CHAPTER 10

The Mystery of the Business Cycle
PBOBABILITT THEORT GIVES US A GREAT MANT GLIMPSES into the future, and is therefore an excellent method of forecasting — ^particularly for people who do not live by the clock. While probability can tell us little about the temporal sequence of events, it teaches us much about their juxtaposition, so that we can see a host of like individual processes as a recognisable pattern rather than a kaleidoscope.
Admittedly, probability can also be applied to fixed sequences of events, as for instance in genetics, where fathers invariably precede their sons. But, by and large, time sequences play no more than a secondary role in probability theory, which is static rather than dynamic, and considers constantly recurring, rather than constantly changing, events.
It is for this very reason that probability theory, despite its doubtful antecedents, enjoys the favours of conservative economists and sociologists, while radicals reject or simply ignore it. According to probability theory, all human aggre- gates follow a definite pattern: only the fewest of men can be rich, can reach very old age, or attain great fame, nor will this distribution be altered by revolutionary changes in economics, geriatrics, or politics.
The law of income distribution
In economics, this doctrine is associated with the name of Vilfredo Pareto (1848-1923), who was convinced that the existing distribution of incomes was the soundest possible basis of economic and social equilibrium. But even if it was not, little could be done about it, since the distribution was governed by an immutable mathematical and mechanical law, as incontestable as any law of physics.
261
252 CHAPTER 10
Pareto^ did not derive his conclusions from observations of the largest group, i.e. people with small incomes, for this might have been embarrassing. Instead, he divided society into an uneven number of equal groups — a very small number of groups representing very high incomes, and a very large number representing very small incomes. Graphically, this distribution could be represented by a splendid curve, and — on the logarith- mic scale — by the straightest of straight lines.
The elegant result, on paper, was taken for proof positive that national incomes were not distributed by mere chance, but that they were governed by a universally valid natural law. In these circumstances it was deemed unethical, not to say unnatural, to suggest reducing very high incomes. Moreover, Pareto had clearly shown that any such attempts to upset the "natural" economic order would simply lead to a few groups moving up, at the expense of others, but would not affect the total picture to any appreciable extent.
This remarkable "discovery" made its appearance in economic textbooks throughout the world, and came to be known as Pareto's Law. Though it is still upheld by some economists today, it was long ago shown to be, and always to have been, statistically inaccurate. ^ Had it been true, all attempts to improve the economic lot of the poor at the expense of the rich, would have been doomed to failure from the start.
According to Pareto, not even social revolutions could change the economic order, for all economics were said to have an inherent and permanent mechanism which, being greater than any one social system, inevitably led back to the status quo ante. All coercive re-allocations of the national income were therefore deplorable interruptions of an otherwise harmonious state of equilibrium.
Unreal and psychologically fallacious though Pareto's theory was, it nevertheless contained a grain of truth, for redistribu- tions of incomes are generally nothing else than re-allocations of the national cake between different, though numerically equal groups, and have little effect on the population as a whole. It is simply a case of Peter paying Paul.
Another factor militating against successful social demands
^ V. Pareto: Cours d'Economie Politique ( 1897).
2 Eric Roll: A History of Economic Thought (London 1945), p. 414.
THE MYSTERY OF THE BUSINESS CYCLE 253
for a higher share in the national economy, is inflation. Very significantly, most people simply ignore the fact that wage increases can be offset by lower profits, and take it for granted that, as wages go up, so prices must rise. Inflation has therefore become an orthodox method of maintaining social equilibrium — Paul robs Peter to pay Peter a generous increase, and Peter is delighted. Monetary "reforms" of this kind are a cheap way of keeping people contented. Thus when Franklin D. Roosevelt launched his New Deal, which involved far-reaching social and economic reforms, the proportion of the U.S. national income represented by wages and salaries went up by barely one per cent.
Pareto was therefore not quite so silly as he sounds to many mc/dern economists. Even in the Soviet Union, where economic distinctions are admittedly smaller than they are in capitalist countries, there is a marked tendency to widen the gap. While maximum differences in real wages were 10:1 (apart from very exceptional cases) in the first 20 years after the Revolution, differences of 50 : 1 are fairly common today. No one seems to mind that leading scientists earn fifty times as much money as unskilled workers, and, in fact, the ratio seems to be growing greater still.
The discovery of business cycles
Pareto' s Law was one of the last great attempts to set up a universally valid social and economic law based on statistics. Since all such attempts have generally proved to be utter failures, economists have increasingly turned away from them. John Maynard Keynes and his disciples make a point of shunning all general prognoses and look upon economic processes, not so much as tending towards equilibrium, but as being characterised by marked deviations, particularly right at the top and right at the bottom of the social scale.
These deviations can be modified and even avoided, if the consequences of unemployment, investment, inflation and deflation are appreciated in time. All these processes are inter- related, and economists, like physicians, must occasionally prescribe antidotes. Naturally, it would be even better to prevent
254 CHAPTER 10
the illness from starting in the first place, but economists are not primarily prophets, and all they can do is to tell mankind that certain causes have certain effects, and how the worst effects can be prevented.
The real prophets among economists are the believers in cycles who, having observed that certain economic phenomena — and particularly crises — recur with great regularity, conclude that these phenomena are periodic. Thus, they confidently assume that the economic picture of the future will be a repeti- tion of the past.
When periodic theory made its first appearance at the beginning of the 19th century, economics was still a young science and, like all young sciences, it was very self-conscious. Ever since Adam Smith in his Wealth of Nations^ had extolled the advantages of free trade, it was a general axiom that, if only reason and freedom were allowed to replace all economic restrictions, things were bound to become better and better. Since then it has appeared that free trade is not quite the good fairy Adam Smith had made it out to be; it leads not only to merciless competition — ^which was deemed most desirable — but also to most undesirable crises from which all the contestants emerge as losers.
