A Marxist Guide to Capitalist Crises
“A Marxist Guide to Capitalist Crises,” an eBook created from the key posts on the Critique of Crisis Theory blog, is currently in production. We’ll be sharing the completed chapters between our regular postings.
Chapter 34: The Long Cycle — Summary and Conclusions
In the preceding chapters, we examined whether the capitalist economy experiences cycles considerably longer than the industrial cycles of approximately ten years. As we saw, it has been proposed by various economists — both bourgeois and Marxists — over the last 100 years that, in addition to the 10-year industrial cycles and shorter inventory cycles, there also exists a long cycle of approximately 50 years’ duration.
As part of our exploration of the thesis of the existence of a long cycle, we have examined the concrete history of the cycles and crises that have occurred in the global capitalist economy from the crisis of 1847 to the crisis of 2007-09. Over these 161 years, we have seen decades when economic growth surged ahead and other periods dominated by prolonged depression or stagnation. Let’s briefly review our findings.
Changing patterns of cycles and crises
While industrial cycles of approximately ten years have been a remarkably persistent feature of capitalism since 1825, there have been periods when these cycles have been suppressed by world wars and other periods when we have had only partial cycles.
For example, the two world wars of the 20th century suppressed, to a considerable degree, the entire process of expanded capitalist reproduction. Since industrial cycles can arise only within the broader process of expanded capitalist reproduction, wartime suppression of expanded capitalist reproduction suppresses the industrial cycle.
We also find that a war economy cannot be the solution to the industrial cycle because capitalism can only exist as the expanded reproduction of capital. In fact, the temporary suppression of industrial cycles by the war economy demonstrates that the only way to end the industrial cycle of boom and bust is to end capitalist production.
Immediately after the super-crisis of 1929-33, itself part of the aftermath of the World War I war economy, there was no complete industrial cycle. The brutal deflationary policy of the Roosevelt administration in 1936-37 prevented the cyclical recovery of 1933-37 from developing into a real boom. Instead, the war economy of World War II replaced the recovery that followed the 1937-38 recession before it could develop into a boom. Therefore, we saw only partial industrial cycles from the super-crisis of 1929-33 until after World War II.
No full industrial cycle between 1968 and 1982
There was also no complete industrial cycle between 1968 and the beginning of the Volcker shock in 1979. During the 1970 and 1974-75 recessions, governments and central banks attempted to force recoveries through deficit spending and monetary expansion. Under the prevailing conditions, these repeated attempts to force a recovery simply led to panicky flights from the dollar and paper currencies in general into gold, causing the recoveries to abort. Full industrial cycles of a more or less 10-year duration reappeared only after the Volcker shock of 1979-82.
Material basis for 10-year industrial cycle
Marx, Keynes, and Schumpeter all agreed that periodic replacement of the elements of fixed capital, especially factory machinery, forms the material basis of the 10-year industrial cycle. About every ten years, there is a wave of renewal and expansion of fixed capital as new machines replace old machines in existing factories, and new factories filled with state-of-the-art machines are constructed. The resulting over-investment inevitably ends with the overproduction of commodities, leading to flooded markets.
Once the crisis breaks out, the formation of new fixed capital falls off rapidly. Even after the rundown of inventories leads to inventory restocking, there is still a large amount of idle, overproduced fixed capital left over from the last boom. However, about ten years after the last boom, the need to renew a significant amount of now aging factory machinery kicks off a new boom that soon leads to over-investment in fixed capital, general overproduction of commodities, and crisis.
Material basis for short cycle
The Kitchin inventory cycle, which is less pronounced, is also widely accepted among economists who study business cycles. Marx himself made occasional references to this cycle. Joseph Schumpeter called these cycles Kitchin cycles in his 1939 book Business Cycles after the statistician and South African mine owner Joseph Kitchin (1861-1932), who described them.
Since Kitchin, these inventory cycles have been widely accepted among bourgeois business-cycle experts. About every three or four years, inventories are overproduced, leading to a minor recession or slowdown, what the business press calls an “inventory adjustment.” A major recession occurs when a downturn in the Kitchin cycle coincides with a peak in the 10-year investment cycle of fixed capital.
These two cycles are called cycles rather than mere waves because each phase of these cycles necessarily leads to the next phase. While accidental factors certainly affect these cycles, and major wars can temporarily suppress them altogether, these cycles are not accidents but the necessary consequence of the capitalist mode of production.
Is there a 50-year cycle?
This takes us to the proposed 50-year cycle, often called the Kondratiev cycle after the Russian economist Nikolai Kondratiev, who described them in the 1920s. For it to be a true cycle, each stage of the 50-year cycle would necessarily lead to the next phase. Or do these cycles reflect accidental causes? If the latter is the case, they are not cycles at all. Assuming that long cycles exist, we have to explain the material basis of such a cycle.
One explanation proposed by Kondratiev was that the 50-year cycle reflects the lifetime of very long-lived elements of fixed capital. Not all elements of fixed capital are replaced every ten years. It is not hard to find machines that are considerably older than ten years and still used in factories. Buildings and bridges certainly last much longer than ten years. In this case, the material basis of the long cycle would be similar to that of the 10-year industrial cycle. This theory has, however, won very little support among any school of economists, whether Marxist or bourgeois.
In his Business Cycles, Joseph Schumpeter suggested that the 50-year cycle arises from a complex pattern of innovations. A wave of major innovations eventually peters out into a wave of minor innovations. The result is great waves of investment every 50 years that gradually taper off, leading to the depression phase of the Kondratiev cycle. Eventually, the depression in the cycle will prepare the way for a new phase of entrepreneurial innovations and a new upswing in the Kondratiev cycle. Schumpeter proposed in Business Cycles that the super-crisis of 1929-33 was caused, at least in part, by a trough in all three cycles that occurred in the early 1930s.
