Other problems for capitalism arise due to increases in productivity which occur as a result of capital investment or new working practices which aim to increase short term profits for the company. The need to maximise profits results in more and more investment in order to improve the productivity of the workforce (i.e. to increase the amount of surplus value produced). A rise in productivity, however, means that whatever profit is produced is spread over an increasing number of commodities. This profit still needs to be realised on the market but this may prove difficult as capitalists produce not for existing markets but for expected ones. As individual firms cannot predict what their competitors will do, it is rational for them to try to maximise their market share by increasing production (by increasing investment). As the market does not provide the necessary information to co-ordinate their actions, this leads to supply exceeding demand and difficulties realising the profits contained in the produced commodities. In other words, a period of over-production occurs due to the over-accumulation of capital.

Due to the increased investment in the means of production, variable capital (labour) uses a larger and larger constant capital (the means of production). As labour is the source of surplus value, this means that in the short term profits must be increased by the new investment, i.e. workers must produce more, in relative terms, than before so reducing a firms production costs for the commodities or services it produces. This allows increased profits to be realised at the current market price (which reflects the old costs of production). Exploitation of labour must increase in order for the return on total (i.e. constant and variable) capital to increase or, at worse, remain constant.

However, while this is rational for one company, it is not rational when all firms do it, which they must in order to remain in business. As investment increases, the surplus value workers have to produce must increase faster. If the mass of available profits in the economy is too small compared to the total capital invested then any problems a company faces in making profits in a specific market due to a localised slump caused by the price mechanism may spread to affect the whole economy. In other words, a fall in the rate of profit (the ratio of profit to investment in capital and labour) in the economy as a whole could result in already produced surplus value, earmarked for the expansion of capital, remaining in its money form and thus failing to act as capital. No new investments are made, goods cannot be sold resulting in a general reduction of production and so increased unemployment as companies fire workers or go out of business. This removes more and more constant capital from the economy, increasing unemployment which forces those with jobs to work harder, for longer so allowing the mass of profits produced to be increased, resulting (eventually) in an increase in the rate of profit. Once profit rates are high enough, capitalists have the incentive to make new investments and slump turns to boom.

It could be argued that such an analysis is flawed as no company would invest in machinery if it would reduce it's rate of profit. But such an objection is flawed, simply because (as we noted) such investment is perfectly sensible (indeed, a necessity) for a specific firm. By investing they gain (potentially) an edge in the market and so increased profits. Unfortunately, while this is individually sensible, collectively it is not as the net result of these individual acts is over-investment in the economy as a whole. Unlike the model of perfect competition, in a real economy capitalists have no way of knowing the future, and so the results of their own actions, nevermind the actions of their competitors. Thus over-accumulation of capital is the natural result of competition simply because it is individually rational and the future is unknowable. Both of these factors ensure that firms act as they do, investing in machinery which, in the end, will result in a crisis of over-accumulation.

Cycles of prosperity, followed by over-production and then depression are the natural result of capitalism. Over-production is the result of over-accumulation, and over-accumulation occurs because of the need to maximise short-term profits in order to stay in business. So while the crisis appears as a glut of commodities on the market, as there are more commodities in circulation that can be purchased by the aggregate demand ("Property sells products to the labourer for more than it pays him for them," to use Proudhon's words), its roots are deeper. It lies in the nature of capitalist production itself.

A classic example of these "objective" pressures on capitalism is the "Roaring Twenties" that preceded the Great Depression of the 1930s. After the 1921 slump, there was a rapid rise in investment in the USA with investment nearly doubling between 1919 and 1927.

Because of this investment in capital equipment, manufacturing production grew by 8.0% per annum between 1919 and 1929 and labour productivity grew by an annual rate of 5.6% (this is including the slump of 1921-22). This increase in productivity was reflected in the fact that over the post-1922 boom, the share of manufacturing income paid in salaries rose from 17% to 18.3% and the share to capital rose from 25.5% to 29.1%. Managerial salaries rose by 21.9% and firm surplus by 62.6% between 1920 and 1929. With costs falling and prices comparatively stable, profits increased which in turn lead to high levels of capital investment (the production of capital goods increased at an average annual rate of 6.4%).

Unsurprisingly, in such circumstances, in the 1920s prosperity was concentrated at the top 60% of families made less than $2000 a year, 42% less than $1000. One-tenth of the top 1% of families received as much income as the bottom 42% and only 2.3% of the population enjoyed incomes over $10000. While the richest 1% owned 40% of the nation's wealth by 1929 (and the number of people claiming half-million dollar incomes rose from 156 in 1920 to 1489 in 1929) the bottom 93% of the population experienced a 4% drop in real disposable per-capita income between 1923 and 1929.