Cycle research
Not even the most orthodox economists could afford to ignore this incontestable fact, though they continued to hold fast for as long as they could to the French economist Bastiat's contention that free trade must needs lead to complete harmony.^ As crises multiplied, however, even the most die-hard conservatives realised that the combined acts of human volition which governed free trade, were in turn governed by what appears to be a superhuman force that played havoc with man's plans at periodic intervals.
Economists now faced an entirely new task. Even if they could not tell precisely how to avoid the next economic crisis,
^ Adam Smith: Inquiry into the nature and causes of wealth of nations
(1776).
2 C.-F. Bastiat: Harmonies economiques (1849).
THE MYSTERY OF THE BUSINESS CYCLE 255
they could at least observe and record these periodic upsets, and use their records for predictive purposes. A new branch of economics was bom — Business Cycle Research — whose main task it was to determine which way the economic wind was blowing.
But before this task could be tackled effectively, an untold amount of theoretical and statistical spade work had first to be done. Though papers and records on business cycles had reached mountainous proportions, no individual theory had proved its absolute worth. So, statistical research proceeded apace and economists vied with one another to collect better and bigger data. Business cycle research, being largely historical and retrospective, can only foretell the future once it has probed the past as fully as possible.
Today, hundreds of public, semi-public, and private institu- tions all over the world devote themselves exclusively or predominantly to business cycle research. In addition, many banks, industrial companies, and other businesses have special offices devoted to the same task. If we counted all the statisti- cians and other staff who are directly or indirectly involved in this work, we should find that they far outnumber the staff of meteorological institutes.
As so often happens, quality has not kept up with quantity. Though cycle research in general has become mere routine, some of its offshoots are extremely unsalubrious. Particularly in America, there is a regular Stock Exchange branch whose practices are very reminiscent of those of soothsayers. Then again, cycle theory has often degenerated into mere scholastic hair splitting, whose practical results are in no relationship to the amount of work that has gone into it. But, all in all, the achievements of this young branch of predictive science are far from insignificant, and it would be wrong to dismiss it, simply because of such false prophecies as those made on the eve of the 1929 crisis, or immediately after the Second World War.
Crises The fact that modern cycle research began in the 19th century.
256 CHAPTER 10
does not mean that the idea of cycles in nature is of relatively recent origin. Since time immemorial, farmers have been keenly interested in cyclical phenomena and in arranging their work accordingly. Planetary orbits, the weather, the seasons — in short everything sent by the heavens, has a more or less c}'clical character, and human beings working out of doors could not fail to notice it.
In other respects, too, cycles have always played a large role in agriculture. Thus arable land had to be left fallow from time to time (which later gave rise to the practice of the rotation of crops) for, just like men or beasts, the soil needed time to recuperate from strenuous labours. The fact that reckless avarice does not pay dividends on the land, was known to every farmer.
In industry, however, everything seemed to be different. The only rhythmical interruptions here, were a minimum of time for sleeping and a day off for observing the Sabbath. Otherwise, work went on continuously, particularly after the invention of the steam engine when interruptions of work meant letting the boilers go out. Continuity of production became an economic principle, and since machines were becoming better and better, there seemed to be no limit to what man could produce. But then, quite suddenly, the limit was nevertheless reached, not admittedly on the productive, but on the distributive side. The greater the quantity of goods which poured from the factory, the more difficult it became to dispose of them. The first such great interruptions of the productive process were significantly called "trade crises" to show that trade, rather than over-production, was to blame. Clearly the fault lay with the customers, who refused to buy what was being offered them. Only a few imper- tinent outsiders, like the Swiss economist de Sismondi (1773- 1842)^ blamed over-production and low wages, but public opinion and more orthodox economists paid no heed to him.
The question when the first of these crises occurred is still very hotly contested. While it was formerly thought that the great distributive slump of 1816, on which de Sismondi based his theory of crises, was the first economic crisis in the modern
1 Simonde de Sismondi: Nouveaux principes de I'Economie politique (1819).
THE MYSTERY OF THE BUSINESS CYCLE 257
sense, more recent investigations have shown that crises had occurred as early as the 18th century. Still, these very early crises could not as yet have been periodic and were certainly local affairs. Thus the years 1787-1793, i.e. the time of the French Revolution, were years of great prosperity in England. In what follows, we shall therefore concentrate on more recent crises, and begin with the general depression which followed immediately after the Napoleonic wars. This crisis, although apparently due to extraordinary causes, was followed by a similar upheaval in 1825, i.e. after ten years of peace, by another in 1836, and by a third, affecting the whole of Europe in 1847. Crises therefore seemed to recur every nine to eleven years, each lasting for two to three years, each followed by gradual improve- ments, and some by exceptional prosperity. But crises rather than prosperity were in the public's eye, since crises caused bankruptcies, public scandals, dumping of goods, mass-immigra- tion, suicides, political disorders, etc. Thus the idea of business cycles was not yet born, and people spoke of periodic crises instead, in much the same way that they had previously spoken of epidemics.
The Juglar-cycle
The first man to mention business cycles as such was the French economist Clement Juglar who, turning from his studies of periodic population phenomena — ^marriages, births and deaths — to an investigation of fluctuations in the rate of interest of the Banks of France and England, discovered a surprising rhythm in them.
In 1 860, after ten years of further study, Juglar published the results of his observations and calculations in a book that said far more than its title, Des crises commerciales et leur retour periodique en France, en Angleterre et aux Etats- Unis ( Of com- mercial crises and their periodic return in France, England, and the United States), would have led one to suspect. The book propounded the revolutionary doctrine that, far from being isolated events, crises were mere phases in long business cycles, whose purpose it was to slough off the dead tissue accumulated during the previous period of prosperity. In other words, crises
258 CHAPTER 10
were mere trials by fate (the original Greek meaning of the word) during which the wheat was separated from the chaff. While individuals might suffer, the economy as a whole was purified and strengthened by this somewhat painful, but altogether wholesome, purgative.