Among the economists who came to reject the long-cycle theory was the well-known U.S. Marxist — or Keynesian Marxist, as we can describe him — Paul Sweezy. During the Depression and Sweezy’s formative years as an economist, many economists influenced by Keynes believed that a permanent tendency toward stagnation marked a “mature” capitalist economy.
Schumpeter, who rejected this view in favor of his view that the stagnation of the Depression was largely cyclical and therefore temporary, dubbed these economists stagnationists. Sweezy accepted this label and considered himself a stagnationist until the end of his days.
Sweezy — influenced by Marx, Keynes, and Schumpeter — was primarily interested in the question of stagnation and monopoly throughout his long life as an economist. Sweezy assumed that a strong tendency toward expansion had marked the freely competitive capitalism of the Smith, Ricardo, and Marx era. Twentieth-century capitalism, however, was dominated by monopoly and, therefore, obeyed quite different economic laws. Unlike freely competitive capitalism, which generates economic growth, monopoly capitalism generates stagnation.
Why no permanent Great Depression under monopoly capitalism?
But hadn’t Sweezy explained too much? Sweezy believed that the Great Depression was the natural condition of monopoly capitalism when no offsetting factors were in operation. Since the Depression was, according to him, the normal state of monopoly capitalism, it did not need any particular explanation. If Sweezy’s theory was correct, what needed explanation was, why didn’t the Depression span the entire history of monopoly capitalism?
Sweezy concluded that innovations were one of the factors along with massive military expenditures and other unproductive expenditures, especially after World War II, that, for a considerable period, could override the stagnationist tendency of monopoly capitalism.
According to Sweezy, one great early 20th-century innovation that saved monopoly capitalism from stagnation was the automobile. Sweezy believed that the automobile had been decisive in saving U.S. monopoly capitalism from stagnation during the 1920s and had played a major role, along with government — primarily military — spending, in staving off stagnation after World War II.
The automobile not only meant that industrial capitalists like Henry Ford had to invest huge amounts of capital in auto production, but it also required the development of a vast infrastructure of highways and gasoline stations. The rise of the automobile generated a soaring demand for gasoline, providing markets for newly drilled oil and refineries to distill and convert the crude oil into gasoline.
In addition, the automobile made modern suburbanization possible, where workers live a considerable distance from their workplaces, whether factories or offices. Suburbanization meant that a considerable amount of housing and infrastructure had to be built to serve the needs of the growing number of people living in the suburbs many miles (kilometers) from their workplaces.
The construction industry not only had to build homes in the suburbs, it also had to construct huge suburban shopping malls. This stimulated the building industry and reacted in turn on the concrete, steel, and many other industries. All this created a huge wave of investment that, according to Sweezy, powerfully counteracted the tendency toward stagnation both before and after the Depression.
Here, we see Schumpeter’s influence on Sweezy. Where Sweezy differed from Schumpeter, however, was that Sweezy couldn’t see anything cyclical about the rise of the automobile industry. It was, according to Sweezy, a historical accident. There might or might not be a similar new innovative industry on the scale of the automobile industry in the future. According to Sweezy, nothing in the laws of motion of monopoly capitalism makes the emergence of new industries on the scale of the automobile inevitable.
Trotsky versus long-cycle theory
A very different Marxist, Leon Trotsky, also rejected the theory of long cycles. In an article partially aimed at counteracting the influence of Kondratiev, written in 1923, entitled The Long Curve of Capitalist Development, Trotsky agreed and indeed strongly emphasized that the concrete history of capitalism showed alternating periods of rapid economic growth and stagnation. But Trotsky denied there was anything cyclical in the succession of long periods of rapid economic growth under capitalism followed by periods of stagnation.
Instead, Trotsky suggested that the succession of rapid economic growth and stagnation epochs could only be explained by factors external to the capitalist system. For example, he mentioned the opening of new continents and natural resources, revolutions and counterrevolutions, and world wars as the factors responsible for the succeeding epochs of rapid capitalist growth and capitalist economic stagnation.
Ernest Mandel and long-cycle theory
Beginning in the 1960s, Trotsky’s avowed follower Ernest Mandel was far more sympathetic to the concept of a long cycle. As the leader of perhaps the largest faction of avowed Trotskyists, as they called themselves, Mandel was embarrassed by Trotsky’s rejection of the long-cycle theory, since this opened him to charges by members of rival Trotskyist groups that he was not a real Trotskyist.
Therefore, Mandel, retreating a bit, preferred the term long waves. But what factor would, according to Mandel, cause capitalist economic development — expanded capitalist reproduction — to be broken up into long waves lasting considerably longer than the 10-year industrial cycles?
Mandel saw the succession of periods of rapid capitalist economic growth and stagnation or semi-stagnation as reflecting long-term changes in the value rate of profit. By value rate of profit, as opposed to the market rate of profit, we mean what the rate of profit would be if all value and surplus value is realized and commodities sold at their value or direct price and the turnover of variable capital is as rapid as the technological conditions of production, including transportation, allow. What Mandel had to explain is the cause of the fluctuations in the value rate over periods considerably longer than the 10-year industrial cycles.
Like many Marxists, Mandel was fascinated by Marx’s law of the tendency of the rate of profit to decline. In Volume III of Capital, Marx described this law as the most important law of political economy.
