Business Cycles

I. GeneralArthur F. Burns

BIBLIOGRAPHY

II. Mathematical ModelsTrygve Haavelmo

BIBLIOGRAPHY

I GENERAL

Economic change is a law of life. Nowadays, we commonly associate economic instability with business booms and recessions, and we have become accustomed to speaking of these vicissitudes in economic fortune as the “business cycle.” However, economic instability has been man’s lot through the ages, whether he has made his living by hunting and fishing, by cultivating crops, or by practicing the arts of commerce, industry, and finance. Economic history discloses endless variations of economic conditions. Even the meaning of “good times” keeps changing as the aspirations of people and their performance undergo changes. But relative to the standards of each age and place, some years are prosperous, others dull, still others depressed.

Types of economic movement

The outstanding feature of modern industrial nations is the growth of their economies. Thus, the population of the United States has risen steadily, year in and year out. So too, with very few exceptions, has the stock of housing, industrial plant, machinery, school buildings, highways, and other major forms of capital. The gross national product—that is, the total output of commodities and services—has fluctuated continually, but has done so along a rising secular trend. So also has output per capita, per worker, or per man-hour worked. In short, the American economy, viewed in the aggregate, has been basically characterized by growth of resources, growth of output, and growth of efficiency.

When we look beneath the surface of aggregate economic activity, we find some industries and communities growing rapidly, others growing only gradually, and still others declining. These divergent trends reflect a host of influences—among them, business innovations, population changes, shifts in consumer preferences, the discovery of new mines or oil fields, the exhaustion of old mines or timberlands, and changes in governmental policies. For example, the capital invested in American railroads and their volume of traffic increased rapidly during the nineteenth century, responding to the economic growth of the country and in turn stimulating it. But the railroads also grew at the expense of coaches, canals, and other waterways, which they gradually superseded by offering better service or charging a lower price. Years later the competitive trend was reversed, as new methods of transportation came into being—first, trolley lines, then buses, trucks, passenger automobiles, pipelines, airplanes, and improved waterways. These battled the railroads for traffic as vigorously as railroads in their youth had fought their commercial rivals. More recently, railroads have begun to retaliate through the use of radically new freight cars and other innovations. Such divergence of industrial trends is one of the expressions of economic progress.

Business cycles have been intimately connected with the lopsided surges of development that mark economic progress. However, business cycles are not the only type of fluctuation to which economic life is subject. During certain hours of the day, most of us are at school or at work; during other hours we relax in whatever way suits our tastes or needs. This daily cycle in activity is so regular and dependable that we take it for granted. The same is true of the weekly cycle which brings its day or days of rest. Whatever difficulties or opportunities the daily and weekly cycles may have posed for our remote ancestors, our own lives and social institutions have become adjusted to their repetitive course. We know that shops will be closed at certain hours and on certain days, and we plan our shopping accordingly. We know that the nation’s production will decline abruptly when factory workers put down their tools in the late afternoon, but we also know that their jobs do not cease on that account and that they will take up their tools again the next morning or when the weekend is over. In view of the extreme brevity and regularity of these cyclical movements, we pay no attention to them in judging whether business is improving or worsening.

Much the same is true of the seasonal fluctuations that run their course within the period of a year. Partly because of vagaries of the weather or the calendar, partly because of changes in business practice, the annual cycle is less regular than the daily or weekly cycle. Nevertheless, we expect business in general to be more brisk in the spring than in the summer or winter, and we ordinarily find it so. We expect department store sales to reach a peak during the Easter shopping season and a still higher peak before Christmas, and so we find it. We expect unemployment to be at its highest in February and at its lowest in October, and so it usually is. Workers in seasonal trades may not cherish the fluctuation to which they are subject, but they can reasonably count on returning to their jobs when the dull season ends and can plan their lives accordingly. In view of the substantial regularity of seasonal fluctuations, businessmen as well as economists usually put them out of sight when they seek to determine whether a particular branch of trade—or the economy as a whole—is expanding or contracting.

