WHAT KONDRATIEV SAID

Having established quite clearly, at least for the United States, that prices have accelerated and decelerated in a succession of long waves, it behooves us to take a closer look at what Nikolai Kondratiev (for whom the long waves are named)said, and how subsequent scholars have tried to explain the 50-year long wave.)


1. Two good treatments of Kondratiev can be found in Nathan H. Mager, The Kondratieff Waves (1987), and a four-part series by Ralph D. Cato that was published in Futures: The Magazine of Commodities, March-June 1986. Kondratiev wrote six pieces on long waves between 1920 and1928; all except the first and fourth were initially published in the Economic Bulletin of the Conjuncture Institute, which he edited. His book The World Economy and Its Condition during an dafter the War was published in 1922. "Some Controversial Questions Concerning the World economy and the Crisis" (1923) and "On the Notion of Economic Statics, Dynamics, and Fluctuations" (1924) appeared in the Economic Bulletin. "Long Business Cycles" was published in Problems of Economic Fluctuations (1925); "Major Economic Cycles" (1926a; published in 1928 with an extensive critique by D. l. Oparin) and "Dynamics of Industrial and Agricultural Prices" (1928a)were published in Economic Bulletin. Kondratiev's work is largely known via his 1925 article, published in abridged form as "Die langen Wellen der Konj unktu r," Arch iv fur SozialuJissenschaft undSozialpolitik 56 (1926b), and his 1928 article, published as "Die Preisdynamik der industriellen undlandwirtschaftlichen Waren," Archiv fur Sozialwissenschaft und Sozialpolitik 60 (1928b). Anabridged version of the 1926 article was translated into English by W. F. Stolper and published as Kondratiev was born in 1892. He was sufficiently precocious that after joining the Chayanov Institute in 1916 he became vice-minister for food in Alexander Kerensky's provisional 1917 government. His mentor was Mikhail 1.Tugan-Baranovskij, Russia's greatest prerevolutionary economist, and when he died in 1919, Kondratiev prepared the eulogy (part of the reason for his later downfall).

Surviving the November Revolution, he became active in the left-wing agrarian Socialist Revolutionary Party, and in 1920 he was appointed founding director of the Conjuncture Institute, a post he held until 1928. The prime purpose of the institute was to study capitalist crises and to predict the crisis that would lead to the collapse of capitalism.

During this time, Kondratiev was also a professor at the Timiryazez Agricultural Academy and an associate of the U.S.S.R. Popular Commissariats of Finance and Agriculture. He devised the "peasant indices"_price indexes for products bought and sold by farmers_and in 1923-1924 he was in charge of drafting the First Five-Year Plan for Soviet agriculture. He was more conservative than many of his colleagues, and he believed that Soviet agriculture could be collectivized without eliminating the wealthier peasant class, the kulaks. He was at the British Exhibition in 1924 and in the United States for economic talks in 1925.

But Kondratiev's hypothesis that long waves were an organic part of the capitalist system_that there was complete interdependency in capitalism between political and social development, war, economics, and finance (particularly his view that capitalist crises were self-correcting and that successive cycles were therefore repetitive)_brought vigorous criticism. In a 1926 paper he wrote:"Each consecutive phase [of the long wave] is a result of a cumulative process during the preceding phase, and, as long as the principles of the capitalistic economy are conserved, each new cycle follows its predecessor with the same regularity with which the different phases succeed each other" (italics added).Clearly, he was a heretic; capitalism was not supposed to be able to renew itself.

Kondratiev also deviated by supporting the Socialist Revolutionary Party's populist agrarian position that expropriated land be redistributed to the peasants who cultivated it.He was removed as director of the Conjuncture Institute in 1928 and the institute was dissolved. In 1928, the Soviet Russian Encyclopedia said of his long-wave idea, "This theory is wrong and reactionary." He was arrested in1930 and charged with heading an illegal "Working Peasants' Party." In March "The Long Waves in Economic Life," Review of Economic Statistics 17 (1935); and a critique by estonian émigré and research director of the Federal Reserve Bank in New York, George Garvy,"Kondratieff's Theory of Long Cycles," appeared in the Review of Economic Statistics in 1943.There is also Nikolai Kondratieffs "The Long Wave Cycle" [l 928], translated by Guy Daniels, with 1931 he was a witness at the Menshevik trial, and both his long-wave hypotheses and his views on agriculture were denounced. In the Gulag Archipelago,Alexander Solzhenitsyn says that Kondratiev was put into solitary confinement and became mentally ill before his death.

Subsequently, the Great Soviet Encyclopedia called the "Theory of Long cycles" a "vulgar bourgeois theory of crises and economic cycles." "The concept of the long cycle in theory is directed against the basic Marxist thesis concerning the inevitability of economic crises under capitalism, and it conceals the unsolvable contradiction of capitalist society." However, the October 9, 1987,issue of Science contained the following news item (p. 149):

Kondratiev Rehabilitated

The Soviet Union has rehabilitated the economist Nikolai Kondratiev. He is best known for his theory of economic cycles, which has recently seen a resurgence of interest among western economists as an explanation of the link between technological innovation and economic growth. Kondratiev was one of a number of academic economists who were arrested during the purges of the early 1930s because of their opposition to the economic policies of Josef Stalin. He subsequently disappeared after a show trial. Another of those whose works can now be openly studied in the U.S.S.R is Alexander Chayanov, a staunch opponent of the mass collectivization of agriculture who supported the gradual transformation and modernization of peasant small holdings through cooperative farming. Chayanov was shot in 1939.

Kondratiev's rehabilitation, as the Soviet Union strives to restructure, is not inconsistent with the favorable assessment accorded his work by Western economists until the advent of the Keynesian paradigm thrust his ideas aside: "It wasN. D. Kondratieff who brought the phenomenon fully before the scientific community and who systematically analyzed all the material available to him on the assumption of the presence of a Long Wave, characteristic of the capitalist process" (Joseph A. Schumpeter, Business Cycles [1939], 1: 164).

Working first with wholesale price levels in England, France, and the United states, Kondratiev found long-term cycles that he said ran from trough to peak to trough as follows: 1789-1814-1849; 1849-1865(U.S.)/1873(Europe)-1896;1896-1920-(?). He called them "waves of an average length of about 50 years."Writing in 1925, he predicted an inevitable plunge in commodity prices ahead.He then examined interest rates and found, as expected, an inverse relationship between the prices quoted for interest-bearing state bonds (French rentes, Britishconsoles) and wholesale-price cycles. Examining other series, he noted that some supported and others were inconsistent with his central idea. He believed that the relationships were not accidental. Cyclical turning points coincided in many series, and coincident long-wave behavior occurred in many countries.