Like so many other important discoveries, Juglar's business cycle remained unnoticed for quite some time. Even so well- read a man as Karl Marx ignored it completely when, seven years later, he developed his own revolutionary theory of crises in his Das Kapital. True, there were basic differences between the two economists, for while Juglar's theory was bourgeois — he had no wish to destroy the capitalist system simply because it led to periodic crises — Marx looked upon crises as proofs posi- tive of the degeneracy of that system. Where Juglar saw the economy moving forward in cycles of equal duration, Marx saw one gigantic downward spiral with crises recurring more and more frequently, and becoming more and more violent until rock bottom, and the end of capitalism, were reached simultaneously and inevitably. Then the workers would usher in the new age of socialism, and the spiral would finally come to an end.
The next hundred years were to prove Juglar right and Marx wrong. While some crises were admittedly much more violent than the terrible crisis of 1857 on which both economists had largely based their theories, others were not. Only in one respect was Marx proved right. Crises in the second half of the 19th century and the early 20th century occurred more fre- quently than they had during the first half of the 19th century, but even then the spiral did not narrow a great deal: the interval between crises was reduced from 9-11 years to 7-9 years.
Moreover, each crisis now seemed to have a different cause. In 1866 the cause was a financial miscalculation, in 1873 a pro- perty speculation, in 1907 a credit squeeze, and in 1929 wild stock exchange speculation. Each crisis looked different, with the result that a host of theories was put forward, only to be disproved on the next occasion.
Only one point seemed to be generally accepted: crises were a regular symptom of the economy. True, in times of prosperity this fact was glossed over and many people deluded themselves with the hope that periodic crises were gone forever, parti- cularly during the 1920s when Americans were convinced that
THE MYSTERY OF THE BUSINESS CYCLE 259
the golden age was here to stay. From now on, depressions could be met with reduced profit-takings, and all was well in the best possible of worlds. And so it was until — 1929.
Although many European economists were also deceived by the American "economic miracle", not all were equally credu- lous. It was precisely in the 1920s that Business Cycle Research came into its own, and began to carry out detailed investigations of even the smallest market and interest fluctuations, in order to derive generally valid laws from them. When the 1929 crisis finally came, even the greatest American "miracle" economists had to admit that their European colleagues had been right and that the old doctrine of business cycles was as valid as ever it had been. Thus, on the eve of the Second World War, economists on both sides of the Atlantic were agreed that every boom was invariably followed by a slump. The only question was whether the next slump might not be so severe as to put an end to pros- perity for ever. Perhaps Marx rather than Juglar had been right after all.
Kitchins and Kondratieffs
Meanwhile, business cycle theoreticians had come up against yet another difficulty. Although 7-9 year Juglar cycles existed beyond all doubt, these cycles alone could no longer explain the total economic picture. Perhaps there were other cycles in addi- tion to Juglar's, interfering with it in much the same way that sound waves interfere with one another — ^producing higher crests here, and lower troughs there. The first positive dis- covery of this kind was made in the 1920s, when Joseph Kitchin^ managed to isolate independent cycles, each lasting about 40 months, in the economy of Britain and the U.S. from 1890- 1922. A parallel investigation by the American W. L. Crum, which went back as far as 1866, came to much the same result.
These shorter cycles followed similar courses to their big brothers'. Here too, long periods of prosperity were followed by sudden reversals in which the stock market slumped, consumption fell off and unemployment increased for a number of months, ^ Review oj Economic Statistics, January 1923.
260 CHAPTER 10
until the pendulum gradually swung back again. The existence of such miniature cycles could be demonstrated in all phases of the Juglar cycle, on whose basic rhythm, however, they had no effect. Thus even during the prosperous 1920s the Kitchin cycle caused small depressions in 1924 and again in 1927 which, though no real crises, led to temporary economic set-backs, for which the term recession was coined. From then on, business- cycle theoreticians would be at odds in deciding whether business set-backs were the beginnings of a new crisis or merely of a new recession.
In addition to these shorter cycles, economists also discovered very long, historical, cycles, the so-called "great waves", each lasting 40-50 years, and each characterised by an upward and a subsequent downward trend lasting from 20-25 years. For instance, in Germany there was an upward swing lasting from 1843 to 1873, and again from 1895 to 1913, while there was a depression from 1874 to 1894. The first man to analyse these long periods systematically was N. D. Kondratieff,^ after whom the cycles came to be known as Kondratieffs, to distinguish them from the shorter Juglars and Kitchins.^ According to Kondratieff a "long wave" lasts for almost forty-five years, though other economists think that they may last fifty or even fifty-four years. ^
Nor are economists agreed on the historical beginnings of these long cycles. While the first long wave reconstructed from reliable statistical material is said by some to have begun at the end of the 18th century and to have lasted into the 1840s, the second to have lasted into the 1 890s, and the third up to the end of the Second World War or until a few years later,* others claim that the two world wars must be ignored since they merely interfered with the smooth functioning of the cycles.
Some modern historians, and Toynbee^ in particular, have tried to relate long business cycles to political history and
1 N. D. Kondratieff: Die langen Wellen der Konjunktur in Archiv der Sozialwissenschaft, December 1926.
2 J. A. Schumpeter: Business Cycles (New York 1939), p. 169.
3 Simon Kuznets: Economic Change (N.Y. 1953), p. 109.
* E. R. Dewey and E. F. Dakin: Cycles, the Science of Prediction
(N.Y. 1947).
6 Arnold Toynbee: A Study of History, Vol. IX, p. 322.
THE MYSTERY OF THE BUSINESS CYCLE 261
especially to war-and-peace cycles. Unfortunately, none of these efforts have led to tangible results. At most, we can agree that long wars — e.g. the Napoleonic wars, the wars of national liberation in the 1850s and 1860s, and the First World War — coincided with upward trends of the three corresponding Kondratieff's cycles, while the Second World War did not (unless the third long wave is allowed to end in 1940 rather than in 1945).
Many leading economists rightly object that two or three parallel events occurring at intervals of 45 or 50 years are no proper basis for a cycle theory, or for economic predictions. In any case, long waves are of small practical use, for neither governments nor private enterprise can generally afford to plan for so many years ahead.