Business cycles differ in vital respects from these daily, weekly, and annual cycles. First, the recurring sequence of changes that constitutes a business cycle—expansion, downturn, contraction, and upturn—is not periodic. In other words, the phases of business cycles repeat themselves, but their duration varies considerably and so too does their intensity and scope. Second, since business cycles last from about two to ten years, they are considerably longer than the other cycles. Third, business cycles have a more powerful tendency to synchronize industrial, commercial, and financial processes than do the shorter cycles. Thus, the daily and weekly cycles in total production have no counterpart in inventories, bank loans, or interest rates, while seasonal fluctuations vary widely from one business activity to another. Fourth, although custom has left its imprint on the daily and annual cycles, they are part of the natural environment of man. Business cycles, on the other hand, are a product of culture. They are found only in modern nations where economic activities are organized mainly through business enterprises and where individuals enjoy considerable freedom in producing, pricing, trading, and saving or investing.

When economic plans and decisions are made independently by millions of business firms and households, some imbalance is frequently bound to occur between output and sales, or between output and the stock of equipment, or between inventories and outstanding orders, or between costs of production and prices. This much can be reasonably anticipated by everyone. However, the locus of the imbalance, its timing and magnitude, and the adjustments to which it leads can rarely, if ever, be foreseen with precision. In short, the business cycle lacks the brevity, the simplicity, the regularity, the dependability, or the predictability of its cousins. For all these reasons, although the business cycle is often the vehicle of progress, it also spells instability for society. When the economy starts on a downward course, no one can be sure how many months the recession will last, whether it will degenerate into a depression, how many business firms will go bankrupt, how far prices will decline, and—most important of all from a human standpoint—how many men and women will become unemployed. Although the United States and other countries are learning rapidly how to adapt to business cycles and to bring them under control, they remain troublesome.

Business cycles are not merely fluctuations in aggregate economic activity. They are also fluctuations that are widely diffused throughout the economy, and this fact distinguishes them from the convulsions of economic fortune that characterized earlier times as well as from the other short-term variations of our own age. Continuous and fairly pervasive fluctuations do not arise in a nation’s economy until its activities of production, distribution, and consumption have become closely interwoven through division of labor, the making and spending of money incomes, a system of banking and credit, a mode of production relying extensively on fixed capital, and some ease in communication and transportation. Since these institutions emerged gradually in the Western world, the phenomenon of business cycles itself developed gradually and no precise date can be assigned for its first mature expression. It appears, however, that business cycles have existed in the United States, Great Britain, and France for nearly two hundred years, and that they have marked the economies of other modern nations practicing free enterprise since the latter part of the nineteenth century— if not longer. Earlier centuries, while free from business cycles, did not escape the ordeal of economic instability. This is evident from the hardships that frequently accompanied or followed bad harvests, epidemics, wars, earthquakes, monetary upheavals, high-handed acts of rulers, civil disorders, and similar fortuitous events.

In recent decades, the Soviet Union and other nations that organize economic activity through state enterprises and governmental edicts have also escaped business cycles; but they have not escaped economic fluctuations. Variations in harvests, political purges, wars, monetary revolutions, and misadventures, as well as successes of planning, have left their mark on the aggregate economic activity of these nations. Of course, episodic or erratic disturbances also powerfully influence the course of economic activity in the United States and in other developed nations that practice free enterprise, but they appear to do so by hastening or retarding, by strengthening or opposing, the economic processes that of themselves tend to generate cyclical movements. The ragged contours of most business cycles testify to the role of random disturbances, and so too does the strong individuality of successive business cycles.

Business cycles also need to be distinguished from specific cycles—that is, cycles in specific activities, such as mining coal or trading in securities, which have about the same order of duration as the business cycle but may or may not match its timing. Occasionally, specific cycles appear to be superimposed, so to speak, on longer cycles marked by their own rises and declines. Huge swings, lasting about 10 to 25 years, have been common im building construction in various countries. Waves of this order of duration, but consisting of accelerations and retardations of growth rather than of actual rises and declines, also appear to have characterized aggregate economic activity in the United States. These Kuznets cycles, as they are often called, reflect variations in the intensity of successive business cycles. A distinction between major and minor cycles, such as Hansen makes, likewise involves a grouping of successive business cycles. On this view, the interval between the troughs of severe depressions is a major cycle, so that some major cycles may include only one business cycle while others include two or more. Long waves of about fifty years—usually called Kondratieff cycles—have also been alleged to characterize aggregate economic activity of Western nations. The existence of these waves, while suggested by price movements, has not yet been established.