On this point_that long waves have been a global rather than specifically

More recently, reflecting the contemporary resurgence of interest in the phenomenon, Joshua Goldstein concludes in his book Long Cycles (1988) that"there is remarkable consensus on dating among thirty-three scholars from all theoretical schools of the long wave debate" (p. 74). "Empirical analysis strongly corroborates long waves in price data_both before and after the onset of industrialization in the late eighteenth century. Price waves go back as early as the sixteenth century.... Since the late eighteenth century, price waves appear in, and are synchronous among, various European countries, reflecting the expansion of the core of the world system and its increasing integration in the industrial era" (p. 209).

To better interpret his long-wave idea, Kondratiev studied descriptive data on capitalist nations. He discerned three empirical patterns that he suggested may aid in understanding long waves:

1. Before and during the beginning of the rising wave of a long cycle, the society's economic life undergoes considerable changes. These changes are usually seen in production and exchange techniques preceded by significant technical discoveries and inventions, gold-production and monetary-circulation conditions, and/or involvement of new countries in worldwide economic relations.

The first upwave (1780s-1815) began at the height of the Industrial Revolution and at a time when the United States was emerging into the world's economic markets. The second upwave (1844-1875) was preceded by many technical inventions, including the turbine, harvesting machine, telegraph, steamboat, and sewing machine. The wave's rise was accompanied by expansion of the U.S. role in the world market and a considerable increase in gold production due to discoveries in the United States and Australia. The third upwave (1896-1920) also preceded by important technical inventions, including the vacuum pump, gas motor and engine,telephone, electric locomotive, railroad and streetcar, and transformer; these inventions triggered a second Industrial Revolution, particularly in the chemical and electrical industries. The wave's rise was accompanied by an increase in gold production, establishment of the gold standard in many countries, and the emergence into the world economy of Australia, Argentina, Chile, Canada, and others.

2. The period of the rising wave of a long cycle is characterized by a far greater number of social upheavals and radical changes to society than the period of the downward wave.

The first upwave (1780s-1815) covers the period when the United States gained independence, the French Revolution occurred, there were wars between Napoleonic France and other nations, Russia and Turkey fought, and Poland was partitioned. The second upwave (1844-1875) was marked by revolutionary movements in Italy, Germany, Austria, France, and Hungary, by the Crimean War, the founding of Rumania, the U.S. Civil War, the Franco-Prussian War, and the founding of the German Empire. The third upwave (1896-1920) spanned conflict or wars between Japan and China, Turkey and Greece, and Italy and Turkey, the spanish-American War, the Anglo-Boer War, the Russo-Japanese War, the first and second Balkan Wars, World War 1, and revolutions in Russia, Turkey, China, Germany, and Austro-Hungary.


3. The periods of downward waves of the long cycles are characterized by prolonged and serious depressions in agriculture.

During the first downwave (1815-1844), while the prices of both industrial and agricultural commodities dropped, those in agriculture exhibited relatively greater declines. Starting in 1818, land rents declined until 1840. In England, commissions were established by Parliament in 1820, 1821, 1822, 1833, and 1836 to study the causes and depth of the depression in agriculture. During the second downwave(1875-1896) a great depression in agriculture affected almost all major European nations and the United States. In several instances the price level of agricultural commodities declined absolutely and land rents dropped considerably. Again, commissions were established to study the causes and possible solutions to the depression in agriculture. Kondratiev noted that there were signs of an agricultural depression as the world entered the third long-cycle downwave; the signs were most pronounced in the United States, where commissions on the agricultural depression were formed in 1921 and 1924.


Kondratiev concluded that "the historical material relating to the development of economic and social life as a whole conforms to the hypothesis of long waves." Therefore:

_ Prosperity and depression, with all their profound social implications, are tied to the upswings and downswings of the long waves.
_ During long-wave downswings, "agriculture suffers an especially pronounced long depression."
_ Also during long-wave downswings, "an especially large number of important discoveries and inventions in the technique of production and communication"occur, but these must wait for the next wave's upswing to be implemented on a large scale.

_ Worldwide gold production increases on the wave's upswing, and the global market for commodities and manufactured goods is enlarged by the assimilation of new national territories and colonies.

_"lt is during the period of the rise of long waves (during the period of high tension in the expansion of economic forces) that, as a rule, the most disastrous and extensive wars and revolutions occur." "The rising wave of a long cycle [also] is associated with the replacement and expansion of basic capital goods, and with the radical regrouping of, and changes in, society's productive forces."

Kondratiev's Soviet critics focused on the issue of causation, arguing that the long waves have been created by external circumstances such as inventions, war sand revolutions, international migrations, the assimilation of new countries into the world economy, and fluctuations in world gold production. Capitalism, they said, has internal contradictions that prevent self-renewal. Kondratiev's response was clear: "These considerations . . . are not valid. Their weakness lies in the fact that they reverse the causal connections and take the consequence to bethel cause, or see an accident where we really have to deal with a law governing the events" (italics added). To Kondratiev, the alleged causes were neither exogenous nor random_they were part of the rhythms of the long wave, obedient to and symptomatic of it. The long wave is driven, he believed, by a set of self correcting processes that are inherent to capitalism. For this belief, he died.


LONG-WAVE THEORIES AFTER KONDRATIEV

Kondratiev's long-wave formulation attracted many adherents, each of whom formulated his or her own explanation for the rhythms. Historian Walt Whitman Rostow used the long-wave idea to structure his interpretation of world economic history, centering his explanation of price waves on the problems of overshoot and lag that drive world commodity prices. Craig S. Volland introduced materials-substituting technology into the equation. John R. Borchert described successive technology transitions as leading to epochs of growth. And others, such as J. J. van Duijn, worked on the S shape of infrastructure life cycles.Other formulations include capital-lifespan theories, Joseph A. Schumpeter's innovation-wave concepts, Gerhard Mensch's neo-Schumpeterian metamorphosis model, and neo-Marxist profit-cycle notions.

Despite their differences, what emerges cumulatively from these contribution sis the idea that the long-wave rhythms have something to do with the innovation and diffusion of major new transportation and energy technologies, with infrastructure development, and with collective behavior marked by lag and overshoot. But exactly what the relationship is_what the pattern of causation is_will remain unclear until we have dealt with issues of growth (the cycles of which have a 25-to-30-year periodicity) in Chapters 3 and 4, and have been able to draw together the synchronized rhythms of growth and prices in Chapter 5.

Rostow on Commodity-Price Movements

Rostow pointed to the fact that the evidence for 50-year long waves was in commodity prices. It was logical, then, to seek an explanation in commodity prices. Long-wave fluctuations, he said, could be understood only by relating forces set in motion by a leading sector in growth stemming from the introduction and progressive diffusion of a new technology; forces set in motion by changes in the profitability of producing foodstuffs and raw materials, whether from the side of prices or that of technology, including their effects on investment in new territories and mines, or capital movements, interest rates, terms of trade, and domestic international income distribution; and forces set in motion(notably in housing and infrastructure) by long waves of international or domestic migration, or other forces changing the rate of family formation, the demand for housing, and the size of the work force. All are aspects, he said, of the process of adjustment toward a never-attained dynamic equilibrium.