Marx’s law of the tendency of the rate of profit to fall
Long before the time of Marx, political economists had assumed that the long-term tendency of the rate of profit was downward. This was assumed even by later schools of economics, including the original marginalists and Keynes. Only after World War II did it become popular among bourgeois economists to deny the historically downward tendency of the rate of profit. However, the reasons that Marx gave for the tendency of the rate of profit to fall were completely different than those given by bourgeois economists, whether before or after him.
Marx’s explanation is an extension of his overall theory of value and surplus value. Suppose the level of labor productivity in all branches of production is unchanged. This will mean the value of commodities will be unchanged. If this is true, the value of such elements of capital as raw and auxiliary materials, machinery, and factory buildings is a mathematical constant. Marx called this kind of capital constant capital.
The other type of capital consists of the commodity labor power. Labor power alone not only produces value but a value greater than its own value. Workers not only produce a value that replaces the value of the capital they consume — the necessary labor — but a value above and beyond it, a surplus value, by performing unpaid surplus labor.
This was Marx’s greatest single discovery in economics, though he made many others. Behind the appearance of the exchange of equal quantities of labor, there is the same old phenomenon of unpaid surplus labor that marked chattel slavery, feudalism, and other forms of exploitative class society.
Unlike the value contained in raw and auxiliary materials, machinery, factory buildings, and so on, whose value is preserved in the reproduction process, variable capital creates additional value. The value represented by purchased labor power — variable capital — expands and, in mathematical terms, is, therefore, a variable, not a constant.
While classical economists such as Adam Smith and David Ricardo considered the primary division within capital to be between fixed and circulating capital, Marx, in contrast, discovered that the primary division within capital is between constant capital, which does not produce surplus value, and variable capital, which alone produces surplus value. Therefore, before the transformation of values — or direct prices — into prices of production, the rate of profit on constant capital is zero. Only variable capital will have a rate of profit of more than zero.
Marx then observed that in the course of capitalist development the ratio between constant capital and variable capital shifts in favor of constant capital. This causes what Marx called the organic composition of capital, the ratio between constant and variable capital, to rise. The real cause of the tendency of the rate of profit to fall, however, is hidden from the capitalists engaged in everyday business, as well as the vulgar economists, by the transformation of values into prices of production — which gives rise to the illusion that both constant capital and variable capital produce surplus value.
All things remaining equal, the rate of profit will fall as the organic composition of capital rises, though various counter-factors tend to work in the other direction. Most important of these is the ratio of unpaid to paid labor, what Marx called the rate of surplus value, or the rate of exploitation. With the rise of the productivity of labor which lowers the value of commodities that enter into the value of labor power, the rate of profit on the variable capital rises, partially offsetting the growth of the constant part of capital that yields no profit.
Another important counter-tendency is the cheapening of the elements of constant capital, which slows its rate of growth in value terms. Marx, therefore, distinguished the technological composition of capital from the organic composition of capital. The technological composition of capital rises much more rapidly than the organic composition of capital.
In addition, anything that increases the rate of turnover of variable capital, such as the opening of new markets or improved transportation methods, also increases the rate of profit over a given period, such as a year.
Because there are counteracting factors, the falling rate of profit is only a tendency, according to Marx. Counter-tendencies can override the tendency of the rate of profit to fall for a more or less prolonged period. Therefore, Marx’s law of the tendency of the rate of profit to fall does not rule out a rise in the rate of profit over considerable periods. Mandel built his theory of long waves on this foundation.
Mandel’s theory of semi-cycles
Marx expected the rate of profit to fall only over considerable periods of time and not from year to year. Mandel, as we saw in an earlier chapter, ended up with a semi-cyclical theory. Essentially, he held that over some 20 years — about two ordinary 10-year industrial cycles — the rising organic composition of capital leads to a fall in the value rate of profit sufficient to cause a considerable slowdown in the rate of accumulation of capital or economic growth.
This slowdown is not accidental, according to Mandel; it is a necessary result of the most basic laws that govern the capitalist system. But why doesn’t the stagnation or semi-stagnation simply last indefinitely or even gradually intensify as the organic composition continues to rise? Here, like Sweezy, Mandel turned to accidental factors that may or may not recur in the future.
According to Mandel, each expansionary long wave of capitalist development historically had a different cause. The only thing these causes had in common was that they all temporarily increased the rate of profit, overriding the long-term downward trend of the profit rate.
According to Mandel (and Marx and Engels), the gold discoveries in California and Australia led to an extremely rapid expansion of the capitalist market, in other words to a very rapid growth of monetarily effective demand. In the late 19th century, the rapid spread of colonialism caused the profit rate to increase. The fruit of this rise was the wave of accelerated capitalist economic growth that began in the 1890s and continued until 1913.
The last long wave of capitalist economic expansion, the one that occurred between 1948 and 1967, Mandel believed was rooted in a sharp rise in the rate of exploitation and profit caused by the huge defeats that the working class had suffered in the years that followed World War I. These included the rise of a series of fascist and military dictatorships, especially the fascist dictatorship of Adolf Hitler, which crushed the working class movement throughout continental Europe.
This led to a sharp rise in the rate of surplus value, the ratio of unpaid to paid labor. The sharp rise in the rate of profit on variable capital for several decades after World War II swamped the effect of the rise in the organic composition of capital that occurred during this period.
This explanation works better in explaining the postwar prosperity in West Germany, the rest of Western Europe, and Japan than in explaining the boom in the United States. The inter-war years — leaving aside the special case of the Russian Revolution — were a disaster for the European working classes and the Japanese working class, where a military dictatorship was imposed in the 1930s and 1940s.