The terms used by economists to describe the phases of business cycles are rich in diversity but are gradually becoming standardized. The “peak” of a business cycle marks the end of “expansion” and the beginning of “contraction.” The “trough” marks the end of contraction and the beginning of expansion. Frequently, “prosperity” is used interchangeably with “expansion,” although it is better practice to restrict terms such as “prosperity” or “boom” to the higher reaches of particular expansions when full employment is closely approximated. The term “recession” does double duty. It is widely used to refer to the transition from expansion to contraction, just as “recovery” or “revival” is used to refer to the transition from contraction to expansion. Contractions of varying intensity are also commonly distinguished by the terms “recession” and “depression”; the former refers to a moderate contraction of aggregate activity that lasts in the neighborhood of a year, while the latter refers to a severe contraction or to one which, while moderate, lasts distinctly longer than a year. The term “crisis” originally was used to de-note the financial disturbances that frequently occurred during the transition from expansion to contraction, but later it came to be applied to any transition from expansion to contraction. Nowadays, the term “crisis” is usually reserved for a violent disruption of financial markets without regard to the stage of the business cycle in which such a disturbance occurs.

Growth of knowledge about business cycles

In view of the complexity of business cycles and the innumerable differences between them, their essential features and causes have long been a matter of debate. The lack of full or precise economic statistics, which was especially serious before World War I, inevitably contributed to uncertainty about the actual course of business cycles and their causes. But as public concern about crises, inflation, depressions, and unemployment grew, economists have also pressed their investigation of this range of problems.

During much of the nineteenth century, interest was focused on commercial crises—that is, the sharp rise of money rates, scramble for liquidity, drop of prices, and spread of bankruptcies that frequently marked the culmination of a boom. With the emergence of the concept of a business cycle, various economists became concerned with the entire round of events that preceded and followed a crisis. The business cycle itself, however, was still viewed as centering, in the main, in activities of commerce and finance. Some economists traced its causes to natural forces, others to psychological factors, and still others to the workings of the monetary and banking system. Toward the end of the century, interest began to shift to phenomena of industry and employment, and more particularly to the great fluctuations that characterized the capital goods industries. This change of outlook reflected the growth of manufacturing, transportation, and public utility enterprises in modern nations, the relative decline of agriculture, and a growing realization that the transition from prosperity to recession could occur without a crisis or panic but not without a substantial increase of unemployment. In later decades, numerous explanations of the business cycle were developed that gave a large role to investment—usually to investment in fixed capital but sometimes to investment in inventories. Economists stressed different factors that had a bearing on the investment process—such as population growth, territorial expansion, stock of capital, the state of optimism, new technology, bunching of innovations, the rate of change in consumption, variation of interest rates, and changes of costs, prices, and profits. Or else they attributed primary significance to particular features of economic organization—such as industrial competition, uncertainty of demand, or the inequality of incomes. More frequently than not, the various theories differed mainly in their points of emphasis and therefore served to supplement one another.

The truly outstanding contributions to knowledge of business cycles were made by a small number of economists. Clément Juglar, 1819–1905, pioneered by demonstrating, in the course of a massive factual study of prices and finance, that crises were merely a passing phase of a recurring, wavelike fluctuation in business activity. Mikhail Tugan-Baranovskii, 1865–1919, was the first influential economist to see in the fluctuating rate of growth of the fixed capital of a country the main cause of its business cycles. Knut Wicksell, 1851–1926, clarified the cumulative processes of the business cycle by analyzing the consequences of a discrepancy between the rate of return on investment, which was liable to shift because of technological or other real changes in opportunity, and the market rate of interest. Albert Aftalion, 1874–1956, developed the implications for the business cycle of certain industrial facts—the long period required to bring new fixed capital into being, the long life of capital goods, and the capacity of minor changes in consumption to generate large changes in the net additions to the fixed capital required by business firms. Joseph A. Schumpeter, 1883–1950, viewed economic growth itself as a cyclical process and attributed the business cycle to the bunching of innovations, which forced difficult readjustments on old enterprises but in the end resulted in a more effective use of existing resources. Wesley Clair Mitchell, 1874–1948, carried factual investigation of business cycles far beyond earlier efforts, sharpened the concept of a self-generating cycle in a business system, and clarified the interrelations of costs, prices, and profits during a business cycle. John Maynard Keynes, 1883–1946, stressed the dynamic role of investment in altering the level of national income, formulated a consumption function which treats consumer spending as a passive response to national income, and with the aid of this function clarified the process whereby an increment of investment, besides adding directly to a nation’s income, raises it indirectly by stimulating larger consumer spending. Through the contributions of these pioneers and of many other economists and economic statisticians, notably Warren M. Persons, Simon Kuznets, and Jan Tinbergen, significant advances have been made in recent decades in describing with some precision the major features of business cycles and also in understanding the processes whereby they are generated.