There were, he concluded, two principal kinds of trend periods (The World economy, pp. 109-110):

1. Periods when prices in general, agricultural and raw-material prices in particular, and interest rates are rising or are high relative to previous and subsequent periods; when agriculture is expanding rapidly; and when income distribution tends to shift in favor of agriculture and profits, while urban real wages are under pressure (1790-1815, 1848-1873, and 1896-1920):

1790-1815: The expansion of the major leading sector, cotton textiles, was dampened by war and the necessity to expand agriculture to feed britain's growing population. The war expanded the iron industry in Britain and led to its takeoff in New England in the United states.

1848-1873: The population-food imbalance operated in full force. Prices for grain shot up in the 1850s. The price of cotton also rose, and urban wages came under severe pressure. The British railroads expanded, but overall growth was affected by the American Civil war. Prices, interest rates, terms of trade, income distribution, and flows of capital were distorted in this period by a series of min or wars and, for the first time, by the attractions of gold mining.

1896-1920: Small wars led to relative shortages of foodstuffs and raw materials, the expansion of military outlays, and finally a great war. Industrial real wages came under pressure. By the late 1890s, the railroads were no longer a leading economic sector. Steel and its subsectors were decelerating. Electricity, new forms of chemicals,and the automobile sectoral complex were moving forward very rapidly. The Boer War was expensive, but in the United States, the prices of agricultural products fell.

2. Periods when the previous trends are reversed (1815-1848, 1873-1896,and 1920-1936):

1815-1848: The price of foodstuffs fell in this postwar period. The speedy expansion of the cotton industry in the 1830s led to a sharp increase in prices in the period 1833-1835. Although the period was not affected by wars, population pressure and the potato crisis of 1845-1847 in Ireland dislodged the food-population balance.

1873-1896: The railroads peaked out as a leading sector, but steel took root,and machine tools and steam engines for ships were born. At this time electricity and new forms of chemicals began to move from the stages of invention to an early phase of innovation. In th elate 1800s, new economies (Sweden, Japan, Russia, and Italy)took off.

1920-1936: Although the prices of agricultural products were still low, this did not encourage a sufficiently accelerated diffusion of the new leading economic sectors in advanced economies.

The answer to long waves, Rostow concluded, lies in the fact that the opening up of new sources of food and raw materials requires substantial periods of time."The lags involved in responding to a relative rise in food or raw material prices,and the fact that the response often required the development of whole new regions, led to an overshooting of world requirements and a period of relative surplus. A relative fall in the prices of food and raw materials then followed . . .until expanding world requirements caught up with the excess capacity" (Getting from Here to There [1979], p. 22). Commodity prices then shifted upward again, stimulating the opening up of new sources of food and raw materials.


Volland on Materials-substituting Technology

Taking Rostow one step further, Craig S. Volland argued in his article "A Comprehensive Theory of Long Wave Cycles" (1987):

_ Long waves are caused by the growth of dominant technologies that strain the supply of natural resources upon which they are dependent.

_ The mechanism that induces the wave is rising commodity prices. In the intermediate term (10-20 years), rising real prices initiate the search for new sources; capital requirements are huge and lead times are very long. In the long term (40-60 years), an exponential increase in demand causes resource-based technologies to become vulnerable to substitution by new technologies because the search for new sources leads to lower-grade ores, heavier hydrocarbons,smaller fields, and frontier areas where the cost of recovery is great.

_ The depressionary leg of the cycle ends as new technologies gain acceptance and relieve the strain. These innovations may allow more-efficient use of exist

ing materials, may be dependent on different materials in plentiful supply that can be more easily utilized, may be only mildly resource intensive, or may be based on renewable resources.

"At first blush," Volland says, "this would appear to be a rehash of Rostow's excellent work. This is not the case. Rostow assumed the rise of commodity prices was due to transient shortages that eventually were relieved by additional investment in the production of those same commodities. He did not take into account the clear historical record of price and technology-induced substitution between key commodities, particularly primary energy sources" (ibid., p. 130).


Borchert on Technology Transitions and Epochs of Growth

This notion of technology-induced substitutions was implicit in Minnesota geographer John R. Borchert's epochal study "American Metropolitan Evolution"(1961). Focusing on transportation technologies and energy sources in their initial phase of deployment, he argued:

_ Before 1830 the location and spread of American cities was primarily influenced by wagon and sail technology. Eastern seaports dominated, and centers rising in importance were located on the nation's inland waterways.

_ From 1830 to 1870, canals, iron rails, and the steamboat were the dominant innovations. Ports with large rail territories grew to dominance.

_ Between 1870 and 1920 the major influences were the steel rail and the oceangoing vessel, combined with the major thrust of industrialization. The Northeastern Industrial Belt lay at the heart of a national rail network, with processing centers at rail nodes.

_ After 1920, cities were reshaped by the automobile, while air travel and long distance communications began to recast regional relationships.

It is new technology, Borchert said, that shapes the pattern of urban and regional development, serving as the leading growth sector. Recasting his supportive materials, we can see in Figure 22 how the inland waterways and the use of captive industrial water power reached their peak-development zenith in the inflationary spiral of 1864-1865 and declined thereafter. Railroad development and the coal industry peaked in the inflationary spiral of 1919-1920 and subsequently declined (Figure 23). Motor vehicles, surfaced-road development, and oil production were all still heading upward in the 1950s (Figure 24), but a more recent rendering of patterns of U.S. energy consumption shows that petroleum consumption peaked in the most recent inflation crisis (that of 1981), just as woodland coal use peaked in the inflation crises of the 1860s and around 1920, respectively (Figure 25).

What these graphics reveal is that the life cycles of energy use and dominant transportation technology have common features (Figure 26). The initial innovation

(i.e., the first commercial application) may occur long before widespread acceptance. With such acceptance, there then is a period of accelerating deployment, a turning point, and a period of decelerating deployment until peak development is achieved during an inflationary crisis. Retrenchment follows as substitute technologies take over. The time period between initial acceleration after one inflation crisis and peak deployment in the next is that of a single longwave. As Volland suggests, inflationary pressures on an old technology spark the growth of an alternative that diffuses until it, too, pressures resources and reaches the inflationary limits of its growth.