But they were also years of great advance for the working class in the leading capitalist country of the epoch, the United States. The building of powerful industrial unions for the first time in basic industry, as well as winning unemployment insurance and social insurance, no matter how inadequate, by the U.S. working class under the New Deal certainly made it more difficult for the U.S. capitalists to raise the rate of profit by increasing the rate of surplus value within the United States. This is why the great majority of U.S. capitalists viewed Franklin Roosevelt with bitter hatred.
But it is true that after World War II economic growth was much faster in West Germany and the rest of Western Europe and Japan, except Britain, than it was in the United States. Therefore, we can concede that there is a strong element of truth in Mandel’s explanation regarding Western Europe and Japan.
Perhaps Mandel’s views were over-colored by the Western European experience. Mandel lived in Western Europe throughout his entire life. Economic growth during the first several decades after the war was indeed considerably slower in the United States and Britain, where the working class had not suffered under fascist or military dictatorship, than it was in the imperialist countries that had experienced such dictatorships after World War I.
After Mandel’s death in 1995, the British Marxist Bill Jefferies, a Mandel follower, pointed out in his doctoral thesis that the overthrow of the planned economies of the Soviet Union and Eastern Europe and the new opening of China to world capital after 1978 worked as powerful counteracting factors against the long-term tendency of the rate of profit to fall. Following the 1970s, the amount of labor power available for capitalist exploitation skyrocketed. Jefferies was correct on these points.
From the other side of the class divide, Alan Greenspan, the former chief of the U.S. Federal Reserve System, has emphasized the role of a vast increase in the quantity of cheap labor in causing what Greenspan saw as the great capitalist prosperity that began in the early 1980s and that had not ended when he wrote his memoir, The Age of Turbulence.
These events not only increased the rate of surplus value on average on a world scale, but with so much cheap labor available, the capitalists have had a greatly reduced incentive to replace living labor — variable capital — with dead labor — constant capital. If the planned economies of the Soviet Union and Eastern Europe had not been dismantled, and if China and Vietnam had followed the Soviet pattern of economic development, the organic composition of capital would have been considerably higher, the average rate of surplus value and the rate of profit on the world market would have been lower around the turn of the 21st century than it was.
The new capitalist-run factories that have sprung up in China since 1978 and other cheap labor countries tend to be far less automated than the highly automated factories found in the United States, Western Europe, Japan, and other relatively high-wage areas. Generally, the factories not being shut down in the United States, Western Europe, and Japan are highly automated or, in Marxist terminology, have a very high organic composition of capital. All this over decades has worked against the long-run tendency toward a higher organic composition of capital and a lower rate of profit.
Jefferies, a disciple of Mandel, is much more consistent and willing to face unpleasant political realities than his master was in the final years of his life. He drew the logical conclusion that since the 1980s, the capitalist economy has once again been in a period of accelerated economic growth, or, in Mandel’s terminology, in a “long wave with an undertone of expansion.”
But doesn’t the panic of 2007-09 refute Jefferies’ view in practice? It didn’t necessarily at first. True, Jefferies, going overboard, made the mistake of predicting that there would be no worldwide recession when the first signs of a developing crisis began to manifest themselves in global financial markets in the middle of 2007. However, other periods that economic historians consider “long waves” of economic prosperity have also seen sharp panics followed by worldwide, if brief, recessions.
For example, there is the case of the panic of 1857, which occurred in Marx’s lifetime, and the panic of 1907, which occurred during the prosperous years — admittedly near their end — that preceded World War I. If the 2007-09 crisis had been followed by a dramatic rebound as occurred in 1858 and 1908, it would not have necessarily meant an end to a long period of capitalist prosperity that Jefferies believed was underway.
However, the recovery from the crisis of 2007-09 proved very slow, particularly in light of the sharpness of the crisis. Europe even slipped back into recession in 2011. However, the general pattern of sub-par growth stubbornly persisted in most countries right down to the COVID shutdowns of 2020.
In light of these developments, Jefferies’ prediction that the crisis of 2007-09 would first not occur at all and then that it would turn out to be only a brief interruption in a period of rapid growth like the crisis of 1857-58 proved false.
Where did Jefferies go wrong?
What we may call the Mandel-Jefferies theory of long waves is based on problems that the capitalist class faces in producing a quantity of surplus value sufficient to maintain a value rate of profit high enough to sustain vigorous capitalist expanded reproduction. Mandel, who was a considerably more subtle economic thinker than Jefferies is, did note in his book Long Waves of Capitalist Development that the question of effective monetary demand — the realization of surplus value in the form of monetary profit — should also be worked into the theory of long waves. However, in his book Mandel concentrated only on the production of surplus value and the value rate of profit. And that was his mistake.
The evolution of the world capitalist economy between 2007 and 2020 indicates that a theory of either crises, stagnation, long cycles, or long waves that fail to treat the problem of markets as an independent variable is, at best, incomplete. It is not enough for the capitalists to produce the surplus value; they must also realize it on the market in the form of money.
Fluctuations in the production of gold
This brings us to what is perhaps the oldest explanation for long waves or long cycles: fluctuations in the production of money material — gold. According to Mandel, the Dutch Marxist J. van Gelderen — who Mandel credits along with Parvus as the founders of long-wave theory — noticed that rises in gold production preceded accelerated waves of growth of the capitalist system. Karl Kautsky, in his book The High Cost of Living, also noted the relationship between rising gold production and periods of accelerated waves of growth of the capitalist economy, as well as the relationship between prolonged depressions and depressed gold production.
More recent Marxists, however, have generally ignored the role of gold production when they discuss long-cycle or long-wave theories. Mandel played down the role of gold production in his book Long Waves, except for the 1848-1873 “expansionary wave.”
