But prices of financial assets reflect the aggregate value placed on them by the participants in those markets. So it seems that any information that might tell us when people's attitudes or preferences regarding such assets are changing would be imminently important to investors and decision makers. Apparently not so. Very little of what passes for economics ever examines the dynamics of the society that underlies those financial markets. As a result, the current view has produced highly complex linear forecasting models that perform reasonably well between turning points in the economy but unimpressively at those turning points. It has also given rise to the Random Walk theory as the dominant explanation of price behavior in the associated markets. Recently, however, this paradigm has come under increasing attack for its inadequacy in explaining observed market phenomena.

If we are willing to view market behavior from outside of the accepted paradigm and entertain the possibility that government action is only of secondary importance to endogenous societal forces, some intriguing possibilities emerge. We might conclude that markets are inherently unstable (as the Keynesians contend) but that any attempt by government to stabilize them will likely sacrifice economic growth (as the classicists would argue). Instead of treating economic inputs such as taxes, interest rates and money supply as primary causal variables, these might instead be viewed as intermediate variables. The primary causal variables that produce these intermediate responses would then only be discoverable by characterizing the behavior of the primary participants in the markets, people.

The only manageable approach to the problem of understanding how people behave is to study their behavior in groups, searching for reliable societal patterns over time. As a starting point, it is reasonable to assume that the economic output of a society is determined by the number of productive people, their productive capacity, and the social priority placed on production. The respective causative forces behind each of these variables is population growth, technological innovation, and cultural ethos. As we shall see, associated with each of these causative forces are both secular and cyclical components.

POPULATION GROWTH AND THE 40 YEAR CYCLE

The more people there are in a society, the greater the economic output we should expect from that society. Therefore, to compare the productive output between two nations, or within one nation over time, it is helpful to first divide that output by the number of people producing it. The real per capita gross national product of the United States increased 888 percent between 1873 and 1990, while the real S&P 500 index adjusted for population over that same time period increased by only 21 percent, nearly unchanged (see Figure 1). A case could be made that all production gains from technology are paid out to investors in the form of dividends, leaving the real per capita value of the productive assets unchanged. The effect of the secular increase in population then would be to increase the total value of productive assets proportionately.

However, population in the United States does not increase at a steady linear rate but rather displays a 40 year cyclic pattern. Figure 2 shows annual births since 1909 and clearly demonstrates cycle peaks at 1920 and 1960. Decennial data prior to 1909 indicate peaks near 1835 and 1880. When the number of productive workers is increasing rapidly, it would be reasonable to expect that the economy would also expand rapidly. Since workers are most productive when they reach middle age, having maximized their job skills and responsibility levels, we should anticipate that a bulge in the birth rate would manifest itself in higher economic growth some 40 to 50 years later. Indeed, research by H. S. Dent, Jr. indicates a strong correlation between stock market tops and the peak spending ages of 47 to 49. For example, the baby boom terminating in 1920 resulted in a stock market peak 48 years later in 1968. We can expect then that the cyclic phenomenon of population growth will manifest itself in a 40 year boom-bust cycle in the stock market, and three earlier peaks do occur in 1853, 1881, and 1929.

TECHNOLOGICAL INNOVATION AND THE 60 YEAR CYCLE

Nikolai Kondratieff, the Russian agricultural economist, published an article in 1926 regarding what he referred to as the Long Wave. He analyzed a variety of of economic variables from several Western countries and identified a business cycle lasting between 50 and 60 years. During downswings commodity prices, workers wages, industrial production, and interest rates tend to fall while during upswings they all tend to rise. Although the existence of the Long Wave is still debated, recent research by H. B. Stewart clearly shows a cycle in total energy use in the U.S. averaging 57 years since 1850. A strong case can be made that when energy use falls below its long term trend line, economic activity is at a cyclical low, while energy use above the trend line corresponds with high economic activity. Interestingly, energy usage peaks based on five year data (1855, 1910, 1970) correlate closely with peaks in long bond yields (1861, 1920, 1981). This U.S. credit cycle of bond yields is presented in Figure 3.