Of pigs and rubber plants
Far more important for practical purposes are the cycles govern- ing individual branches of the economy. Thus United States economists have discovered an eighteen-year cycle for the building trade, though Dewey and Dakin's prediction that this cycle would reach a low in 1952 has not been fulfilled. Building construction in 1952 was not only very much more intense than before the war, but 20% higher than in 1947, when the predic- tion was made. 14-8 year cycles are said to govern cattle and black pepper, 23 months cycles to govern the American textile industry, 34 months cycles to govern the sugar market, etc. Almost every important industry has at one time or another been said to have a cycle, though few such cycles have ever proved reliable prognostic criteria.
There is much better evidence in favour of cycles in which two independent economic phenomena follow each other at regular intervals, and in which one phenomenon can be predicted from the occurrence of another.
One of the most famous cases is the pork-price cycle discovered by the German Institute of Business Cycle Research in the 1920s. Statisticians had noticed for many years that the price of pork underwent marked periodic fluctuations, not, however, all
262 CHAPTER 10
of the same intensity. As a rule, prices, within a cycle, fluctuated by from 30-50%, quite independently of the general state of the market. Since the demand for pork was fairly constant, these fluctuations were clearly the result of changes in the conditions of supply.
In fact, every 40 months or so, farmers would either flood the market with pork or else cause an acute shortage. The reasons are quite simple: farmers are usually adaptable people, and if pigs fetch high prices they switch to pig rearing. Now it takes about 15 months to rear a pig for market, and a prior few months to reorganise a farm. Thus roughly 1 8 months after the market began to boom, a glut of pigs starts to pour into the butchers' shops. Butchers, in turn, realising that pork is becoming more and more plentiful, reduce their prices until pork rearing is no longer a profitable business. The farmers devote their attention to other livestock or crops instead, until there is a new scarcity of pigs, and so ad infinitum.
Although most pig breeders ought to have learned their lesson, they could never resist the lure of high prices. Even when business cycle experts explained to them the error of their ways in detail,^ they took little notice and continued to breed pigs as they always had done — too much or too little. The Institute itself, however, had scored a great hit, and for many years the German Institute for Cycle Research was able to predict pork prices very accurately, 12 months in advance. Only in 1933, at the peak of the general economic crisis, did their predictions fail, because millions of people could no longer afford to buy pork, i.e. the cycle collapsed for an entirely different set of reasons.
The history of the pork cycle is one of the more impressive pages in the annals of serious economic prognosis. Even so, it would be wrong to base general economic laws on it. The cycle only worked because pig breeders acted according to classical laissez-faire axioms: high prices — many pigs, low prices — ^few pigs. Now, this axiom cannot be applied universally, even to completely free agricultural economies since the peasants of many primitive countries would never dream of increasing production when prices rise. If they manage to obtain more
^ A. Hanau: Die Prognose der Scbweinepreise. Special edition of Viertel- jahrshefte zur Konjunkturforschung (Berlin 1928).
THE MYSTERY OF THE BUSINESS CYCLE 263
money for their products than they require for their immediate needs, they much prefer to take things easy for a bit.
This is precisely what happens in Malaya: if rubber prices rise, the Malay native taps fewer trees, but when the market falls, he is forced to exploit his plantation to the full, since otherwise he and his family would go hungry.^ On the other hand, in parts of Indonesia where the main crop is rice, and rubber is only a secondary industry, the natives are extremely hard- working and respond quite differently to market fluctuations. The moment the rubber prices rise, they tap the trees very heavily to get the best of the boom. Thus, in the case of rubber whose price depends on the international market and a great many regional factors, it is very much more difficult to predict price fluctuations than in the case of German pork.
Trading hopes
The world's most important raw materials are usually offered for sale in Commodity Markets, ostensibly to safeguard the producers against sudden price fluctuations. In fact, actual sales represent only a fraction — and often an insignificant fraction — of the exchange transactions, the vast majority being dealings in futures. Thus American commodity exchanges will sell you coffee and cocoa beans, wheat, cotton, and 1001 other articles, that have not even been planted.
A great many of these articles are, in fact, never delivered at all. When the agreed delivery date is reached, the vendor simply pays the purchaser the difference between the agreed price and the ruling market price, or, if the market has dropped, he collects the difference himself. Thus commodity transactions are, in fact, speculations with future markets rather than with tangible articles. The actual risk to the individual is smaller than it might appear, for, whenever the market seems to be going against them, speculators can usually sell their risks to somebody else.
Thus dealing in "futures" are not, in reality, what their name implies — anyone can contract out of this kind of future at any time. In this way "futures" become short-term transactions, and ^ J. W. F. Rowe; Markets and Men (Cambridge 1936), p. 129,
264 CHAPTER 10
speculators can always try to sail with the wind. On the exchange floor the distant future contracts into hours or minutes — a newly acquired "future" can be disposed of at any moment.
However, since such transactions are somewhat costly and not entirely without risk, even professional speculators prefer to stick to their "futures" for at least a few weeks, and are therefore keenly interested in predictions.
Psychology plays a greater role in stock exchange predictions than it does in most other business spheres. Prices are not so much dictated by urgent requirements of vendors or purchasers, as by assessments of the public mind. Thus, buyers always believe that public demand for certain shares will rise, while the vendors always think the precise opposite. If this were not the case, there would be no need for stock exchanges at all, since straightforward investments could be handled very efficiently by banks and other public institutions.
The stock exchange is therefore a market for hopes and expectations, in which the future is bought or sold at a discount. Needless to say, this involves attempts to look into the future, and the stock exchange is fully conscious of this prophetic task. Each working day, a consort of prophets assembles on its floor in order to determine what quotations will result from their joint efforts. Their ability to do so seems — to them, at least — beyond any doubt, for jobbers rarely look upon themselves as mere gamblers.