However, in spite of this, US capitalism was booming and the laissez-faire capitalism was at its peak. But by 1929 all this had changed with the stock market crashing — followed by a deep depression. What was its cause? Given our analysis presented above, it may have been expected to have been caused by the "boom" decreasing unemployment, so increased working class power and leading to a profits squeeze, but this was not the case.

This slump was not the result of working class resistance, indeed the 1920s were marked by a labour market which remained continuously favourable to employers. This was for two reasons. Firstly, the "Palmer Raids" at the end of the 1910s saw the state root out radicals in the US labour movement and wider society. Secondly, the deep depression of 1920-21 (during which national unemployment rates averaged over 9%) combined with the use of legal injunctions by employers against work protests and the use of industrial spies to identify and sack union members made labour weak and so the influence and size of unions fell as workers were forced to sign "yellow-dog" contracts to keep their jobs.

During the post-1922 boom, this position did not change. The national 3.3% unemployment rate hid the fact that non-farm unemployment averaged 5.5% between 1923 and 1929. Across all industries, the growth of manufacturing output did not increase the demand for labour. Between 1919 and 1929, employment of production workers fell by 1% and non-production employment fell by about 6% (during the 1923 to 29 boom, production employment only increased by 2%, and non-production employment remained constant). This was due to the introduction of labour saving machinery and the rise in the capital stock. In addition, the high productivity associated with farming resulted in a flood of rural workers into the urban labour market.

Facing high unemployment, workers' quit rates fell due to fear of loosing jobs (particularly those workers with relatively higher wages and employment stability). This combined with the steady decline of the unions and the very low number of strikes (lowest since the early 1880s) indicates that labour was weak. Wages, like prices, were comparatively stable. Indeed, the share of total manufacturing income going to wages fell from 57.5% in 1923-24 to 52.6% in 1928/29 (between 1920 and 1929, it fell by 5.7%). It is interesting to note that even with a labour market favourable to employers for over 5 years, unemployment was still high. This suggests that the neo-classical "argument" that unemployment within capitalism is caused by strong unions or high real wages is somewhat flawed to say the least (see section C.9).

The key to understanding what happened lies the contradictory nature of capitalist production. The "boom" conditions were the result of capital investment, which increased productivity, thereby reducing costs and increasing profits. The large and increasing investment in capital goods was the principal device by which profits were spent. In addition, those sectors of the economy marked by big business (i.e. oligopoly, a market dominated by a few firms) placed pressures upon the more competitive ones. As big business, as usual, received a higher share of profits due to their market position (see section C.5), this lead to many firms in the more competitive sectors of the economy facing a profitability crisis during the 1920s.

The increase in investment, while directly squeezing profits in the more competitive sectors of the economy, also eventually caused the rate of profit to stagnate, and then fall, over the economy as a whole. While the mass of available profits in the economy grew, it eventually became too small compared to the total capital invested. Moreover, with the fall in the share of income going to labour and the rise of inequality, aggregate demand for goods could not keep up with production, leading to unsold goods (which is another way of expressing the process of over-investment leading to over-production, as over-production implies under-consumption and vice versa). As expected returns (profitability) on investments hesitated, a decline in investment demand occurred and so a slump began (rising predominantly from the capital stock rising faster than profits). Investment flattened out in 1928 and turned down in 1929. With the stagnation in investment, a great speculative orgy occurred in 1928 and 1929 in an attempt to enhance profitability. This unsurprisingly failed and in October 1929 the stock market crashed, paving the way for the Great Depression of the 1930s.

The crash of 1929 indicates the "objective" limits of capitalism. Even with a very weak position of labour, crisis still occurred and prosperity turned to "hard times." In contradiction to neo-classical economic theory, the events of the 1920s indicate that even if the capitalist assumption that labour is a commodity like all others is approximated in real life, capitalism is still subject to crisis (ironically, a militant union movement in the 1920s would have postponed crisis by shifting income from capital to labour, increasing aggregate demand, reducing investment and supporting the more competitive sectors of the economy!). Therefore, any neo-classical "blame labour" arguments for crisis (which were so popular in the 1930s and 1970s) only tells half the story (if that). Even if workers do act in a servile way to capitalist authority, capitalism will still be marked by boom and bust (as shown by the 1920s and 1980s).

