards a rational satisfaction of needs. This would enable the use of available productive forces to create well-rounded personalities rather than a burgeoning mass of useless or harmful commodities. Increasing social wealth will then mean a radical reduction of labour-time, halving the working day and working week, generalized tertiary education, and strongly increased social consumption. Employment in education and health, science and art would expend rapidly while the commodity and money economy could wither away.

Only within such a framework can people really gain mastery over the unfettered forces of production, instead of being their slaves as they are today. Only within such a framework can the main goals of socialist society be realized, which I would like to express here paradoxically as follows: a slowdown of economic growth, because an abundance of rationally consumable goods already exists; maximum development of 'inequality', in the sense of the personal development of all human individuals as unique beings, as the historical preconditions of socio-economic equality for this have already been created; and a definitive humanization of relations among people, because from now on the development of each has the development of all as its precondition.


Translation by Alex de Jong.