Speculators, unlike gamblers, believe that they act with due consideration and foresight. Speculation is a kind of planning, even though the speculator himself rarely influences the success or failure of his plan, but rather participates in somebody else's — a company chairman's, nature's, an engineer's, or a govern- ment's. These, and a host of other factors, are the subjects of his speculations, and since their outcome is rarely known in advance, the successful speculator must in some way become a prophet.
Before he can do so, he must know as much as possible about the shares he proposes to deal in and also about the general economic outlook — it is unusual for individual shares to run counter to the general trend. They may rise steeply while the market itself rises slowly, or drop steeply in a weak market, but only exceptionally will they fall during a boom or rise during a
THE MYSTERY OF THE BUSINESS CYCLE 265
slump. General economic trends are therefore the most important subject of all stock exchange predictions.
n^all Street Prophets
Particularly in America, there are a number of special offices and publications exclusively concerned with predictions of stock exchange, as distinct from general economic, trends. While the two are obviously related, the stock exchange may react badly where the economy as a whole reacts well. Thus while the economy needs large reserves for future expansion or building, the stock exchange usually prefers large dividends. Tax reforms, too, can be an incentive to production, and at the same time have a depressing effect on the market.
Above all, political news and rumours influence the stock exchange much more strongly than they do the rest of the economy. The stock exchange is a most sensitive barometer of current trends, and responds to the slightest change. A day on which the market does not change is most unpromising, and if this tendency persists the market usually turns dull — ^prices begin to fall. Some speculators invariably contract out of a dull market, and it is therefore important to keep the exchange seething with excitement. Politics supply the most excellent stimulus.
Professional experts vie with one another in dishing up political sensations, and particularly "inside" information not known to the editors of daily newspapers. Thus the most sought- after stock exchange bulletins are usually headed "strictly confidential", and the source of their information is often the editor's own fertile brain.
Some stock exchange prophets prefer to consult the spirits, instead, and one of the most famous Wall Street forecasters of recent times, Frederick N. Goldsmith, was wont to consult a comic strip published by one of the American dailies. Other readers might merely laugh at the strips, but Goldsmith knew better — the spirit of a deceased Wall Street speculator had given him the key to their hidden meaning. Nor were these strips the sum total of his inspiration, for, thanks to a gifted
266 CHAPTER 10
medium, he was in daily contact with the spirit of the late John Pierpont Morgan — ^no wonder that 90-95% of his tips were said to be correct and that subscribers to his circular letter paid him 25 dollars for the privilege of sharing his inspirations. His annual income from these activities was 39,000 dollars, and he was allowed to continue his beneficent activities for 50 years. Only when he had reached the ripe old age of 83, did the authorities begin to take note and to put a stop to his remarkable performance.^
Now, secret information from the spiritual realm is the exception rather than the rule, even in Wall Street. Most speculators prefer more down-to-earth information and, because nowadays so much depends on the decisions of governments, Washington has become an important centre of tips even for the New York Stock Exchange. No dealer can afford to ignore rumours with an authentic Washington stamp, and rumours, true or false, have become part of their stock-in- trade.
Most conscientious experts have begun to classify all rumours according to their degree of credibility, and one of the most prominent American sheets dealing with this kind of information which, by the way, is not exclusively addressed to stock exchange speculators, classifies its information by means of ticks. Thus, a doubtful and unverifiable story is given one tick, one that seems less doubtful, two ticks, and one that has been authenticated by the editor, three ticks. ^ Even so, the rumour that Eisenhower would not seek re-election was given three ticks just before he stood for nomination, and so was the story that France, after the electoral successes of the Communists and Poujadists in Spring 1956, was closer to a revolution than she had been at any time during the past eight years. Another three-tick story was that Israel and Egypt would go to war — seven months earlier than they did.
As one can see, such predictions are anything but reliable, and often look ridiculous in the light of subsequent events. Neverthe- less, it is a fact that not only on the stock exchange but also in normal business, men often base important decisions on assump-
1 Time, 27th October 1948.
2 Personal from Pearson. A weekly interpretative Washington News Letter edited by Drew Pearson (Washington 1956).
THE MYSTERY OF THE BUSINESS CYCLE 267
tions and rumours that are subsequently proved to have been utterly false. Luckily, in stock exchange speculations at least, it does not so much matter whether a rumour is true or false — the only thing that counts is how many people believe in it. For this very reason, diplomats and statesmen who try to use private information for playing the market, often lose their shirts. True, they have advance knowledge of political events, but they know little about the psychology of the stock exchange and therefore derive false conclusions from the correct premisses. A classical example was that of Friedrich von Holstein — one of Bismarck's closest confidants — who played the market so un- successfully, that despite all his private information, he eventually owed his bank 300,000 marks. ^
The postman always knocks twice
Since private information, however reliable, is a very in- adequate a key to success in the stock exchange, American fore- casters, in particular, have tried to use more "reliable"criteria. Illogical though price fluctuations may appear to be to the layman, they must surely have a logic of their own. Thus if one ignores individual ups and downs and investigates over-all fluctuations instead, clear periodic patterns are said to emerge.
For instance, if a dull market suddenly becomes cheerful, a slight reversal may be expected three days afterwards. This is called a "technical reaction" in stock exchange circles, though it is really a psychological reaction: investors who have had to wait a long time to sell their shares at a profit take the first opportunity to do so. Conversely, a sudden fall in the market is often followed by a slight improvement within a few days, when those who have waited to buy shares at a lower price suddenly step in. However, these small and transitory reactions have no bearing on the main trend or on the general economic situation, nor are they regular enough to be relied upon. Sometimes, the
^ F. von Holstein: Stock Exchange Correspondence edited by Ernst Feder (Berlin 1925) — R. Lewinsohn and F. Pick: Sinn und Unsinn der Borse (Berlin 1933).
268 CHAPTER 10
main trend may continue for up to a fortnight without any reaction, but, by and large, jobbers have done well by taking this factor into account.
Far more questionable is the problem of more persistent stock exchange patterns. Before the war, many Wall Street experts thought they could detect a 40-day cycle of slight reversals in the general trend. This cycle offered the skilled manipulator excellent and repeated chances of profit-taking. Of late, the 40-day cycle seems to have been discarded by many in favour of a 24-day cycle.