To take another example, America's 100 largest firms, employing 5 million persons and having assets of $126 billion, saw their average amount of assets per worker grow from $12,200 in 1949 to $20,900 in 1959 and to $24,000 in 1962 [First National City Bank, Economic Letter, June 1963]. As can be seen, the rate of increase in average assets per worker falls off over time. The initial period of high capital formation was followed by a recessionary period between 1957 and 1961. These years were marked by a sharp increase in unemployment (from 3 million in 1956 to a high of 5 million in 1961) and a higher unemployment rate after the slump than before (an increase of 1 million from 1956 figures to around 4 million in 1962). [T. Brecher and T. Costello, Common Sense for Hard Times, chart 2]

We have referred to data from this period, because some supporters of "free market" capitalism have used the same period to argue for the advantages of capital investment. This data actually indicates, however, that increased capital formation helps to create the potential for recession, because although it increases productivity (and so profits) for a period, it reduces profit rates in the long run because there is a relative scarcity of surplus value in the economy (compared to invested capital). This fall in profit rates is indicated by the decrease in capital formation, which is the point of production in the first place within capitalism, as well as by the increase of unemployment during that period.

So, if the profit rate falls to a level that does not allow capital formation to continue, a slump sets in. This general slump is usually started by overproduction for a specific commodity, possibly caused by the process described in section C.7.2. If there are enough profits in the economy, localised slumps have a reduced tendency to grow and become general. A slump only becomes general when the rate of profit over the whole economy falls. A local slump spreads through the market because of the lack of information the market provides producers. When one industry over-produces, it cuts back production, introduces cost-cutting measures, fires workers and so on in order to try and realise more profits. This reduces demand for industries that supplied the affected industry and reduces general demand due to unemployment. The related industries now face over-production themselves and the natural response to the information supplied by the market is for individual companies to reduce production, fire workers, etc., which again leads to declining demand. This makes it even harder to realise profit on the market and leads to more cost cutting, deepening the crisis. While individually this is rational, collectively it is not and so soon all industries face the same problem. A local slump is propagated through the economy because the capitalist economy does not communicate enough information for producers to make rational decisions or co-ordinate their activities.

"Over-production," we should point out, exists only from the viewpoint of capital, not of the working class:

"What economists call over-production is but a production that is above the purchasing power of the worker. . . this sort of over-production remains fatally characteristic of the present capitalist production, because workers cannot buy with their salaries what they have produced and at the same time copiously nourish the swarm of idlers who live upon their work." [Peter Kropotkin, Op. Cit., pp. 127-128]

In other words, over-production and under-consumption reciprocally imply each other. There is no over production except in regard to a given level of solvent demand. There is no deficiency in demand except in relation to a given level of production. The goods "over-produced" may be required by consumers, but the market price is too low to generate a profit and so production must be reduced in order to artificially increase it. So, for example, the sight of food being destroyed while people go hungry is a common one in depression years.

So, while the crisis appears on the market as a "commodity glut" (i.e. as a reduction in effective demand) and is propagated through the economy by the price mechanism, its roots lie in production. Until such time as profit levels stabilise at an acceptable level, thus allowing renewed capital expansion, the slump will continue. The social costs of such cost cutting is yet another "externality," to be bothered with only if they threaten capitalists' power and wealth.

There are means, of course, by which capitalism can postpone (but not stop) a general crisis developing. Imperialism, by which markets are increased and profits are extracted from less developed countries and used to boost the imperialist countries profits, is one method ("The workman being unable to purchase with their wages the riches they are producing, industry must search for markets elsewhere" - Kropotkin, Op. Cit., p. 55). Another is state manipulation of credit and other economic factors (such as minimum wages, the incorporation of trades unions into the system, arms production, maintaining a "natural" rate of unemployment to keep workers on their toes etc.). Another is state spending to increase aggregate demand, which can increase consumption and so lessen the dangers of over-production. Or the rate of exploitation produced by the new investments can be high enough to counteract the increase in constant capital and keep the profit rate from falling. However, these have (objective and subjective) limits and can never succeed in stopping depressions from occurring.

Hence capitalism will suffer from a boom-and-bust cycle due to the above-mentioned objective pressures on profit production, even if we ignore the subjective revolt against authority by workers, explained earlier. In other words, even if the capitalist assumption that workers are not human beings but only "variable capital" was true, it would not mean that capitalism was a crisis free system. However, for most anarchists, such a discussion is somewhat academic for human beings are not commodities, the labour "market" is not like the iron market, and the subjective revolt against capitalist domination will exist as long as capitalism does.