The Kondratieff wave is largely a technological phenomenon. While it is widely agreed that the gains in living standards we enjoy can be attributed to the secular contribution of technological innovations, few are aware that innovations are generated cyclically. Research going back to 1800 indicates that basic technological innovations emerge in clusters, the rate gradually rising and falling over a period of decades. In fact, the rate of innovation introduction reaches a maximum when total energy use troughs and a minimum when energy use peaks. Since energy usage correlates with the level of economic activity, it can be concluded that innovations are introduced into the marketplace in response to declining demand for existing products. Economic activity in the form of consumption continues to fall for 25 to 30 years as resources are devoted to costly R&D projects, producing more and more innovations. Then the rate of basic innovations peaks as entrepreneurs employ the new technologies in a 30 year explosion of new industries that create products consumers want to purchase. This Long Wave effect produces a 60 year boom-bust cycle in the stock market with tops in 1853, 1909, and1968 that coincide closely with energy usage peaks . Notice that the peaks at 1853 and1968 are also 40 year spending wave tops while the 1909 peak falls between spending wave tops.

Why is the technology cycle and the resulting credit cycle 60 years in length? The answer is found in the concept of generations. Various researchers have suggested that people born over a certain fixed period of time will tend to share similar experiences, goals, and values with their cohorts that are distinct from those of people born either before or after that period of time. Typically this period of time is around 20 years in length. One model of generations proposed by Strauss & Howe identifies a repeating cycle of four generational types lasting roughly 80 years. They label the four types idealists-reactives-civics-adaptives with each generation playing a well defined role in the process of social change. Idealists break down established institutions, reactives promote new values, civics build new institutions, and adaptives run these new institutions. Then a new generation of idealists appears and the process repeats. Although this social change model was developed without reference to the birth wave of Figure 2, it is interesting to note that it shows generations arising near peaks and troughs of the wave. Specifically, the last five generations emerge in 1901, 1925, 1943, 1961, and 1982.

Since the technology cycle is 60 years long, it can be argued that while social change occurs over four generations, technological change is a three stage cycle of inventors- entrepreneurs-marketers. The inventor generation develops new technologies, the entrepreneurial generation creates new products and industries from these new innovations, and the marketing generation refines and promotes products into every conceivable niche of the marketplace. Each of the four social types takes a turn as one of the technology types, forming a combined cycle that takes 240 years to repeat. As an example, the well known Baby Boom generation of 1943 to 1960 is in the idealist category in the social change model and the inventor category in the technology change model. This is a highly volatile combination that produces dramatic changes in the world. The last time idealists played the role of inventor was in the middle eighteenth century when they launched what we now call the Industrial Revolution and the Age of Enlightenment. Prior to that, idealistic inventors produced the Scientific Revolution and the Protestant Reformation of early sixteenth century Europe. The current Information Revolution and New Age Enlightenment are products of the Baby Boomers that are likely to be as significant as these earlier movements.

CULTURAL ETHOS AND THE 30 YEAR CYCLE

The secular economic growth of a society is greatly affected by its attitude toward and acceptance of change. If the cultural ethos is conservative, the society will seek to maximize economic security at the sacrifice of economic gain. On the other hand, a progressive culture will endure the uncertainty of change for the sake of greater economic growth. This conservative-progressive tradeoff was best exemplified in the twentieth century by the stable low growth Soviet economy on the one extreme and the volatile high growth economies of first the United States and later Japan on the other.

Research has shown that high growth societies tend to have a high degree of political freedom and to be more individualistic than socially cohesive. By using these two measures, Webber & Wildavsky were able to categorize all societies into one of four political regimes: individualist, authoritarian, egalitarian, or collectivist. Individualist societies are politically free with low social cohesion and are controlled by markets. The United States is the best example of this type of regime. Authoritarian societies combine low social cohesion with political constraint and have strong central governments. The African continent is dominated by such political regimes. In egalitarian societies, social cohesion is strong in an environment of political freedom, leading to a political arrangement of majority rule as is predominant in Scandinavian countries. Bureaucracies control collectivist regimes where social cohesion is strong and political constraints are many. The Soviet Union and many Eastern European countries fall into this category.

Although a society may be dominated by one particular regime, all four forces exist within all societies. In the United States, each force dominates in a rotating fashion, producing a social cycle lasting about 30 years, which aligns well with the 30 year political cycle of public purpose vs. private interest identified by A. Schlesinger. The collectivist period is marked by high regulation and taxes and high government expenditures resulting in an inflationary low growth economy. Figure 4 demonstrates the regularity of inflation peaks over the last two centuries, and R. Batra has documented correlating cycles in wholesale price changes, money supply growth, and regulatory growth. The individualist period follows with low taxes, deregulation, and reduced expenditures that often produces a deflationary high growth economy and a build-up of debt. Another low growth economy follows when the authoritarian period raises taxes and regulation while cutting expenditures further producing a period of economic contraction. The economy booms again in the expansionary egalitarian period accompanied by moderate taxes and regulation and increased government spending. As the economy cycles through the four economic stages of inflation - deflation - contraction - expansion, the markets follow. Commodities dominate during inflation, stocks and bonds in deflation, bonds and cash in contraction, and stocks during expansion.