This paper presents in nontechnical language the main results of modern research on the nature and causes of business cycles. It should be borne in mind, however, that the concrete manifestations of the business cycle keep changing and that numerous aspects of business cycles are still obscure. These facts justify extensive new research. The investigations that economists have currently under way focus on speculative model building, econometric model building, historical studies of individual cycles, statistical studies of fluctuations in individual processes or in the economy at large, experiments with forecasting techniques, and studies of business-cycle policy. This variety of approaches sometimes leads to methodological controversies. But no serious student of business cycles any longer questions that empirical research must be guided by an analytic framework or that speculative theorizing must be tested by an appeal to experience.

Cyclical behavior of aggregate activity

The business cycle involves to some degree the entire system of business—the formation of firms and their disappearance, prices as well as output, the employment of labor and other resources, costs and profits, the flow of incomes to individuals and consumer spending, savings and investments, exports and imports, trading in securities as well as commodities, the extension and repayment of loans, the money supply and its turnover, and the fiscal operations of government. Since there is no unique way of combining all these activities, the business cycle cannot be fully depicted by any single measure. However, the behavior of the entire congeries of fluctuations is indicated reasonably well for recent decades by statistical series of fairly comprehensive economic coverage—such as industrial production, total or nonagricultural employment, the flow of personal income, bank clearings or debits, and the gross national product.

The picture of a typical business cycle which emerges from these statistical records and also from earlier historical descriptions is that of a sustained rise in aggregate economic activity followed by a sustained, but smaller and shorter, decline. Activity at the peak of a business cycle is not merely higher than at the immediately preceding and following troughs. With very rare exceptions, it is also higher than at the preceding peak and lower than at the following peak. Likewise, the trough of a business cycle is usually higher than its immediate predecessor. In view of these typical characteristics, a business cycle almost always includes a visible element of growth. It is not merely an oscillation. The expansion, which ultimately carries aggregate activity to new heights, is typically most rapid in its early stages—the more so when it follows a severe contraction than when it follows a mild one. Although the rate of advance usually tapers off as the expansion proceeds, at times it reaccelerates as an expansion draws to a close without, however, regaining its initial speed. During contractions the rate of decline is usually fastest in the middle stages.

Between 1854 and 1961 the average length of business cycles in the United States was 49 months, with the average expansion lasting 30 months and the average contraction 19. The duration of individual cycles varied considerably—from 10 to 80 months for expansions, from 7 to 65 months for contractions, and from 17 to 101 months for full cycles. In a sense, aggregate activity was “depressed” over longer intervals than the duration of contractions may suggest, since some time must elapse before recovery can restore activity to the level attained at the preceding cyclical peak. On the other hand, the level of activity in the months immediately following a peak is often only a little lower than at the peak. During the ten business cycles from 1919 to 1961, when expansions averaged 35 months and contractions 15, the increases of industrial production ranged from 18 to 93 per cent and averaged 38 per cent, while the declines ranged from 7 to 66 per cent and averaged 26 per cent. Total output and employment, however, have fluctuated within decidedly narrower ranges. The reason is that they encompass, besides volatile activities like manufacturing and mining, relatively stable activities such as retailing, the service trades, and governmental work. Thus, during the business-cycle contraction of 1957–1958, when industrial production declined 14.2 per cent, total real output fell only 4.6 per cent and employment in nonagricultural establishments 4.3 per cent (see Table 1 and Figure 1).

In other industrial countries the average duration of business cycles has been somewhat longer than in the United States. For example, between 1879 and 1932, 15 business cycles ran their course in the United States, but only 10 in Germany, and 11 in Great Britain and France. Typically, the amplitude of business cycles has also been smaller in other countries than in the United States. Although the business cycles of individual countries often synchronize, some divergence of economic fortune has always been present. In general, the minor cycles of individual nations have followed a relatively independent course, while the larger cyclical movements have tended to be of international scope.