A more persistent American belief is based on the so-called Dow-theory, propounded by Charles Dow and S. A. Nelson at the turn of the century. According to this theory, successful stock exchange speculators have only to watch the movements of industrial and railway shares very carefully. If both groups of shares begin to rise simultaneously, a boom is imminent, if both drop simultaneously, a lasting depression may be expected, and if both move in opposite directions, the future is altogether unpredictable.
The predictions made for many years by Dow's pupil, William Hamilton, the editor of the Wall Street Journal, on the basis of Dow's theory, have been right 50% of the time and wrong just as often, ^ but the prophecies of most other Wall Street prophets fall far short of even this sparse result. In other words, stock exchange speculation might just as well (or better) be based on the toss of a coin. An investigation of 7500 predic- tions by 16 well-known "forecast bureaus" has shown that the vast majority was wrong. ^
The Dow theory in its "classical" form is, at best, a tautology. Far more interesting is a variation on the same theme, based on the well-known fact that the postman always knocks twice. In America, stock exchange fluctuations have always been looked upon as the struggle between optimistic bulls and pessimistic bears. Whenever the bulls fail to drive prices beyond the level that held before the last downward reaction, the time of the bears has clearly come, i.e. the market will drop. Conversely, if after a small upward reaction, prices do not drop below the former
1 G. Leffler: The Stock Market (N.Y. 1951 ), p. 522.
2 A. Cowles: Can Stock Market Forecasters Forecast? in Econometria Vol. I, No. 3 (July 193S).
THE MYSTERY OF THE BUSINESS CYCLE
269
30. Wall Street quotations of industrial shares. A long upward trend ends with a "double top".
level, the bears are out of office, and the bulls are about to take
over.
On stock exchange graphs, these trends are quite clear to see. Thus if, at intervals of a few weeks or even a few months, there appear two crests or troughs of equal intensity, it is usually safe to say that the boom or slump is over. But like all such rules, this one, too, fails from time to time — some postmen leave after only one knock. Booms may turn into slumps without any prior warning, and vice versa. In that case, stock exchange graphs, instead of showing two highs or two lows, will show anything else you like. But, within certain limits, postmen generally persevere, and double knocks — "double bottoms" or "double tops" as American experts call them — are, in fact, tested means of predicting not only stock exchange fluctuations, but also of predicting such mass-psychological phenomena as politics or even fashions: if a trend reaches two successive peaks, it is often the beginning of the end.
270 CHAPTER 10
The cycle-barometer
Significant though the specific psychological trends of stock exchange fluctuations are, the market as such is also influenced by the general economic situation, and market trends opposed to general trends can never be of anything but the shortest duration. To assess general trends, however, is extremely difficult. A century ago, when the first attempts were being made to tackle this problem systematically, all that people had to go by were crises — sudden drops in prices, frequent bankruptcies and set-backs in external trade. When things went more smoothly, it seemed impossible to say anything at all about future trends or to evaluate small fluctuations. True, some experts studied bank rates, or kept records of the prices of the most important raw materials such as coal and iron, but little was known about imports and exports, about the labour market, and about unemployment, since no statistics on these subjects had been compiled. Nor was anything known about turnover, bank credits, stocks, and profits and losses, which were kept strict secrets, unless insolvency or legal proceedings brought some of the facts to the notice of the public.
Private enterprise was far more private and unrestricted than it is today, and this meant that every business shrouded its own activities in a thick veil of mystery. The less the competitors knew about one, the better it was for all concerned. Even the national economy was nothing but a legal, philosophic and poli- tical abstraction, for bureaucratic officials were as reluctant to reveal what they knew as were private entrepreneurs.
Oddly enough, the history of economic analysis followed a parallel course to the history of painting. Having tried to depict nature as realistically as possible and producing a host of quite unnaturally posed pictures as a result, economists like painters turned to impressionism in the 1880s. The economic impres- sionist par excellence was Alfred de Foville who, taking 32 apparently unrelated economic and social phenomena on which statistics had been compiled, viz. post office revenue, legacies, coal output, tobacco sales, inland revenue, box-office receipts, suicides, births, etc., compiled a "barometer" (see Fig. 3l) on which socially favourable trends were marked in white, bad trends in black, and intermediate trends in different shades of
bad
= fair
fairly good |:;:-".-:^=good
FRANCE
Ci'rculaKon postale
CrculaKon telegraphique
Successions
OonaKons
Enpeglstrement
Ventes d'immeubles
Tmbre des quittances
Cxjrs des rentes 3p.%
Clearing -House de Pan's
Emissions frantjaises
Revenu des actions
Transmissions de Htres
Commerce exferleyr ISpeeitti
Importations de matieres premieres
Exportations de produifs fabriques
ChetTiins de Fer : Tonnage
Chemins de Ter : Receltes brures
NavvgaKon internaKonale:1bnnage
tapeur: Rnte moWce
PnoducKon de ta houille
Cbnsommation de la howlle
fVoductvondu fen
Tabacs-.Ventes
Receltes des ttieafres parlsiens
Octroi de Paris
Mont dc Piete de Paris: Engagements
Contributions directes : Anticipations ||I[^Br^^
Cbnfribuiions diredesfFrais de poursuiles ||||B
Condamnaf ions, collocations, liquidations
Faillites
^icides
&cedent des naissances sw les deces
o*- ^incr*
3 1 . De Foville's General Barometer, one of the first attempts to present a pictorial analysis of economic phenomena.
272 CHAPTER 10
grey. Thus, a single glance would give anyone a fair idea of the state of the country's economy. De Foville himself believed that he had founded a new branch of science which he called "economic and social meteorology"/ his colourful chart being a General Barometer. In fact, it was anything but a barometer, for while a real barometer measures a specific phenomenon, viz. atmos- pheric pressure, De Foville's barometer measured phenomena of all sorts. Nevertheless, the term "barometer" was to become part of the vocabulary of many economists, as they made frantic attempts to predict the future from kaleidoscopic data.