Modern Marxist economists, with few exceptions, agreed with most bourgeois economists on this question, and wrongly assumed that with the end of the international gold standard, the level of gold production no longer significantly affects the rate of economic growth. This leads to the view that any shortage of demand can be dealt with through government deficit spending and, if there is a shortage of money, by simply having the central bank print more money, up to “full employment.”
If crises occur anyway, they must arise either because a section of the capitalist class opposes Keynesian policies — usually assumed to be the financial or money capitalists, who fear the effects of otherwise beneficial “mild inflation” because it tends to erode their incomes in real terms — or because an insufficient amount of surplus value is being produced to sustain an adequate value rate of profit.
In reality, the concrete data on the history of gold production from the 1848-51 gold discoveries onward show that periods of extraordinary economic crisis-depression have been preceded by declining gold production, just as eras of rapid capitalist economic growth have been preceded or accompanied by rising gold production.
This is true for the years examined by such early 20th-century Marxists as van Gelderen and Kautsky but also for the following years and decades. For example, a major drop in gold production during and immediately after World War I preceded the super-crisis of 1929-33. The rate of increase of gold production also began to stagnate immediately before the stagflation of the 1970s. The inflationary economic crisis of 1974-75 was preceded by declining gold production, and gold production remained stagnant at a depressed level until the economic crisis of 1979-82, after which it recovered.
This pattern has been repeated in the 2007-09 Great Recession. Between the early 1980s and the turn of the century, the trend in gold production was strongly upward, and after the turn of the century, it began to decline. Within less than a decade the economic crisis of 2007-09 hit, though after that crisis, gold production resumed its rise and hit new highs.
Therefore, both our theoretical analysis and the concrete economic history since 1850 agree that the long-run rate of growth of the market is tied not to the rate of growth of commodity production, as Say’s Law would have it, or to the fiscal and monetary policies followed by governments and central banks as Keynesians and Keynesian-Marxists believe, but to the rate of growth of the quantity of the money material — gold. This is true whether a gold standard in any form is in effect or a system of paper money prevails.
Contrary to the hopes and expectations of bourgeois economists, the end of the gold standard has not freed the growth of the market from its golden chains. This is shown by the economic “stagflationary” crisis of 1968-82 and was confirmed again by the 2007-09 world crisis. The level of gold production provides the missing piece in Mandel’s analysis of long waves: the role of the growth or lack of growth of monetarily effective demand — the market.
Mandel, unlike most modern Marxists who either claim that gold has lost its monetary role with the end of the international gold standard or simply ignore the question altogether, specifically affirmed that gold continues to retain its monetary role and continues to function as the universal equivalent or measure of the value of commodities. But except for the 1848-1873 long wave of expansion that occurred when the gold standard was in effect in the most important capitalist country, Britain, as well as the United States, Mandel played down the importance of the level of gold production on the rate of economic growth. The price he paid left a large hole in his theory of long waves that he himself acknowledged, the role of monetarily effective demand.
Is gold production cyclical?
The level of gold production also provides a potential cyclical element. Remember, the lack of any perceived cyclical mechanism caused both Trotsky and Sweezy to reject the existence of long cycles in capitalist production.
Suppose we have a situation where the general price level is below the labor values of commodities — or, more strictly, market prices are below the prices of production of commodities — such as is generally the case after a major economic crisis or prolonged deep depression. This will mean that the rate of profit in the gold mining and refining industry will be above the average rate of profit that prevails in capitalist industry as a whole.
Capital, which is always in search of super-profits above and beyond the average rate of profit, will then flow into gold production as long as this situation exists. Gold production will be stimulated, and considerable amounts of idle money capital will accumulate in the banks for years. These growing hoards of idle money capital will have the effect of driving down the rate of interest and increasing the profit of enterprise which alone provides the motive to actually produce surplus value.
The industrial capitalists will have to increase their investments sooner or later because if they don’t, more and more of their capital will consist of idle money capital, and the rate of profit will steadily decline on their total capital, including money capital. As the industrial capitalists step up their investments, more and more workers will be exploited, and the production of surplus value rises and the possibilities of actually realizing the surplus value in money form expand as previously idle hoards of money capital are drawn into circulation. The market then experiences one of its sudden expansions that mark the concrete history of capitalist production.
The “boom phase” of this “long cycle” will come to dominate perhaps several 10-year industrial cycles. During these “expansionary” economic cycles, the boom phase lasts much longer than the combined crisis-depression phase. During booms, demand exceeds the supply of commodities at current market prices. Prices, therefore, drift upward as the market equalizes supply and demand.
Eventually, prices rise above their underlying labor values and the prices of production that are determined with some modifications by labor values. This will mean the rate of profit in the gold mining industry will fall below the average rate of profit in industry as a whole as well as absolutely. Capital will begin to flow out of gold mining and refining into other more profitable industries. The rate of increase in gold production will at first decline, and eventually, the level of gold production will begin to fall absolutely.
Declining gold production, whatever monetary system is in effect, whether a gold standard, gold-exchange standard, or a “paper money” system, means that as long as the rate of economic growth remains unchanged, the rate of interest will rise relative to the total profit, causing a drop in the profit of enterprise.
Money will become tight sooner or later, and eventually, a major or series of economic crises will break out. Prices in terms of gold will decline. At some point, prices will again fall below the prices of production, causing the rate of profit in gold mining and refining to once again exceed the average rate of profit in other industries as well as increase absolutely. Capital will again flow into gold mining and refining, and the cycle will repeat. Here we have a true cyclical mechanism where one phase of the cycle leads to the next phase of the cycle.