Cyclical behavior of individual activities

Many, but by no means all, individual activities reflect the cyclical movements of comprehensive aggregates of economic activity. The fortunes of individual firms are often dominated by personal factors or conditions peculiar to their industry or locality. Activities like the production of wheat experience fluctuations that are heavily influenced by the weather and bear little relation in time to business cycles. Activities involving the production of new products, like radio tubes in the 1920s or transistors more recently, may defy business-cycle contractions during the early and rapidly growing stage of their history. Some financial magnitudes, like the money supply, decline during severe contractions but merely experience a reduced rate of growth during ordinary contractions. Others, like commercial bank investments or the cash balances of corporations, tend to move contracyclically. Even activities that generally move with the business cycle sometimes skip a cycle, or undergo an extra fluctuation of their own, or move especially early or late during recessions or recoveries. In short, some economic activities are free from cyclical fluctuations over extended periods or are subject to an independent rhythm, while even the numerous activities that tend to keep in step with the business cycle have specific cycles whose turning points are scattered.

This diversity of movement in various branches of the economy means that expansion in some activities is always accompanied by contraction in others. We find, for example, that expansions in individual branches of production run side by side with individual contractions, whether business as a whole is depressed or prosperous. The turns of the specific cycles are not, however, distributed at random through time. On the contrary, they come in clusters, so that at a time when the troughs in production are bunched the peaks are few, and vice versa. But when the number of troughs in a given month exceeds the peaks, the number of expanding activities must also be larger the following month. Hence, the bunching of cyclical turns results, so to speak, in protracted periods when a majority of individual branches of production experience expansion, followed by protracted periods when a majority experience contraction. Empirically, the periods when expansions preponderate are virtually coterminous with the upward phases of the cycle in aggregate production; that is to say, aggregate production expands when individual expansions dominate. Moreover, when the expanding activities constitute a large majority, the amplitude of the cyclical rise in total production is apt to be larger than when the majority is small. To put this relationship another way, when the cyclical rise of total production is especially large, the industrial scope of expansion also tends to be especially broad. The scope of individual contractions, while usually less extensive than that of expansions, is similarly correlated with cyclical declines in aggregate production. All these relations in the sphere of production hold, and in the same way, between individual branches of employment and total employment, between individual branches of expenditure and total expenditure, and, indeed, between individual business processes and business as a whole.

The shift from a widening to a narrowing scope of expansions usually takes place gradually and follows a cyclical course. Rising activities are only a bare majority at the beginning of a business-cycle expansion. Their number swells as aggregate activity increases, though expansion reaches its widest scope not when aggregate activity is at a peak but perhaps six months or a year earlier. In the neighborhood of the peak, crosscurrents are the outstanding feature of the economic situation. Once the economy turns down, the number of expanding activities becomes smaller and smaller, but the scope of contraction does not widen indefinitely. Perhaps six months or a year before aggregate activity reaches a trough, the proportion of contracting activities is already at a maximum. Thereafter, the majority of contracting activities dwindles, while the minority of expanding activities keeps growing and before long becomes the ruling majority. About the time when that happens, the tide of aggregate activity begins rising again. A continual transformation of the economic system thus occurs beneath the surface phenomena of aggregate expansion and contraction.

The degree of clustering and the precise sequence of the cyclical turns of individual branches of production or employment vary from one business-cycle turn to the next. New and rapidly growing industries tend to move down late at downturns and to move up early at upturns. Activity in the machinery trades tends to move somewhat late at both upturns and downturns. Apart from these tendencies, the sequence within any cluster of cyclical upturns in individual branches of industry usually bears little resemblance to the sequence within the next cluster of either downturns or upturns.

Rather strong repetitive tendencies emerge, however, when production and employment are viewed in relation to other economic processes. Activities preparatory to investment expenditure—such as the formation of new firms, appropriations for capital expenditure by corporations, issuance of building permits, contracts for residential building, orders for machinery and equipment, contracts for commercial and industrial construction, additions to private debt, and new equity issues—typically begin declining while total production, employment, the flow of incomes, and the average level of wholesale prices are still rising. Similarly, these visible preparations for investment typically recover several months before production, employment, incomes, and wholesale prices end their cyclical decline. Cyclical fluctuations in profit margins, in the proportion of corporations achieving rising profits, and in prices of common stocks also tend to lead the tides of aggregate activity, and so too—although less consistently—do the fluctuations of total corporate profits. Other activities that tend to move up early in recoveries and to move down early in recessions are investment in inventories of materials, spot prices of industrial raw materials, and certain marginal adjustments of the work force, such as the average length of the work week and the rate of new hirings.