Havard knows no crisis
At the beginning of the 20th century, the impressionist method was at last found wanting, both for diagnostic and also for prognostic purposes. People began to realise that economic trends had to be evaluated far more realistically than any colourful impressionist schemes allowed. Just as artists dis- carded subtle touches in favour of bold strokes of the brush, economists, too, looked for new means of expression. While the term "barometer" had stuck, the word assumed a new meaning.
This time, the light came out of America, and moreover out of America's Mecca of knowledge: Harvard University. Having made a very careful study of economic fluctuations over many years, the Harvard experts came to the conclusion that general economic trends can be expressed by means of three curves reflecting trends in the stock, commodity, and money markets respectively. Their main guides were, in fact, share prices, raw materials and rates of interest. It appeared from these curves that the stock market was always a jump ahead of the commodity market which, in turn, was a jump ahead of the money market. This had been known before, but the Harvard experts also determined the exact interval between movements in the three markets, and could therefore predict that, if shares rose or fell, raw material prices and rates of interest would follow at a certain date.
Immediately after the First World War, the Harvard 1 A. de Foville: Essai de mitiorologie iconomique et sociale (Paris 1888).
THE MYSTERY OF THE BUSINESS CYCLE 273
Committee for Economic Research began to publish its three curves and to make regular predictions based on them. These predictions caused a great sensation, not only because they were more accurate than any prior estimates of this kind, but chiefly because eminent men were now using the stock exchange for making general economic predictions, when all previous fore- casters had done the precise opposite. Moreover, the Harvard economists were respected as disinterested scientists — ^no one following their advice ran the danger of being taken for a ride to some swindler's direct advantage.
For a time all went well. From 1919 to 1925, the stock exchange curve fulfilled its promise admirably, but in 1925 the spell was broken. Shares kept rising but commodity prices refused to follow. Rates of interest, on the other hand, followed with a vengeance, not least because wild stock exchange speculation drove them up so high that they became usurious. Havard tried its best to repair the barometer, but the mercury seemed to have run out of it.
Thus the Harvard team, like all great experts, failed to predict the great depression of 1929. What was even worse, it betrayed its own theory, by proclaiming that the stock exchange slump was only a transitory phenomen which could have no lasting effects on the commodity and money markets, i.e. on the general economy. In short, there was no real crisis, and no cause for concern. When the public learned better, the Harvard barometer became one more victim of the depression, and with it ended one of the most interesting and fruitful attempts to make regular economic prognoses.
The German barometer
All the same. Harvard had set an example, and some of its disciples fared far better than their teacher. One of the most successful was the Institutfur Konjunkturforschung ( Institute for Business Cycle Research) founded in Berlin in 1925. Under the brilliant leadership of Ernst Wagemann, this Institute quickly gained international recognition. Its triple barometer system was a variation of the Harvard barometer. In Germany, the
274 CHAPTER 10
Stock exchange played a very much smaller role than it did in America and, for that reason alone, share prices had to step into the background. Again, on the commodity market, the German economists investigated turnover rather than price fluctuations, arguing quite logically that prices were largely reflections of supply and demand.
While the first or monetary barometer registered prices, wages, and rates of interest, ^ the second or productive baro- meter registered industrial output, employment, orders, im- ports of raw material and semi-finished goods, bank credits, the circulation of money, and bank reserves of gold and foreign exchange. The fact that money, credit, and production, were all treated under one heading was not so much a question of prin- ciple as of expediency. Since production and business figures took some time to prepare, bank credits were a fair indication of future production, and bank reserves reflected the possible volume of future imports.
The third or trading barometer registered everything con- cerned with internal and external trade: the turnover of indivi- dual sectors of the economy, the production of consumer goods, total imports and exports, and also wage incomes. The three barometers therefore overlapped to some extent, while some factors, e.g. building and estate transactions were completely ignored. Many other objections, too, could have been raised against the German barometer system, which nevertheless proved its practical, if limited, usefulness.
In fact, the Berlin Institute for Business Cycle Research rarely tried to make predictions for more than three months ahead, and, by and large, it made few mistakes. True, much of its success was due to the fact that a number of leading industrial concerns gave the Institute full access to their order books, and once the future orders of leading businesses are known it is not too difficult to predict if production will rise or fall, or if the level of employment will be maintained. The rest of the work was a mere compilation of graphs which expressed the known facts in impressive ways.
A frequent mistake of national business cycle research institutes is that they look upon their own country as the hub
^ Ernst Wagemann: Struktur und Rbythmus der fVeltwirtscbaft (Berlin 1931), p. 177 fF.
THE MYSTERY OF THE BUSINESS CYCLE 275
of the world and ignore the rest of mankind. Although the German institute was less guilty in this respect than its counter- parts in other countries, it, too, allowed itself to be carried away by the "economic miracle" and by ideas of eternal prosperity. Thus it failed to predict the great collapse of autumn 1929, or even to appreciate the seriousness and persistence of the subsequent depression. However, in 1932, it was one of the first to recognise that the worst was over, and that things were about to improve again.
The post-war crisis fails to materialise
During the war, Business Cycle Institutes were put out of independent business altogether, since government decisions put a stop to all market fluctuations. Economic research in- stitutes became government departments, and were supplied with large sums of money to carry out comprehensive investigations. Still, many economists must have longed for the good old times of the free economy, for only extreme dissatisfaction can explain the excessively gloomy prophecies they made at that time.
They had an unprecedented opportunity, and once again they failed bitterly; more bitterly, in fact, than just before the 1929 crisis, for this time no one expected them to predict the great turning point — the end of the war. All that people wanted to know from them was what things would be like once all the horror was over. Particularly in America, on which no bombs had fallen, an extra year of war was not nearly as great a worry as the question how the over-inflated war economy could be switched to peacetime conditions.