This cyclical mechanism is rooted in the most basic law that regulates the capitalist economy, the law of value of commodities. Long cycles, if the hypothesis that we are exploring here is correct, function as the mechanism by which the level of market prices is kept in the long run more or less in line with production prices — themselves determined by the underlying labor values — in the long run.
The long cycle would then be a necessary consequence of capitalist production. Our hypothesis also provides a material basis for the long cycle. Just like the 10-year industrial cycle involves the decennial renewal of fixed capital, and the short Kitchin cycle involves the three-year cycle of commodity capital, the long cycle would involve the cyclical movement of the production of and accumulation of money capital. Unlike Mandel’s theory, both the upper and lower turning points of the cycle are explained. Instead of a semi-cycle, we have a full cycle.
This hypothesis depends on the assumption that the level of gold production is governed by a cycle with at least a 20-year up phase. But the above suggestion depends on the cyclical nature of gold production, and how cyclical is gold production? The historical data show that the level of gold production is determined only partially by cyclical forces. Many non-cyclical factors play an important role.
For example, the discovery of gold in California and Australia in 1848-51 was not cyclical. Therefore, insofar as the rise in world gold production caused the mid-Victorian boom of 1848-1873 — and even Mandel agrees that it was — to that extent, the “boom” wasn’t cyclical either.
Similarly, the development of the cyanide process that made it possible to extract minute quantities of gold from very poor ores was not strictly cyclical either — though here the case could be made that high profits in gold production encouraged high levels of investment and therefore innovation in gold production. The cyanide process’s widespread application in gold mining and refining starting in the 1890s may have indirectly, therefore, had a cyclical element. However, just as the California and Australian gold discoveries were not cyclical, neither were the gold discoveries in the mid-1890s in the Klondike.
As explained in earlier chapters, the sharp and prolonged decline in the level of gold production during and immediately after World War I was linked to the inflation of prices, not just in terms of depreciated paper money but in gold money as well, and was therefore not cyclical either. It was the result of the World War I war economy.
However, the rate of increase in gold production had already slowed to a crawl before the war began, so perhaps World War I only aggravated a crisis in gold production that was already developing. The growing stagnation of gold production in the years immediately preceding World War I, following sharp increases in commodity prices, contained a strong cyclical element though the depletion of the Klondike gold mines was not itself cyclical and neither was the depletion of the South African gold mines at the turn of the 21st century, which preceded the crisis of 2007-09.
To reach the gold deposits, the gold mining companies were forced to drill ever deeper, straining the limits of late 20th- and early 21st-century technology in this regard — not to mention forcing the mineworkers to labor in increasingly hot conditions, straining the limits of human physiology. The depletion of the South African gold mines and the deteriorating natural conditions of production that played no small role in the coming of the Great Recession were not cyclical factors.
Because non-cyclical factors so strongly influence gold production — the discoveries of new mines, the exhaustion of old mines, inflations caused by wars, not only World Wars I and II but smaller wars such as Vietnam, as well as the progress in mining and refining technology — we would not expect to see regular long cycles in the capitalist economy.
The rate of growth of the capitalist economy responds to changes in the level of gold production, whether cyclical or non-cyclical. Therefore, changes in the rate of growth of the capitalist economy tend to have an irregular character and will be far more difficult to predict than the 10-year industrial or three-year inventory cycles of capitalism. The underlying cyclical forces that tend to produce the long cycles that do exist will be greatly distorted or even swamped by non-cyclical forces sometimes working to reinforce and sometimes crossing and sometimes negating them.
For example, would the Great Depression of the 1930s have occurred if there had been discoveries of new, very rich gold deposits on the relative scale of 1848 and the following years during the 1920s? Not likely, in our opinion. Perhaps in that counter-historical case, there would have been the usual cyclical recession around 1929-30 or a little later, with rapid economic growth resuming within a year as happened in 1858. The fact that nothing like this happened is to a certain extent a historical “accident.”
More regular “long cycles” in the 19th century?
During the 19th century, there were 50-year-long swings in prices that did seem to be somewhat regular. This, perhaps, is what inspired Kondratiev to propose his 50-year cycle in the first place. For example, prices trended downward from 1815, the end of the world war that followed the French Revolution, to about 1843. This would represent a declining phase of about 28 years.
Between 1843 and 1873, prices generally rose, with the uptrend greatly strengthened by the 1848-1851 gold discoveries in California and Australia. This would represent an upward trend of 30 years. Including the upward and downward phases, this comes to a 58-year cycle.
Between the price peak of 1873 and the trough of 1896, there were 23 years. The next price peak came in 1920, though World War I greatly influenced this price peak. Subtracting 1896 from 1920 gives us 24 years. Adding 23 and 24 together, we get a 47-year cycle — close to 50 years.
After this, the cycles become less regular. Between 1920 and 1933-34, the trend of prices was strongly downward. However, we get only 13 years of declining dollar prices and 14 years if we use prices measured in gold instead. Between 1933 and 2013, prices in terms of U.S. dollars and other paper currencies rose continuously with only brief and limited interruptions in 1937-38, 1948-49, and 2008-09. Prices have not been cyclical at all in terms of dollar prices and other paper money. Any long cycle or even long wave that existed before 1933 would seem to be no longer visible if we look only at prices in terms of paper dollars.
But if we look at prices measured in terms of gold, we see a different picture. Taking the end of Roosevelt’s 1933-34 devaluation of the dollar as the price trough, we have an up phase, from 1933 to 1970, of 36 years. Adding the down and up phases together gives us exactly 50 years — 1920 to 1970 — for the entire cycle. The down phase of the cycle was compressed by the extreme violence of the super-crisis of 1929-33.