The unlabeled dashed line charts the percentage of electrically driven U.S. factories, and the unlabeled dotted line tracks the percentage of U.S. households wired for electricity.That such epochal arrangements may be the consequence of phase structures and regimes that are the outcome of the non-linear dynamics of deterministic chaos is explored by richard Day and Jean-Luc Walter in their 1989 paper, "Economic Growth in the Very Long run." "Evolution," they say, "is driven by an unstable, deterministic (intrinsic) process,not by a random shock (extrinsic) process" (p. 285). They thus add support to Walt W.Rostow's belief that "a satisfactory dynamic theory of production and prices must render substantially endogenous the sequence of major inventions and innovations_the leading sector complexes_as well as the incremental improvements in productivity embraced under the case of increasing returns" (Why the Poor Get Richer and the Rich Slow Down[1980], p. 42)


The Infrastructure Life Cycle

What J. J. van Duijn suggests in his 1983 book, The Long Wave in Economic life (echoing Simon Kuznets's 1930 Secular Movements in Production and prices), is that "growth is an S-shaped phenomenon": basic innovations give rise to new industrial sectors that develop in an S-shaped life-cycle pattern, New sectors require their own infrastructure. But excess accumulation of physical stock will occur. The leveling off of demand in the innovation-incorporating sectors will accentuate the overexpansion of the capital sector. The combined effect of those two forces is a long-wave downturn. The pattern is that of a five phase product life-cycle in which the phases are introduction (innovation),growth, maturity, saturation, and decline (Figure 27). Overlapping transportation and energy life cycles produced the technological successions that we saw

in Figure 26. Such successions have been central to Eric Jantsch's work on technological forecasting.
In a 1989 paper entitled "The Third Kondratieff Wave," Christopher Freeman offers the following characterization of technological successions, which he equates with Kondratiev-phased modes of global economic development:
1. Early Mechanization Kondratiev, 1770s-1840s
Key industries: cotton, pig iron

reuaing sectors and infrastructures: textiles, textile chemicals, textile machinery, iron-working and iron castings, water power, potteries, trunk canals, turnpike roads

Technological leaders: Britain, France, Belgium
Other industrializing countries: German states, the Netherlands
2. Steam Power and Railway Kondratiev, 1840s-1890s
Key industries: coal, transport, heavy engineering

Leading sectors and infrastructures: steam engines, steamships, machine tools, iron, railway equipment, railways, world shipping

Technological leaders: Britain, France, Belgium, Germany, the United StatesOther industrializing countries: Italy, the Netherlands, Switzerland, AustriaHungary

3. Electrical and Heavy Engineering Kondratiev, 1890s-1940s
Key industries: steel, automobiles, aircraft, telecommunications, radio, alu
minum, consumer durables, oil, plastics
Leading sectors and infrastructures: electrical engineering, electrical machin
ery, cable and wire, heavy engineering, heavy armaments, steel ships, heavy
chemicals, synthetic dyestuffs, electricity supply and distribution
Technological leaders: Germany, the United States, Britain, France, Belgium,
Switzerland, the Netherlands
Other industrializing countries: Italy, Austria-Hungary, Canada, Sweden,
Denmark, Japan, Russia
4. "Fordist" Mass-Production Kondratiev, 1940s- 1990s
Key industries: energy, computers, radar, machine tools, drugs, nuclear
weapons and power, missiles, microelectronics software
Leading sectors and infrastructures: automobiles, trucks, tractors, tanks, ar
maments for motorized warfare, aircraft, consumer durables, processing
plants, synthetic materials, petrochemicals, highways, airports, airlines
Technological leaders: the United States, Germany, other EEC, Japan, Swe
den, Switzerland, the Soviet Union, other EFTA, Canada, Australia
Other industrializing countries: other Eastern European, Korea, Brazil, Mex
ico, Venezuela, Argentina, China, India, Taiwan
5. Information and Communications Kondratiev, 1990s_?

Key industries: "chips" (microelectronics), third-generation biotechnology products and processes, space activities, fine chemicals, SDI

Leading sectors and infrastructures: computers, electronic capital goods,software, telecommunications equipment, optical fibers, robotics, flexible manufacturing systems, ceramics, data banks, information services, digital telecommunications networks, satellites

Technological leaders: Japan, the United States, Germany, Sweden, other EEC, EFTA, the Soviet Union and other Eastern European, Taiwan, Korea,Canada, Australia

Other industrializing countries: Brazil, Mexico, Argentina, Venezuela,China, India, Indonesia, Turkey, Egypt, Pakistan, Nigeria, Algeria, Tunisia, other Latin American

"Transmaterialization"

Related successions take place in the mineral industries: transmaterialization,according to Lorna M. Waddell and Walter C. Labys, is "a characteristic behavior of materials markets through time" in which minerals demand changes as materials linked to mature industries undergo periodic replacement by higher quality or technologically more appropriate materials linked to new industries.The energy-source transitions that we have already seen are part of a much broader pattern of materials-demand successions that also follow five-phase life cycles. The notion is certainly clear enough in Figures 28, 29, and 30, as demand has shifted from the industries of the early twentieth century to those of the postwar boom, and now to those of the modern "high tech" complex.

Capital-Lifespan Theories

Kondratiev believed that "the material basis of the long cycles is the wear and tear, the replacement and increase of the fund of basic capital goods, the production of which requires tremendous investment and its long process." He echoed a number of earlier economists. His mentor Mikhail Tugan-Baranovskij had argued as early as 1894 that the main cause of cycles was the fluctuating rate of growth of fixed capital. In 1896, Knut Wiksell had written about discrepancies in the rate of return on investment caused by technological or other changes inopportunity. Albert Aftalion spoke in 1913 of the long period required to bring capital into being, the long life of capital goods, and the capacity of relatively minor changes in consumption to generate large changes in net additions to fixed capital required by business.

The view is not one on which Marxists look with disfavor. Marx had argued that the obsolescence of capital goods was the chief cause of depressions, with the efficiency life of machines determining the length of expansion periods. In1920, Samuel De Wolff estimated that the depreciation rate for long-lived fixed capital ("capitalist infrastructure" such as railroads, bridges, and manufacturing plants) was 2.615 percent per year, leading to a 38-year lifespan.

After Kondratiev, the idea has continued to be developed. In 1939, MichalKalecki asked what causes periodic crises. His answer was that investment is not only produced; it produces. Investment is a source of prosperity. Increasing investment improves business and stimulates a further rise in investment. But at the same time, investment is an addition to capital equipment. Right from birth it competes with an older generation of this equipment. Inevitably, more new investment creates a crisis for old investment.

Two ideas thus seem to be embedded in capital-lifespan concepts: (1) the idea of the lifespan itself; (2) the notion of substitution of newer capital for old. These ideas were developed by Jay W. Forrester and his associates at the Massachusetts institute of Technology in the 1970s. They built an elaborate fifteen-sector System Dynamics National Model of the U.S. economy in which 50-year long waves were generated consisting of three decades of investment, one decade of market saturation, and one decade of depression. John D. Sterman (1985) emphasizes that this model produces results consistent with the idea of bounded rationality. Forrester (1977) concludes that sufficient causes for long waves are the long lifespan needed to change the production capacity of the capital sectors,the way capital sectors provide their own input capital as a factor of production,the need to develop excess capacity to catch up on deferred demands, and psychological and speculative forces that can cause overexpansion in the capital sector. The sequence is:

1. Decay of the capital plant below the level required for replacement needs

2. Development of shortages that promote demands for capital-intensive pro

duction

3. Self-ordering of capital by the capital sector of the economy in order to produce the capital needed by consumer-goods industries

4. Overexpansion of the capital sector to a capacity greater than needed for replacement to catch up with deferred needs

5. Excess accumulation of capital in the housing, consumer durables, and durable manufacturing sectors

6. Failure of capital users to absorb the output of the overexpanded capital sector s

7. Sudden appearance of unemployment and rapid collapse of the capital sector

8. Decline of excess capital through physical depreciation

See Figure 31.

