Cause of the Kondratieff Cycle

by Ray Tomes

Background

The idea that there was a slow cycle in prices was put forward by Nikolai Dmytriyevich Kondratieff after whom the cycle has been named. In fact he was probably preceded by Van Gelderen and Beveridge in observing the presence of such a cycle, and it may have been known even in ancient times. The wave has a broad sweep which is partly hidden by shorter term fluctuations and individual cycles vary from around 50 to 60 years.

Kondratieff Cycle Period

The period of the Kondratieff cycle is most often given as 54 years, although sometimes as 53 years of anything from 50 to 60 years. Often the period is just assumed to be 54 years because the data is not long enough for a realistic measurement.

Schumpeter suggested a model in which the four main cycles, Kondratieff (54 years), Kuznets (18 years), Juglar (9 years) and Kitchin (4 years) can be added together to form a composite waveform. In this model the Kondratieff cycle is expected to be made up of three Kuznets cycles, each of which is two Juglars and these in turn are an uncertain number of Kitchin cycles.

Perhaps the best historical determinations have been by Ed Dewey who found a period of 53 years in his analysis of battles and wars along with other periods of 6.0, 9.6, 11.2, 17.7, 22.0, 143 and 164 years. In Dewey’s table of cycle periods he uses a figure of 53.25 years being 3 times his base period of 17.75 years to which many cycles periods are related.

My own estimate of the average Kondratieff cycle period based on the harmonics theory is 53.38 years. This is consistent with the average period of the cycle as found in depressions, unemployment, battles and wars and other series. The accuracy of the determined period at present is only sufficient to state that the cycle probably averages between 53 and 54 years.

Past Explanation

An explanation for the existence of the Kondratieff cycle has often been the cycle of innovation, although no real reason for this having a cycle has been given. It does not explain why the cycle has existed even before modern times when the concept of innovation and invention has really applied. An entirely new proposal was put forward by the Author in the late 1970s when he first re-discovered the cycle for himself and was then told that it was called the Kondratieff cycle.

The fact that war cycles show the same period of 53 years indicates that there may be a partly economic basis to the timing of battles and wars, a not surprising conclusion.

Demographics

When I first did economic modelling on computers in 1977 I was not looking for cycles but rather for regression equations that would allow future values of economic variables to be estimated for businesses to assist in decision making. I was rather surprised to find a bunch of cycles jump out at me in my results. One of these was a very long cycle which only had a single wave in the data period I was using.

This long wave was connected to a number of different things, but the most relevant to its cause seemed to be the variations in demographics. The birth rate was lower in the times when major recessions occured and higher when the economy was stable. These birth rate fluctuations affected the whole age distribution of the population, meaning that lumps of people with similar ages exist until those people die of old age. The most recent lump is called the post war (WW II) baby boom.

As well as fluctuations in demographics and prices, this wave showed up in the New Zealand economy that I was studying in terms of trade, which is the relation between import and export prices, and in migration rates. Because NZ was mainly an exporter of meat, this variation was attributable to the cycle of depressions as people buy less meat at such times. Likewise, migration variations can be undertood as resulting from terms of trade changes as people prefer to leave rather than travel to a country with poor economic performance.

Lesson in Depreciation of long lived assets

Before proceeding to the cause of the Kondratieff cycle it is necessary to put forward an explanation of how assets which have long lives have high leverage on economic conditions. For a commodity such as food, if the population fluctuates by 1% then we need a corresponding variation of 1% in food availability and therefore a 1% variation in food production.

The same thing is not true in the case of a long lived asset such as housing. Because houses last for a very long time, a population increase of just 1% will cause a much larger increase in the housing demand, probably around 50%. If we had a stable population and houses lasted for 100 years then we would only need to replace 1% of houses each year. However populations are generally increasing so that each year more like 2% must be added to the stock of houses. This varies from country to country depending on natural population increases and migration patterns. Regardless of these variations between countries, in all cases the effect of a difference between one year and the next of a 1% variation in population growth rate will mean a many times greater effect on house building.

The same thing applies to commercial building and other business setup costs. The capital formation associated with even small population changes is very dramatic. This is an economic effect of migration that is not widely understood. We hear claims that migrants take away local jobs. Actually they cause enormous increase in demand for all capital stocks – housing, jobs, cars etc. that takes about 10 years to catch up on.

When we consider the natural fluctuations in birth rate and what happens to these population lumps as they age we may note two times when the effects are especially great. The first is when people reach an age of about 15 to 25. During this period most people get jobs, cars, houses and household equipment. It makes no difference whether they own the house or business or someone else does. Either way there is increased capital formation. The second time when there is an effect related to age is when people retire. The effect of this is not as great as the first period, because this group generally already has the necessary capital stocks that they need and may in fact move into a smaller, easier to maintain, house. Rather, the effect is due to stopping active employment and therefore not contributing to the production side of the equation. Actually the same is true of children, but because they generally live with their parents, they do not require any expensive capital expenditure.

It so happens that these two periods of higher expense and less production are about 50 years apart, so that successive Kondratieff cycle peaks in population combine to have an even greater effect. But what is this effect and what drives the cycle?

Proposed Cause

When the impact of population age group lumps on capital formation is understood then the cause of the Kondratieff cycle and its period can also be understood. In a period like the 1930s which had a serious depression, the birth rate is much lower than at other times because people cannot afford to have large families when they are struggling to get jobs. The effect of this lower birth rate is then felt about 20 years later when a lower than normal proportion of people are reaching the 15 to 25 year age group which is associated with large capital formation requirements for jobs, cars, houses and household equipment. That means that the resources are not stretched supplying capital items and so resources can be increasingly diverted to immediate provisions. This results in a healthy economy that existed in the 1950s and early 1960s. Full employment is easily achieved and the prosperity that results leads to larger family sizes and increased birth rate.

This prosperity then sows the seeds for the reverse side of the Kondratieff cycle. The higher birth rate means that come the 1970s and 1980s there will again be more people reaching the ages that need jobs, cars, houses and everything else and that jobs cannot be created rapidly enough. This returns to a time of high unemploment as in the 1930s although the depth of the recession was not as great. Again, people respond to the difficult economic times by having smaller families. Again this will influence conditions a generation later and lead to prosperity and full employment in the 2010s.

Although the demographic effect on capital formation requirements is the key aspect of the Kondratieff cycle which determines its period, it would possibly die out if there were not also some external forces that were driving the oscillation. There is some hint of a weather cycle of about the same period which may also be relevant. This is an area were further investigation is justified.

Future Kondratieff cycle

The forecast for the next couple of decades is healthy with the best conditions likely to occur around 2014 but with a decade either side being healthy in terms of employment. The next trough should be centred on 2040 or thereabouts. Such a trough could be reduced in its severity if people understood the Kondratieff cycle and did not make more babies over the next 20 years just because times are good now. If they want to do a favour for their children, then having less children will mean that they will have a better chance of full employment and ease of obtaining houses.