Between 1970 and 1980, prices in terms of gold dropped sharply. Again, the decline in the prices of commodities measured in terms of gold as opposed to paper currencies was very short but also very violent. They recovered after 1980 until 2001. After 2001, as the dollar resumed its depreciation, prices in terms of gold began a downward trend. We get, therefore, a 10-year down phase (1970 to 1980) and a 21-year up phase (1980 to 2001), adding up to a total of 31 years for the latest full cycle, well short of a full 50-year Kondratiev cycle.
If the 50-year cycle were in operation, and taking 1970 as a peak year and the up and down phases of the cycle were of equal lengths, we would expect a price trough in terms of gold around 1995 and the next peak around 2020. Instead, we get a price trough in terms of gold around 1980, 15 years early, and a new, if considerably lower, price peak in 2001.
Therefore, since 1920, instead of regular cycles with more or less equal up and down phases, we seem to have a cycle of 20- to 25-year up phases followed by violent collapses of prices in terms of gold that last about a decade.
Perhaps the apparent regularity of the long cycle in the 19th- and early 20th-century price data was largely accidental. Or it might have reflected the greater stability of capitalism under the international gold standard in an era without world wars.
The price peaks of 1815 and 1920 seem to have been influenced by world wars and therefore were not completely cyclical. The mid-Victorian boom was tied to the cheapening of gold in the wake of discoveries of the rich mines in California and Australia, an event that was also non-cyclical.
Even the stagnation of gold production after the mid-1960s is related to the price inflation of the late 1960s, partly caused by the Vietnam War. Here, however, it seems that the Vietnam War only accelerated a cyclical decline in gold production that was coming anyway, just as apartheid, also a non-cyclical factor, postponed both the peak of gold production until 1970 as well as the collapse of commodity prices measured in terms of gold that quickly followed.
There appears to be an underlying cyclical pattern here, but clearly non-cyclical factors are playing an important role, at least regarding gold production.
Are the theories of long-term changes in the rate of growth of the capitalist economy caused by changes in gold production, on the one hand, and the changes in the long-term rate of capitalist economic growth tied to the rise in the organic composition of capital and fluctuations in the rate of surplus value, on the other, mutually exclusive? Or can they be combined?
It is easy to document changes in gold production, with fairly reliable figures on global gold production available from the 1850s onward. It is much harder to document changes in the organic composition of capital, rate of surplus value and that part of the turnover time that is determined by the technological state of production and transportation — as opposed to the state of business and changes in prices. This makes it hard to document changes in the value rate of profit and, therefore, empirically explore how changes in the value rate of profit exert an independent effect on the long-term evolution of the rate of capitalist economic growth and markets.
Various Marxists have attempted to measure changes in the rate of profit over certain periods. However, this data lacks the reliability or completeness of the statistical data available for world gold production. Charting world gold production trends, compared to the factors behind the value rate of profit, is, after all, simple. However, as we see, changes in gold production are a pretty good indicator of changes in business conditions and prices — in terms of gold — over periods that exceed the 10-year industrial cycle.
A combined theory
Despite the problem of inadequate data on the organic composition of capital, the rate of surplus value, and the technology-determined rate of turnover of variable capital — the value rate of profit — we believe the theory that links long waves to long-term fluctuations in the value rate of profit and the theory that links them to long-term fluctuations in the level of gold production, and thus in the rate of growth of the market, can be combined.
Marx pointed out that the optimum conditions for producing surplus value and realizing surplus value only sometimes coincide. If surplus value is not produced, there is no profit. If the surplus value produced is insufficient to generate high enough profits, there will be economic stagnation.
On the other hand, if the surplus value is produced in quantities more than adequate to support rapid capitalist economic growth, there is no guarantee that the surplus value will be realized in terms of money. And if it is not realized in terms of money, there will be no profit, or if partially realized, perhaps a rate of profit that is too low to support rapid economic growth. Indeed, the conditions that favor the realization of surplus value tend, over the long run, to undermine the production of surplus value.
As a thought experiment, let’s assume the problem of realizing surplus value could be permanently solved under the capitalist mode of production. The market would no longer limit the scale of capitalist production. Production could increase as rapidly as allowed by the value rate of profit. Crises of general overproduction would no longer get in the way.
Would this solve the contradictions of capitalism? The problem of the realization of the value and surplus value of commodities in our Say’s Law world would by definition be solved. But what about the problem of producing surplus value?
The demand for labor power, no longer periodically checked by crises of overproduction, would rise without interruption. With the demand for labor power high and growing, it would only be a matter of time before the balance of forces on the world labor market would shift in favor of the sellers of the commodity labor power. Wages would rise, which, of course, would be very good for the working class.
How would the industrial capitalists react to such a deteriorating situation for them? They would introduce machinery at an accelerated rate to hold down wages. But, introducing machinery at an accelerated rate would mean a sharp rise in the organic composition of capital. The rate of profit would come under great pressure. The combination of a falling rate of surplus value and a rising organic composition of capital would squeeze the rate of profit at two ends at once.
Overcoming the problem of realizing the surplus value would progressively undermine the conditions of producing it. The end result would be a crisis caused by what Marx called an absolute overproduction of capital and not a relative overproduction of commodities — which, also of necessity, means a relative overproduction of capital. Therefore, even if the problem of “demand” — the market — could be solved, we would still not have a crisis-free capitalism.