Forrester writes that each major expansion grows around a highly integrated and mutually supporting combination of technologies (particularly transport and energy) which during good times tends to reject incompatible innovations.The process culminates in excess debt and an overbuilding of the capital sectors,followed by depression. During the depression the excess capital is worn outland fully depreciated. Defaults and bankruptcies clean out the excess debt load.The long wave involves an overbuilding of the capital sectors. Their growth exceeds the capital-output rate needed for long-term equilibrium to a point beyond that justified by the marginal productivity of capital. Finally, the overexpansion is ended by a great depression during which excess physical capital is

physically worn out and is financially depreciated on the account books until the economic stage has been set for a new era of rebuilding.

Graham and Senge (1980) also argue that the switch over to new technologies comes in the Kondratiev downwave. Near the peak of the wave, major industries suffer decreasing returns to capital. The resources required by a particular type of investment become more costly to acquire. As the economy slides toward depression, opportunities arise for inventions that have been waiting in the wings to be implemented. Thus, Ehrensaft (1980) could point to the clustering of agricultural innovations in the downswing as farmers adopted new technologies to increase production in the face of declining prices. These inventions become the basis of the next wave of growth; the next Kondratiev trough is the point of technology transition at which deployment of the new technology exceeds the old. George Ray (1980, 1983) argues that new energy sources are the central elements in the innovations that are adopted. A. Van der Zwan (1980)maintains that technologically advanced sectors shape the recovery from depression. All of which points to the critical role played by innovation.


Schumpeter on Innovation Waves

Kondratiev believed that during the long-wave recession "an especially large number of important discoveries and inventions in the technique of production and communication" occur, but that they must wait for the next wave's upswing to be implemented on a large scale. This was the point of departure for Joseph a. Schumpeter, who tried to build an innovation-wave theory of long cycles.

Schumpeter, once Austria's Minister of Finance and later a professor at Harvard University, undertook as his life's work a detailed account of the economic histories of England, Germany, and the United States after 1787, and provided strong empirical support for the idea of long waves.2

He concluded that any fully comprehensive economic history necessarily had to take account of multiple cycles, and in Business Cycles (1968) he was the first to name three of them after their discoverers: Kitchins (inventory cycles now acknowledged to run 3 -4 years), Juglars (investment cycles running 7- 11years), and Kondratievs (long waves of 45-60 years).3


2. Works by Joseph A. Schumpeter: The Theory of Economic Development (1934), a translation of the first German edition of 1912; Business Cycles: A Theoretical, Historical, and Statistical analysis of the Capitalist Process (1939); Capitalism, Socialism, and Democracy ([1942] 1976).See also Michael 1. Stevenson, Joseph Alois Schumpeter: A Bibliography, 1905-1984 (1985). John maynard Keynes wrote, in his Treatise on Money (1930): "In the area of fixed capital it is easy to understand why fluctuations should occur in the rate of investment. Entrepreneurs are induced to embark on the production of fixed capital, or are deterred from doing so by their expectations of the profits to be made. Apart from the many minor reasons why these should fluctuate in a changing world, Professor Schumpeter's explanation of major movements may unreservedly be accepted"(p. 85).

3. Joseph Kitchin had proposed a 3.3-year farm commodity cycle in 1923, although his analysis of economic fluctuations did distinguish between these "minor cycles" and what he called "major cycles" (Juglars) and "fundamental movements" (Kondratievs). Clement Juglar was the first to iso

Like Kondratiev, Schumpeter accepted that long waves (in Arthur Burn s'swords, "lopsided surges of development that mark economic progress" ["Business Cycles" (1968), p. 227]) were not externally caused, but were the internalregulators of capitalist development. He had already concluded that savings and investment were mere quantitative expressions of demographic growth, and what he sought was an explanation for the qualitative shifts that define development. Technical innovation embodied his notion of qualitative change: innovation as "the setting up of a new production function" (Business Cycles, p. 87).Almost as an echo came the remarks of Robert McCormick Adams in his 1988essay "Contexts of Technological Advance": "It is difficult to escape the view that technological change has a focussed, non-random, and episodic or wavelike character" (p. 43). Such "technological change is not an isolate, but a set of material and behavioral manifestations deeply embedded in a social and cognitive system.... There is a perception that it is an independent causative agent because of massive second-order effects [or] revolutions" (p. 63).

In Schumpeter's view (Business Cycles, pp. 93-101) major innovations require non-negligible time and outlay. They tend to be embodied in "new firms"and to be associated with the rise to leadership of "new men." Old firms are conservative; innovations emerge primarily with new and less conservative firms. And new men insert "a class of facts of the behavioristic type." Old businessmen work within a limited horizon, making decisions about profitability with the foresight provided by well-worn channels: "the repetition of acts of routine." Innovations have to come from new men working outside these channels: "a new frame has to be constructed." Thus, whenever there is evidence that a major breakthrough has been made, it becomes much easier for others togo the same thing and to improve upon it: "They are driven to copying it if they can ... and it becomes easier to do similar things in similar lines ... when certain innovations such as the steam engine, directly affect a wide variety of industries.... First ... innovations do not remain isolated events, and are not evenly distributed in time.... they tend to cluster, to come about in bunches,simply because first some, and then most, firms follow in the wake of successful innovation; second . . . innovations are not at any time distributed over the economic system at random, but tend to concentrate in certain sectors and their surroundings" (italics added).