COMBINING THE CYCLES

When the above three cycles are depicted as simple sinusoids and summed together, the combined economic cycle of Figure 5 is obtained. The combination of 30, 40 , and 60 year cycles produces a cycle that repeats every 120 years. This cycle does an excellent job of following the general trends in the U.S. economy since 1850, registering booms when history has indicated periods of economic prosperity, and busts when the economy has been in wars, depressions, or inflation periods. By reviewing investment performance over the past 150 years, it was empirically determined that each 60 year Long Wave can be subdivided into the following eight economic stages: inflation-deflation-contraction- expansion-???-???-expansion-expansion.

Apparently the four stage social cycle dominates during the first half of the cycle when the Long Wave is in descent. The pattern of inflation-deflation-contraction-expansion is clearly evident in the last two full Long Wave cycles (1857-1883, 1916-1940) as well as the current unfolding cycle (1973-2001?). The phase of the spending wave during this first part of the cycle affects the strength of the various booms and busts. For example, the spending wave was falling during the 1930s which compounded the economic downturn of the Great Depression while it was rising throughout the milder depression of the 1870s. The second half of the Long Wave cycle is less readily predictable, the first two stages being dependent on whether the spending wave is peaking or troughing and on the relative strengths of the three cycles. In the prior two Kondratieff cycles these periods developed as contraction-contraction (1884-1897) and inflation-expansion (1941-1956). The final two stages reflect the dominant strength of the ascending Long Wave, yielding a strong period for the stock market in all of the previous three 60 year cycles (1843-1856, 1898-1915, 1957-1972).

Figure 6 shows the performance of a model portfolio that rotates investments according to the economic stages outlined above. The returns of this portfolio exceed a portfolio of stocks by an average of 2.4 percent annually. Furthermore, the model portfolio exhibits considerably lower volatility risk (beta of 0.6). When performance is adjusted to match stock market volatility (beta of 1.0), the model portfolio yields a return that exceeds stocks by an average of 6.8 percent per year.

FORECASTS FROM THE INVESTMENT CYCLE

The scenario that lies immediately ahead will undoubtedly baffle most economists and market analysts who rely on the existing paradigm for their forecasts. The economic recovery from the 1990 recession has been far more sluggish than their linear models would predict for such a low inflation and interest rate environment. That's because markets, unlike the models that attempt to emulate them, are nonlinear. Market behavior is not determined merely by a set of boundary conditions but also by the history leading up to that set of conditions. Markets are composed of people who possess memories. If a society of people arrive at condition C by passing first through conditions A and B their financial decisions will be quite different than if they arrived at C by passing first through conditions E and D. In the parlance of chaos theory, this is known as a bifurcation.

In terms of the investment cycle, the U.S. economy is currently in the final stage of a contraction wave that began in late 1987. Contraction waves that fall in the descending half of the credit cycle produce particularly severe economic downturns that economists classify as deflationary depressions. These occur approximately every 60 years, the last two appearing in the 1870s and the 1930s. The depression bifurcations are a byproduct of an enormous build up of debt in the economy. When debt expands faster than productivity as it has over the past decade, some sectors of the economy are unable to service their newly acquired debt. This leads to defaults and an increased caution by both lenders and borrowers who have suffered losses. Each upswing in the economy tends to be weaker than the one before as more and more sectors become cautious in the subsequent downturns. At the same time this is going on, technological innovation raises efficiencies in one industry after another as companies faced with buyer resistance invest heavily in cost-cutting equipment. This increase in productivity reduces labor requirements, leading to lower prices and wages. Concern over unemployment compounds the cautious attitude among prospective consumers creating further downward pressure on prices. Falling prices make leveraged investments very unattractive and typically the last year of the depression culminates in a dramatic economic shake-out. Once the debt liquidation transfers assets into stronger hands, the economy launches into a powerful recovery.

In order to determine what course of action we as investors should take, it is instructive to compare the current depression that started in 1990 with those that began in 1873 and 1929. The inflation peaks of 1864, 1920, and 1981 were all followed by a deflationary boom lasting about seven years. Then, within two years of the onset of the ensuing contraction wave, the stock market collapsed in historic one day panics that were labeled with similarly infamous names: Black Thursday on September 18, 1873; Black Tuesday on October 29, 1929; Black Monday on October 19, 1987. In all three cases the market quickly retraced its losses. In each of the first two periods, the contraction wave came to an abrupt end roughly 13 years after the prior inflation peak (1877, 1932). This event was marked by a climactic correction in the stock market, a significant rise in unemployment, a spike in both debt and real long term yields, and a fall in commodity prices. Then an equally dramatic expansion boom lasting about six years immediately followed the shake-out.

