first problem concerns periodicity. The waves in the trade cycle move in regular sequences. The shortest, terminating every four or five years, is the one which concerns democratic politicians most. They feel obliged to keep wary eyes on the economic indices in case these predict unfavourable events coinciding with election time. The most important cycle is of 20-year duration. The cyclical trends in the movement of phenomena like population, migration, immigration and house building were first fully elaborated by Simon Kuznets for the USA.6 The importance of these long cycles is that they terminate in slumps, the amplitudes of which are greater than those experienced during the course of the upswing of the cycle. This is because inventory investment and other volatile forms of investment coincide with a downturn in the building programme, thereby creating a severe recession.7 But Kuznets has admitted that, while there is no difficulty in identifying the long swings, there is a problem in explaining them.8 He had to content himself largely with describing the phenomena.
Does the land speculation hypothesis fit into the 20-year cycle? In the 1930s Homer Hoyt, then a post-graduate student at the University of Chicago, investigated the trends in land values in Chicago over the remarkably long period of 100 years. He discovered a regular cycle of 18-year duration.9 His data is considered to represent the general trend in real estate values in the USA over the period up to the 1930s.10 Hoyt has since up-dated his material, and his results are listed in Table 5:I. They fit neatly into the sequence of business cycles. A peak in land values is missing. Its absence may be explained by the threat and advent of the world war, which disturbed the benign psychological outlook which is necessary for speculation. Consequently, the peak of this cycle was eliminated.11
But what if the trends in land values were simply a response to other variables in the economy? We expect to see a rise in land values, for as national income increases, so does the surplus, or economic rents. Rising land values, then, are a derived phenomenon. So how can we justifiably ascribe the primary power to cause recessions to the land market, the income for which is itself dependent upon the functioning of the labour and capital markets ? Is it not possible that a downturn in national income then results in a drop in land values ? This is the popular view: the causal forces are held to be in the opposite direction to the one postulated by Henry George. To prove a causal connection working from land speculation to the wider economy, we need to trace more than ‘a fairly close correspondence’12 between movements in the real estate market and phenomena like the building cycle.
TABLE 5:I
USA 1818-1929
Sources: 1 Homer Hoyt, ‘The Urban Real Estate Cycle - Performances and Prospects’, in Urban Land Institute Technical Bulletin No. 38, 1950, p. 7. 2 G. Shirk, ‘The 18?-Year Cycle in Total Construction’, Cycles, August 1981, p. 149, Figure 1, and C.A. Dauten and L. M. Valentine, Business Cycles and Forecasting, Ohio: South-Western Publishing Co., 1974, p.277. For the booms of the 1830s and 1850s, see also G. F. Warren and F. A. Pearson, World Prices and the Building industry, 1937, p. 99.
The timings favour Henry George’s hypothesis. Table 5:I shows that the peak in land values is reached 12 to 24 months before the economic recession; i.e., the downturn in land values precedes the decline in general economic prosperity. But this chronological sequence is insufficient to prove cause and effect. Landowners may just be blessed with better predictive foresight than the rest of us (though, in that case, why do so many of them continue to buy land just as the speculative bubble is about to burst?).
We need to demonstrate a transmission mechanism, one by which antecedent behaviour in the land market diffuses itself into the factory, office and corner retail store. One such mechanism may be the activity in the construction industry. If land, because of speculation, costs too much, does this curtail construction and thereby dampen activity over a wider sphere of the economy? The American evidence is tantalising. The peaks in the building cycle follow the peaks in land values and precede general economic recessions! This relationship will be investigated in detail in Chapters 8 to 10. First, however, we need a more detailed account of speculative behaviour.
Land speculation is a two-dimensional activity. It is spatial. It entails the acquisition of control over a clearly-defined piece of territory, such as land on the fringe of an urban area, or large tracts on the frontiers of a colonising society. It is also temporal. Purchases today are calculated to provide a financial gain through resale in the future. Thus, the dealer has to be willing and able to hold onto land for a period of time, and sell when he calculates that prices have reached their most attractive heights.
A useful definition of land speculation has been provided by Prof. Botha. It is, he says, ‘an investment over a relatively short period of time in an asset yielding an unrealistically high rate of return accompanied by a relatively low degree of risk’.13 This contrasts land speculation with speculation in other areas, in which the risk of loss is much greater — for example, in the currency markets. We need to know more about the time scales, however, to see how they relate to the 18-year cycles that have been observed in the US economy.
Botha distinguishes three broad categories of time: secular, long and short-term. In the first case, a second or third-generation land monopolist who sells at a hundred times the original purchase price cannot be said to have speculated; the land was not bought with an intention to capitalise on increasing land values within the lifetime of the buyer. An example of this is the Spanish olive grower who owned five acres near Madrid. He and his family fled from General Franco’s army in 1940. The family split itself between London and New York. The parents died, and after the death of Franco the children decided to investigate their ancestral past. They discovered that the five acres that they had inherited were worth £5m. in 1980, just 40 years after the last olive had been picked.14 During that time, the land had remained idle, but the suburbs of Madrid grew outwards and around the site. So far as the land users of the Spanish capital were concerned, the owner of the five acres was doing a Rip Van Winkle act: he had gone to sleep. Still, the value of the land soared with the prosperity of the community. The children, however, while benefiting from their unearned millions, were not speculative owners in the sense which we wish to emphasise for the purposes of this study.15
At the other end of the time-scale, people who buy and quickly re-sell land for a profit usually do so for less spectacular gains. This sharp wheeling-dealing characterises the tail-end of a cycle in land values, when speculative ‘fever’ has gripped a wider circle of people who, perceiving the huge gains made by those who bought for rock-bottom prices, decide to gate-crash the market.
The more interesting time-period, for Botha, is the intermediate phase, his ‘relatively short period’ which, in terms of his three categories, is the long period. He cites the example of someone holding land for 10 years, having bought it at agricultural use values and allowing it to lie fallow in the knowledge that urban expansion will force up the price in the fullness of time.
This intermediate time-scale is characterised by a conscious decision to methodically identify and acquire tracts which will eventually be required for development; and the willingness