Disturbances from equilibrium arising from innovation cannot be smoothly absorbed. The "disturbances must necessarily be big, in the sense that they will disrupt the existing system and enforce adaptation" (ibid., p. 101). "For actions


late the 7-to-11-year business cycle, averaging 9-10 years. Schumpeter said that three Kitchin cycles occur within each Juglar, and that each Kondratiev long wave was composed of six Juglars. See joseph Kitchin, "Cycles and Trends in Economic Factors," Review of Economic Statistics 5 (1923):10-16 Clement Juglar, Des crises commerciales en leur retour pérodique en France, en Angleterreet aux Ëtats-Unis (1862); and Schumpeter, Business Cycles. See also Arthur F. Burns and Wesley C.Mitchell, Measuring Business Cycles (1946).

which consist in carrying out innovations we reserve the term Enterprise; the individuals who carry them out we call entrepreneurs" (p. 102). (The contrasting view is that of the planned innovation that takes place in the R&D facilities of the large modern corporation.) But entrepreneurial activity upsets the equilibrium of the system. This increases instability and risk. Thus, after the initial burst of enterprise, it is necessary for a new equilibrium to be established. Askew products stream into markets, the agenda switches to the predictability of a new routine and entrepreneurial activity ceases (p. 176). Economic change o revolution thus "goes in units separated from each other by neighborhoods of equilibrium" (p. 138). Moreover, "one cyclical solution produces the next....Recession is a reaction to prosperity.... Windfall gains, rising prices and so on produce waves of optimism" (pp. 140-142). And "if innovations are being embodied in new plant and equipment, additional consumers' spending will result practically as quickly in new producers' spending. Both together will spread from the point or points in the system on which they first impinge, and create prosperity. Two things are then practically sure to happen. First, old firms will react to this situation and, second, many of them will speculate on this situation.... New borrowing will then no longer be confined to entrepreneurs, and deposits will be created to finance general expansion, each loan tending to induce another loan, each rise in prices another rise.... This is what we call ...the Secondary Wave, which superimposes its effects on the Primary Wave"(p. 145). "The phenomena of this secondary wave may be and generally are quantitatively more important than those of the primary wave.... This is one reason why the element of innovation has been so much neglected" (p. 146).

The booms caused by the bunching of innovations will come to an end. New entrepreneurs' demands for labor, capital, and raw materials drive up their prices. New products come on the market and compete with the original innovations; receipts decrease as costs increase. The appearance of new products favors a fall in prices_the timing of which depends on how soon the new products appear_which terminates the boom and leads to a depression.

Schumpeter thus argued that economic development is a discontinuous process, appearing in clusters of entrepreneurial activity that changes the structure of the economy: new techniques, new products, new markets, and new raw materials. Concentrations of entrepreneurial fervor are separated by long periods of consolidation of the new advances, without major new entrepreneurial activity. Major innovations sparked Britain's first Industrial Revolution in the downwave from the stagflation crisis of 1763 and helped spark the upwave of the 1790s: Watt's rotary-motion improvements to Newcomen's steam engine came between 1769 and 1784; the period 1765-1769 saw the introduction of Hargreave's spinning jenny and Arkwright's water frame, followed by Cartwright's machine loom in 1787; in 1761 the Bridgewater canal was built to move coal, and by 1815 Britain had 2,200 miles of canals and 2,000 miles of improved rivers; in the 1790s John L. McAdam introduced his hard-packed

road surface that became standard for turnpikes. Railroads were introduced in the downwave of the 1820s (Stockton and Darlington, 1825; Liverpool and Manchester, 1829). The invention was proved by the success of 6,500 miles of track before the depression of the Hungry Forties, and was a major element of the steam-railroad-iron expansion of the 1850s and 1860s that followed.

But what was involved simultaneously, Schumpeter said, was a process ofcreative destruction. Economic development involves the periodic destructive disruption of the capitalist production system and its creative renewal by innovations that lift the economy to a higher plane of development: "Capitalism,then, is by nature a form or method of economic change and not only never is but never can be stationary.... The fundamental impulse that sets and keeps the capitalist engine in motion comes from the new consumers' goods, the new methods of production or transportation, the new markets, and the new forms of industrial organization that capitalist enterprise creates.... The opening upon new markets, foreign or domestic, and the organizational development from the craft shop and factory to such concerns as U.S. Steel illustrate the same process of industrial mutation.... That incessantly revolutionizes the economic structure from within, incessantly destroying the old one, incessantly creating anew one. This process of Creative Destruction is the essential fact about capitalism" (Capitalism, Socialism, and Democracy [1942 (1976)], pp. 82-83).


Mensch's Metamorphosis Model

Schumpeter-inspired research has multiplied. Recently, a neo-Schumpeterian,Gerhard Mensch, proposed a metamorphosis model, attempting to set Schumpeter's ideas more squarely in a long-wave frame.4 Mensch's key concept is that of a technology stalemate out of which an economy is ultimately eased by clusters of innovations taken from a reservoir of investment opportunities formed by a continuing stream of scientific discoveries. This produces a structural metamorphosis as old activities are cast off and replaced by revolutionary new ones.

Mensch postulates the following causal sequence:

1. A cluster of basic innovations (introducing new branches of industry) and radical-improvement innovations (which rejuvenate existing branches) occurs in response to a technology stalemate. Venture capital is attracted to the new lines of business. New demands are awakened.


4. Gerhard Mensch, Stalemate in Technology (1979). See also J. J. van Duijn, The Long Wave in Economic Life (1983); John Clark, Christopher Freeman, and Luc Soete, "Long Waves, Inventions, and Innovations," Futures 15 (1981); Christopher Freeman, ea., Long Waves in the World Economy (1984); Christopher Freeman, Design, Innovation, and Long Cycles in Economic Development (1986). See also Alfred Kleinknecht, Innovation Patterns in Crisis and Prosperity: Schumpeter's Long Cycle Reconsidered (1987). Kleinknecht, a major figure in the neo-Schumpeterian revival of the 1980s, argues that "waves of important innovations are an endogenous element of the long-wave process. The wave-like occurrence of major innovative breakthroughs is a decisive cause of the fairly simultaneous rise of new branches of industry which foster the long-wave theory"(p. 206).

2. Parallel S-shaped innovation-growth cycles, some of which substitute for older goods or services, characterize the new branches of industry. Initial entry is followed by rapid upswing and by accelerated growth.

3. During the new-product upswing, investment, employment, and incomes increase rapidly, well ahead of prices and inflation. This is the period of disinflation or deflation.

4. Basic and radical-improvement innovations are followed by routine-improvement innovations that rationalize production and increase capital intensity. But these innovations are subject to diminishing returns on the demand side. The growth curve grades over from acceleration to deceleration. This is a natural phase of the product life-cycle of manufactured goods.

5. Corporate growth overestimates domestic markets and produces excess supply and heated competition. Markets become saturated. Almost inevitably,an industrial economy tends to overinvest in any new technology_capital goods in particular.

6. There are two responses to excess supply: attempts to reduce competition by product differentiation and industrial mergers, and attempts to segment domestic and foreign markets, "dumping" excess output. Over investment breeds"pseudo-innovation," which benefits neither buyer nor seller. Rather, existing industries attempt to protect their market shares by means of product differentiation in which the "image" or "packaging" of the product is changed, but no longer are any of its basic qualities altered. The economy then enters a periodic which output stagnates. Limits to growth are encountered, markets are flooded, key resources may be scarce and their prices rise, and there is a slowdown in income and employment growth. Rates of return decline, and capital is therefore not reinvested in existing lines of business. Instead, money moves into speculation.