There is a way other than counting 13 years from 1981 to anticipate the economic shake-out and stock market correction. While the inflation adjusted yield of the government long bond has averaged just above two percent over the past 120 years, it is generally higher than four percent during deflationary depressions. Any dips below four percent, however, have always triggered a stock market correction in each of the last three depressions. The panic of 1873 at the onset of that depression and the subsequent shake-out of 1877 both coincided with real long bond yields between three and four percent, as did the crash of 1929 which initiated a steep three year long correction that climaxed in the shake-out of 1932. The current depression began with a stock market correction in 1990 after real long bond yields fell below 4 percent for the first time since1981. The twelve month average of real long bond yields once again fell below four percent in the third quarter of 1993. If history repeats, the market will soon begin a correction that will mark the beginning of the final shake-out of the 1990 era depression. Based on market action in the shake-outs of 1877 and 1932, we can expect the S&P 500 to bottom out between 230 and 300 sometime in 1995 with long bond yields rising to nine percent, unemployment also around nine percent, and a twelve month CPI of minus five percent. Gold could fall to $250 per ounce. Then the stock market should begin a bull market fueled by the Baby Boom spending wave that will more than triple stock values by the year 2000.

Many would contend that history will not repeat, that the Federal Reserve is far too sophisticated and the government role in the economy far too great this time around to allow another economic shake-out. As Figure 7 clearly shows, however, the behavior of the Fed in the current cycle is remarkably similar to its behavior during the Great Depression era. And federal government intervention will be severely limited by its own budgetary constraints which will be exacerbated by lower tax receipts as the economy slows.

Note: To understand why the market forecast was too pessimistic, read What Went Wrong?

THE INVESTMENT CYCLE AND POLITICAL CHANGE

The investment cycle is not only useful in predicting major turning points in the financial markets. History indicates a reliable political consequence of the cycle as well. This should not be too surprising since it is widely known that congressional elections are strongly influenced by the near-term performance of the economy as perceived by voters. In the three previous Long Wave cycles, the party in control at the inflation peak suffered large congressional losses in the economic shake-out that followed roughly 13 years later. For example, in 1814 the Democratic-Republican party (Democratic forerunners) controlled 65% of the seats in Congress while the Federalist party (Republican forerunners) held 35%. By 1826, Democratic control was reduced to 44%. The Republicans held substantial majorities at the next two peaks in 1864 and 1920. In each case, the Republicans were the minority party following the elections 12 years later. From 1864 to1876 they lost a 29% share to the Democrats and from 1920 to 1932 they lost 39%.

The most recent Kondratieff inflation wave peaked in 1981. In the congressional elections immediately following in 1982, the Democrats held a 59% share to the Republican's 41%. Based on past performance, we should expect to see the Republicans control at least a 60% share of Congress after the 1994 elections. For that to occur, the Democrats would have to lose a substantial 115 congressional seats.

___________

REFERENCES

Aronstein, M.; Minter, C.; Salvigsen, S. 1991. Apocalyptic Vision: the Comstock Partners see more hard times ahead. June 10 Barron's interview, pp.8-30. Batra, R. 1987. The Great Depression Of 1990. Simon and Schuster. New York. Dent, Jr., H. S. 1993. The Great Boom Ahead: Your Comprehensive Guide To Personal And Business Profit In The New Era Of Prosperity. Hyperion, New York. Gleick, J. 1987. Chaos: Making A New Science. Penguin Books. New York. Howe, N.; Strauss, W. 1991. Generations: The History Of America's Future, 1584 to 2069. William Morrow and Company, Inc. New York. Mogey, R. 1992. The Business and Investment Cycles. Cycles 43:324-329. Schlesinger, Jr., A. M. 1986. The Cycles Of American History. Houghton Mifflin Company. Boston. Skene, G. L. 1992. Cycles of Inflation and Deflation: Money, Debt, and the 1990s. Praeger. Westport, Connecticut. Stewart, H. B. 1989. Recollecting The Future: A View Of Business, Technology, And Innovation In The Next 30 Years. Dow Jones-Irwin. Homewood, Illinois. Webber, C.; Wildavsky, A. 1986. A History Of Taxation And Expenditure In The Western World. Simon and Schuster. New York.

back to the top