The vast majority of economic experts were convinced that this transition would involve a violent crisis, and even Colin Clark, an Australian economist and statistician, whose (1942) predictions about the economics of 1960 came closest to the truth, expressed the fear that the anticipated post-war boom would be followed by a fairly severe and prolonged depression, the last of its kind. Subsequent depressions would also occur but be short-lived, and from the late 1940s onward there would
276 CHAPTER 10
be a prolonged upward trend. ^ Americans were much more pessimistic, still. Post-war unemployment figures of 12-15 million — or, counting women engaged in war work, of up to 20 million — were commonly predicted, on the assumption that, the First World War having led to a short but extraordinarily heavy readjustment crisis, there was no reason for believing that things would be different after the Second World War.
Then there occurred the real American economic miracle — the demobilisation of ten million people was effected without any upheavals, the transition to a peacetime economy pro- ceeded extremely smoothly and unemployment figures rose to less than four million — ^half as much as before the war. Un- daunted, the prophets of gloom now proclaimed that the worst had merely been postponed. Demobilisation had simply co- incided with a favourable phase of the Kitchin-cycle, but the crisis was bound to come sooner or later. Like mediaeval astrologers who proclaimed volcanoes and earthquakes whenever un- favourable planets threatened to combine under one sign of the Zodiac, so modem economists threatened mankind with terrible disasters on the occasion of the impending coincidence of the three most important cycles: the KondratiefF, the Juglar, and the Kitchin, All three would reach their nadir in 1948, when a world-shaking depression would begin, to last until 1952.
This prophecy, too, was wholly false — as mankind was to learn with great relief. True, America had a marked recession in 1949, but nothing resembling a real crisis, while Europe con- tinued to make uninterrupted economic progress. Cycle theory has not yet recovered from this last body blow, and its exponents have rightly been reproached with confusing the past with the future.
Models to order
The failure of the generally expected post-war crisis to materialise and the prolonged prosperity which, apart from relatively mild recessions (1946, 1949, 1954), has been so characteristic of recent trends, have revived America's faith in 1 Colin Clark: The Economics of 1960 (London 1942), p. 108.
THE MYSTERY OF THE BUSINESS CYCLE 277
eternal bliss on earth. The spectre of 1929 is slowly but steadily receding.
Is this faith really justified.'' Optimists adduce three weighty arguments in its favour: firstly the American economy has become far less speculative, as post-war stock exchange transac- tions show. Moreover speculation is far more strictly controlled today than it was in 1929, when reckless gamblers could use unlimited credit to drive quotations to giddy heights. Secondly, Keynesian principles can be applied: if there is a downward trend the economy can be helped back on its legs by an injection of a measured dose of inflation, while unhealthy expansion can be counteracted by such deflationary measures as a credit squeeze. These prophylactic and therapeutic financial measures have helped to avoid excesses, or to nip them in the bud, and there is no reason to believe that they will be less successful in the future.
The third and most convincing argument (which economists usually keep under their hats) is rearmament. The United States spends twenty times as much money on its armed forces and military aid abroad as she did before the war. Roughly twelve per cent, and in some years as much as fifteen per cent of the national income is directly or indirectly devoted to military expenditure, giving full employment to 10-12 million people at the state's expense — to as many people, in fact, as were unemployed before the war. While the miserable pre-war unemployment benefits hardly kept the wolf from the door, military expenditure now helps to maintain a general state of boom. A rich country can easily afford to spend an eighth of its income on armaments, and then turn the expenditure into good business, as well. If the enormous military budget — 45,000 million dollars in 1957 — were suddenly to be drastically reduced, there is no doubt that a very serious crisis would ensue. In practice, this is unlikely to happen and so, the economists argue, the future looks rosy, indeed. At worst, there may be minor recessions.
European economists, too, generally subscribe to this new optimism, and though many of them have not abjured cycle theories, they no longer worry about determining what phase in a given cycle the present stage represents. They continue to rely on statistics, and general indexes, but they no longer look
278 CHAPTER 10
upon "barometers" as omens. After all, economic developments are men-made, and men can fashion them to their will. This is precisely the task of modem cycle research. True, economic institutes cannot dictate economic developments — certainly not in a free economy — but they can advise those in whose hands the decisions lie. Economic institutes are therefore increasingly turning therapist and refurbishing their theoretical basis. Now that cycle theories have been discarded, new theories must take their place, and these are only in the process of being constructed. While certain pathological symptoms of the economy are well- known, others are not, and economists still find it difficult to say what curative methods must be applied to individual cases, and above all what dosage must be administered.
The better to deal with all contingencies, economists have recently started constructing "econometric" models of specific economic situations. On the basis of the simple statistical (stochistic) models, actual developments can be compared with theoretical expectations, while the complex models show, for instance, what repercussions on the rest of the economy can be expected from wage increases.
Models of this kind may be called conditional predictions, since, while certain basic assumptions are made, variable factors must also be taken into account, and of these there is an almost unlimited number, particularly when, like most econometrists, we consider political factors as well. Still econometrists appear to be undaunted, and the Dutch econometrist Jan Tinbergen, investigating the American trade cycle, thought nothing of constructing his model with 30-40 variable factors.^
True, the results of these gigantic labours are somewhat frightening to the layman, who may easily throw his hands up in despair when he reads the following formula: (1 4- t') ^0 = - t' (104 -c)-8\(a-l-A- Aa)
— B\ {A — uA) + 8'A - a -\- c + Aa — 004 which simply expresses the expected effects of a modest tax reform on purchasing power. But those who are used to mathe- matical symbols will not be discouraged. The only thing that matters, in econometry as in anything else, is not the scientific facade or the splendour of the tools, but the actual results. So
* Jan Tinbergen: EinfUhrungin die Okonometrie (Zurich 1952), p, 252,
THE MYSTERY OF THE BUSINESS CYCLE 279
far, econometry has been unable to tell us much about the economic picture of tomorrow — ^nor is that its purpose. What econometry can do and has done is to predict with great accuracy what effects will result from what economic changes — and that is great progress, indeed.