7. The end result is a technology stalemate in which growth is replaced by stagnation, and in which large-scale organizations seek to maintain the appearance of growth by controlling output and raising prices, thereby inducing stagflation. During this period, one sees increasing protectionism, yet over concentration in leading industries sets the price spiral into gear. Large-scale industry and large-scale labor push up wages and prices. Large-scale organizations that growing the period of pseudo-innovation require and promote conservative patterns of investment: risk-taking in new ventures is minimized.5


5. In this context Carlota Perez argues, in "Structural Change and the Assimilation of New technologies in the Economic and Social System" (1981), that old technologies are perpetuated after new technologies are introduced, due to institutional forces such as the formation of special-interest groups and the protection of specialized labor by trade unions. There is also a merger-wave literature, which says that mergers have come in cycles in the United States. Martin Schnitzer (1987)argues that mergers afford a way of protecting against decline, for if a firm can gain control over competing plants, it has a better opportunity to fix prices and control output. The first great wave of mergers occurred between 1887 and 1904; these mergers took place in such industries as petroleum, iron and steel, copper, sugar, lead, and salt, and their common feature was the drive toward

8. The price spiral ends in a sharp collapse. But because returns in older,established industries were eliminated during the period of stagflation, and the prospect of speculative profit vanishes in the ensuing collapse, new venture capital becomes available, seeking high-growth investment opportunities. The appearance of venture capital at this time results in a rush of attempts to convert many of the inventions that have appeared since the preceding period of basic innovation into useful techniques or products_that is, into a cluster of basic innovations that precipitates another long-term growth upswing. New industries attract capital and labor from stagnant sectors, circumvent older resource scarcities, stimulate demand for new kinds of goods and services, and generally introduce the reinvigorating effects of a structural transformation of the economy. Another cycle begins.

The key to Mensch, then, is the technology stalemate that occurs in a stagflation crisis, and the emergence of a pool of venture capital that seeks out innovative alternatives in the ensuing era of declining prices. But as we shall see later,Mensch's formulation has several critical weaknesses. He does not allow for intervening depressions, nor does he recognize that the growth impulses co mein Kuznets-cycle bursts. Because he, like so many other authors, treats Kondratiev waves as growth cycles, he confuses growth with the S-shaped technology successions that do have Kondratiev periodicity.

Mensch writes that he has discerned three major innovation clusters since1800_in 1814-1828, 1870 - 1886, and 1925 - 1939. See Figure 32. These coincide with the European growth accelerations from the inflationary crises of1814 (Britain), 1873 (Britain, France, Germany), and 1920 (Britain)/1923 (Germany). The comparable U.S. crises occurred in 1814-1815, 1864-1865, and1920, each followed by a decade of deflationary growth. But several authors have suggested that Mensch failed to include a full record of innovations and that this is why he found clusters that fit into the European timing. Certainly, if his clustering notion has merit, the proof should be demonstrable by comparing the United States with Europe, since the timing of the inflation crises and the ensuing growth accelerations is somewhat different in each. I know of no such proof.

Mensch also errs because the bandwagon effect he postulates is not immediate, but takes off well after the innovations have been put into place and their viability has been established. Christopher Freeman notes that the diffusion time monopoly_the consolidation of producers into firms dominating the market. The second cycle of mergers occurred between 1916 and 1929; these mergers_primarily in the mass-production and entertainment, automobile, automobile parts, motion picture, movie theater, and appliance industries_took the form of vertical integration by oligopolists. A third cycle of mergers occurred between 1954 and 1968; this wave was characterized by a movement toward conglomerates. A fourth cycle of mergers_of the conglomerates_began in the 1980s; this cycle resembles the third one in the sense that it has witnessed mergers across different kinds of industries, but on a larger scale. See Martin Schnitzer, Contemporary Government and Business Relations (1987);

for all groups to benefit from adopting a new technology is approximately thirty years. If there is "bunching," Freeman says, it arises from imitation and from a clustering of technically related inventions that is "rather more likely [to be]related to breakthroughs in fundamental science and technology, to bursts ofinvestment, and to periods of very strong demand (including booms and wars)"than to "explosive bursts of imaginative entrepreneurs [anal a swarming of innovative behavior" (Freeman, Clark, and Soete, Unemployment and Technical innovation [1982], p. 321). A related criticism appears in Solomos Solomou's1986 paper "Innovation Clusters and Kondratieff Long Waves in Economic Growtlh": "Mensch's ideas are faulted on a number of counts, the two most important being the use of an unrepresentative innovation selection and the use of an inappropriate statistical test. An examination of the Mensch, Van Duijnand Kleinknecht innovation selections rejects the idea of regular innovation clusters during long wave depression phases. The evidence also rejects the weaker hypothesis of Van Duijn that innovation flows have followed a longwave path; structural changes in the twentieth century have been confused with long waves. The evidence presented here vindicates the Freeman et al. conclusion that the flow of innovation is not responsive to long-wave phases" (p. 111).


MARXIST LONG-WAVE THEORY

The initial Marxist reaction to Kondratiev was spearheaded by Leon Trotsky,who formulated a "curve of capitalist development" in his "Report on the World crisis" to reinforce the argument that long-wave cycles are not internally regulated. Rather, he argued, long waves are periods of growth and decline that lie between capitalist crises_turning-points with external causes, such as wars and revolutions. He reasoned that each epoch of revolution eroded the mainstays of points, the external forces changed the slope of the dynamic equilibrium, producing a curve of capitalist development that moved inevitably toward the final crisis and collapse.


The Profit Cycle

Trotsky was forced out of the Soviet Union and was ultimately murdered, but Marx and Engels had already argued that changes in the rate of profit on both new and replacement investments lay at the root of long swings in capitalist countries. The Belgian Trotskyite Ernst Mandel drew together his hero's notion of the crisis-prone tendencies of capitalist development and Marx's profit cycles to produce the contemporary Marxist theory of long waves. Mandel argues in his 1975 book, Long Waves of Capitalist Development, that long waves have their roots in long-term fluctuations in rates of profit. For Marxists, profit arises from the surplus value over that received by workers in the productive process. Profits provide capital. But competition forces capitalists to use capital to mechanize the productive process, replacing labor. The rate of profit r =sl(c + v), where s is surplus value, c is fixed capital in machines, etc., and v is variable capital items used in the productive process. If s and v remain static, as capital c is increased, r, the rate of profit, must decline, leading to Marx's "Law of the Tendency for the Rate of Profit to Fall." If profits fall, so does the possibility for further capital accumulation. Hence an expansionary phase of growth necessarily must give way to stagnation. During an upswing, demands for means of production in the consumer-goods sector raise the output and profits of capital-goods producers. Investment funds flow to that sector, where above-average rates of profit can be earned. But competitive capital intensification forces profits down. As this occurs, investment funds flow back to the consumer-goods sector. These movements in the rate of profit drive the capitalist industrial cycle through a sequence of accelerated capital accumulation, overaccumulation, decelerated accumulation, and underinvestment in the capital-goods sector as investment drives growth in the consumer-goods sector. Others are not so sure. In his 1987treatise, Phases of Economic Growth, Solomos Solomou rejects Kondratiev phasing for prices, investment, and productivity between 1850 and 1973.