Now, as a thought experiment, assume the opposite situation that could arise in the real world. Suppose due to exhaustion of gold mines, the problem of the realization of surplus value is greatly intensified. The demand for labor power would now grow very slowly. On the labor market, the balance of forces would strongly favor the buyers of labor power — the capitalists — over the sellers — the working class. Wages would fall more and more over time. The rate of surplus value would skyrocket. It would become vastly easier to produce surplus value.
With labor so cheap, the capitalists would increasingly favor labor-intensive methods of production over capital-intensive methods, as the bourgeois economists say. The organic composition would rise very little if it didn’t fall.
In summary, easing the difficulties of realizing surplus value intensifies the problem of producing it. On the other hand, intensifying the problem of realizing surplus value eases the problem of producing it.
Now assume that we have alternating periods of rising gold production, which makes the realization of surplus value relatively easy but the production of surplus value increasingly difficult, and periods of declining gold production, which makes the realization of surplus value much more difficult but makes the production of surplus value easier.
Constructing an ideal long cycle
Let’s begin with the up phase of the cycle. Gold production is expanding, and monetarily effective demand is rising rapidly. The demand for the commodity labor power rises sharply. At first, there is plenty of unemployment and surplus labor. For a while, the production of surplus value has been quite easy for industrial capitalists. The rate of profit will be high, and business will boom. Production of commodities occasionally outraces gold production, but a brief recession quickly corrects the situation, and the economy recovers rapidly.
As the boom continues, however, labor shortages begin to appear, first among skilled workers but later among even unskilled workers. Wages rise, and more importantly, the rate of surplus value falls. As the production of surplus value becomes increasingly difficult, the industrial capitalists react by replacing living labor with machinery.
This temporarily increases the rate of profit for individual capitalists before prices adjust to new lower labor values or prices of production. But in the long run, the pressure on the rate of profit gets even more intense as the organic composition of capital rises. More and more of the total capital consists of constant capital, which produces no surplus value no matter how productive it is in physical terms. Less and less of the total capital consists of variable capital, which alone produces surplus value.
Under our assumptions, the remaining variable capital per worker unit of purchased labor power produces less surplus value. The total mass of surplus value is increasing at a lesser rate than the total value produced by the working class. With the quantity of unemployed labor power increasingly scarce, it is much harder to increase industrial production rapidly. Installing more powerful machines or constructing extremely capital-intensive factories takes time, especially if construction labor is scarce.
With demand exceeding supply at existing prices, prices start rising rapidly relative to their labor value-determined production prices. The rate of profit falls faster in the gold mining and refining industries than it falls in capitalist industry as a whole. The production of gold levels out and starts to decline.
The capitalist system will face increased difficulty in the production of surplus value at the very moment it is becoming more difficult to realize surplus value. Soon, a major economic crisis or series of economic crises will develop.
The demand for labor power will decline dramatically. As unemployment again grows, wages are slashed. It has now become more profitable for industrial capitalists to favor the use of labor-intensive over capital-intensive production methods. The rise in the organic composition of capital is greatly reduced if it isn’t reversed.
The devaluation of the commodities that make up the elements of constant capital also works in the direction of lowering the organic composition of capital. Eventually, protracted depression conditions, or perhaps a shorter period of acute crisis, lowers the prices of commodities in terms of gold below the values of commodities once again.
This raises the rate of profit in the gold production industry once again above the average rate of profit in industry as a whole. At the bottom of the long cycle, a combination of the growing ease of producing surplus value coincides once again with the increased ease of realizing surplus value. In addition, the abundance of money capital means a lower rate of interest relative to the total profit and a sharply rising profit of enterprise. Economic growth speeds up once again, and the cycle or wave is repeated.
So we see here a mechanism for a cyclical or quasi-cyclical process operating through a series of 10-year industrial cycles. If we abstract external shocks, perhaps we will have two 10-year industrial cycles dominated by boom conditions, followed by two industrial cycles dominated by crisis and depression conditions. In the real world, this cycle will be partially broken up by such inevitable accidents as wars, revolutions, counterrevolutions, new gold discoveries, inventions in gold mining and refining, and the exhaustion of gold mines.
In addition, changes in the natural conditions of production involved in the production of the elements of constant capital, such as copper mines, nickel mines, coal mines, oil wells, and so on, also will have an important effect on the organic composition of capital and the rate of profit.
The same is true in agricultural production. Indeed, changes in agricultural production can affect both the organic composition of capital and the rate of surplus value. These are among the accidental factors that can either reinforce or cross the long cycle described above.
Sometimes, these accidental processes will reinforce and intensify the underlying cyclical trend. At other times, such accidents will cross the underlying cyclical trend and weaken or override it. This seems to coincide fairly well with what we see in the economic history of the last 170 years.
With the complexities of real-world cycles, how long can capitalism go on?
Could capitalism, in the absence of a working-class revolution, last indefinitely? If that is the case, then socialism is not an economic necessity but merely a moral imperative.
Or are there certain economic or historical limits to the process that must eventually end capitalist production? And if this is true, exactly what are the factors that must bring capitalism to an end?
Strictly speaking, this takes us beyond crisis theory, the subject of this work. The section of Marxist theory that examines the questions of the historical and economic limits to capitalist production is sometimes called “breakdown theory,” though there are many interpretations of what “breakdown” means in this context.
But, the question of capitalism’s ultimate economic limitations is so intertwined with crisis theory that it cannot be avoided. Crisis theory, therefore, “blends” into “breakdown theory.” Or what comes to the same thing, crisis theory blends into the question of the fate of modern society.
We will explore this question in the final section of this work.
A Critique of Crisis Theory 