A new cycle can only begin, Mandel says, when some externally generatedchange permits a new round of above-average surplus profits to be earned; a capitalist crisis occurs because the profit possibilities of the old industries have been exhausted. Capitalism would fail if it were not for external stimuli that ease it out of crisis and change its character. Examples include: (1) the sudden flight of capital to other countries, (2) a rise in the surplus value of capital as a result of increased exploitation of the working class by entrepreneurs, (3) a falling the price of fixed capital, and (4) innovation.

Thus, Mandel says, the British long wave from 1790 to 1848 was fueled by the access of new manufacturing industry to pre-capitalist-agricultural regions that provided cheap labor and raw materials. The late-nineteenth-century long

wave drew surplus profits from the imperialist expansion of developed capitalist economies into colonial regions. The boom following World War 11 was fueled by the surplus profits of giant corporations that monopolized the most productive techniques within their own countries. The curve of capitalist development moved from competitive capitalism to imperialism to monopolistic late capitalism. Each stage involved a new wave of technical change. The succession of long waves was produced by the dialectic of endogenous (profit cycle) and exogenous(surplus profit opportunity) factors.
In this light, Mandel (Late Capitalism [1975], pp. 131-132) advances the following time-line. It bears little relationship to other long-wave chronologies,and I can find little empirical support for it, save Andrej V. Poletayev's recent essay "Long Waves in Profit Rates in Four Countries" (1989), in which peaks in the rate of return are identified in the early 1870s, soon after 1900, and in the 1950s.
Long Wave "Main Tonality" Origins of This Movement
1793-1825 Expansive; rate of profit rises.
Artisan-produced machines; agriculture lags. behind industry; rising prices for raw materials. Fall in real wages with a slow expansion of the world market (South America).

1826-1847 Slackening; rate ofprofit stagnates.

Dwindling of profits made from competition with precapitalist production in England and Western Europe. Expansion of the World Market decelerates.

1848-1873 Expansive; rate of profit rises

Transition to machine-made machines. Massive expansion of the world market following the growing industrialization and extension of railway construction in the whole of Europe and North America as a result of the 1848 Revolution.

1874-1893 Slackening; rate ofprofit falls then stagnates, then rises slightly.

Machine-made machines are generalized.The commodities produced with them no longer produce a surplus profit. The increased organic composition of capital leads to a decline in the average rate of profit. In
Western Europe real wages rise. The results of the growing export of capital and the fall in the prices of raw materials only gradually permit an increase in capital accumulation. Relative stagnation of the world market.

1894-1913 Expansive; rate of profit rises, thenstagnates

Capital investments in the colonies, the breakthrough of imperialism, and the generalization of monopolies profiting even further from the notably slow rise in the price of raw materials (promoted by the second technological revolution, with its accompanying steep rise in the productivity of labor and the rate of surplus-value), permit a general increase in the rate of profit, which explains the rapid growth of capital accumula tion. Vigorous expansion of the world market (Asia, Africa, Oceania).

1914-1939 Regressive; rate of profit falls sharply.

The outbreak of World War I, the disruption of world trade, and the regression of material production, determine growing difficulties in the valorization of capital, reinforced by the victory of the Russian Revolution and the narrowing of the world market it provoked.

1940/1945 Expansive; rate of profit first rises then to fall slowly starts to fall.

The weakening (and partial atomization) of the working class determined by fascism and World War II permits a massive rise in the rate of profit, which promotes the accumulation of capital. This is first thrown into armaments production, then into the innovations of the third technological revolution, which significantly cheapens constant capital and thus promotes a long-term rise in the rate of profit. The world market shrinks through autarky, world war, and the extension of non-capitalist zones (Eastern Europe, China, North Korea, North Vietnam, Cuba).

Without a secure empirical footing, Mandel's profit-cycle model remains an ideologically preformed construct rather than a useful contribution to longwave theory.

Political Business CyclesNotwithstanding the lack of a firm empirical footing, Marshall takes Marxian long-wave analysis a step beyond Mandel in his 1987 monograph, Long Waves of Regional Development. He writes: "The development of industrial capitalism can be divided into three historical periods. These are the period of competitive capitalism, consisting of the long waves from the late eighteenth to the late nineteenth century; the imperialist long wave from the late nineteenth century to the second World War; and the post-war long wave of late capitalism" (p. 19)."Cyclical crises during each wave are indecisive in the sense that they are of insufficient impact to sweep away social, economic and political obstacles to capitalist regeneration. The upward turning-points of the long waves presume a decisive crisis which has the function of transforming the social and economic structure, preparing the path for a fresh and durable phase of renewed economic expansion. These decisive crises involved social upheavals hinging upon the transformation of the capitalist labor process" (p. 70, italics added). But then marshall provides a very confusing diagram (Figure 33). His curiously selected turning points_1790, 1825/1826, 1847/1848, 1873/1874, 1893/1894, 1913/1914, 1940/1945,1966/1967_reveal how heavily he relies upon Mandel's construct as he attempts to develop "a historical framework for economic and social analysis by weaving together the perspectives of the course of the long waves,the overall periods of capitalist development . . . and the evolution of the labor process which marks each phase of industrial and technological transformation"
(P- 97). ~

Marshall does break with Mandel in one respect, however, by introducing the alternative causation proposed by Michal Kalecki in his notion of political business cycles. According to Kalecki (1939, 1943), the capitalist and working classes both place demands upon government. At the end of a boom phase,under conditions of full employment, capitalists cannot tolerate the greater bargaining power of labor and thus exert pressure for governments to enact deflationary policies leading to recession and rising unemployment. Recessionary conditions, on the other hand, stimulate labor to put pressure on government through militancy and the ballot box to adopt reflationary policies to increase employment and wages. The dialectic is sufficient, Kalecki argues, to produce the oscillation of boom and recession. But other Marxists, such as David M.Gordon (1978), argue that the intervention by government is unnecessary: rising unemployment in the downswing disciplines labor and facilitates the ensuing upturn. As Glyn and Sutcliffe (1980) note, rising wages resulting from improved bargaining positions drive down profits and ultimately lead to the scrapping o fold lines of production. This, they feel, is a better explanation of the profit cycle.

What remains important, Marshall says, is the classic Marxist concept of

capitalist crisis (revealing thereby that his ideology shapes his "facts"). He says that one element of the crisis is the profits squeeze. Each long wave has been generated by a new way to create surplus profits and reverse that squeeze. But the boom-crisis cycle has led to the progressive concentration of capital in fewer hands, and the future of capitalism depends upon how the next decisive crisis is handled by the key actors: multinational corporations, international bankers,and nation-states. Whether we accord this argument credibility depends upon whether we are comfortable with Marshall's ideologically shaped view of what drives long waves, the facts of the case notwithstanding. I prefer to get my facts straight first, and the facts of growth cycles are quite different, as I will show in chapter 3.