On the other hand, many economic processes or activities tend to lag in the course of business cycles. Outstanding among these are labor costs per unit of output, interest rates charged by banks on business loans, mortgage yields, retail prices, business expenditures on new plant and equipment, the installation of new industrial facilities, and aggregate business inventories. Of course, the cyclical turns in these lagging processes tend to precede opposite turns in aggregate activity.

The internal composition of the economy keeps changing in the course of a business cycle but not only on account of differences in cyclical timing. Just as individual activities do not rise or fall in perfect unison, so also they do not rise or fall by any uniform percentage during a business cycle. Some economic magnitudes—for example, retail sales and bank interest rates on business loans—move within a range that is narrow relative to their level. Others—especially business profits, capital gains or losses, and orders for investment goods—have enormous fluctuations. These and many other differences of cyclical amplitude are a recurring feature of business cycles. The turmoil that goes on within aggregate economic activity during a business cycle is, therefore, in no small part systematic.

In a typical business cycle, aggregate production fluctuates over a wider range than do aggregate sales. Moreover, sales by manufacturers fluctuate more widely than sales by wholesalers, while the latter fluctuate more than sales by retailers. The production of durable goods—both those destined for producers and those destined for consumers— fluctuates more widely than that of nondurables. Industrial production usually fluctuates more than the level of industrial prices at wholesale, which in turn fluctuates more than the level of retail prices or of wage rates. The cyclical fluctuation in the number of man-hours worked is larger than the fluctuation in the number employed, and the latter is larger in commodity-producing industries than in the service trades. Wage disbursements fluctuate within a wider range than salary payments or the flow of property income to individuals but within a much narrower range than profits. Corporate profits also fluctuate much more widely than dividend payments or total personal income. Consumer expenditures fluctuate still less than personal income, while personal savings fluctuate more than personal income but less than corporate savings. Cyclical amplitudes are larger in private investment expenditure as a whole than in consumer expenditure and they are also larger in consumer spending on durable goods than on nondurables or services. Again, amplitudes are typically larger in construction contracts than in the volume of construction executed, larger in business orders for machinery and equipment than in their production or shipments, larger in additions to inventories by business firms than in gross or net additions to their fixed capital, and larger in additions to inventories of the firms manufacturing durable goods than of those manufacturing nondurables. Finally, new security issues fluctuate more widely than trading on the stock exchanges, stock prices more than commodity or bond prices, short-term interest rates more than bond yields, open market interest rates more than customer rates, extensions of consumer installment credit more than repayments, imports more than exports, governmental revenues more than expenditures, and so through the gamut of processes that make up the economy.

Since disparities of cyclical amplitude and timing, such as those just noted, tend to be repeated in successive business cycles, the proportions that critical economic factors bear to one another tend to change in a systematic manner during a business cycle. For example, investment expenditure fluctuates much more widely relative to its size than does consumer spending; hence the ratio of investment to the gross national product tends to move with the business cycle, while the ratio of consumer spending to gross national product traces out an inverse movement. The amplitude of cycles is larger in total production than in sales; hence, inventory investment passes through a cycle of accumulation and liquidation that closely matches or even leads the cycle in aggregate activity, while the movement of total inventories lags both in recoveries and recessions. Government expenditures usually fluctuate within a smaller range or bear a much looser relation to the business cycle than do revenues; hence, the budgetary surplus, taken in an algebraic sense, tends to move with the business cycle. One more illustration will have to suffice. The rate of increase of the labor force varies little between expansions and contractions of aggregate activity; employment, on the other hand, moves strongly and synchronously with the tides in activity but typically rises more slowly than the labor force both at the beginning and toward the very end of expansion. Unemployment, therefore, typically turns up before aggregate activity starts receding and turns down only after economic recovery is already under way.

The empirical features of business cycles will be further elucidated in later pages. The point to note now is that our generalizations are largely based on intensive studies of the business cycles that have occurred in the United States during recent decades, although considerable confirmation has also been provided by studies of other countries— notably, Great Britain, Canada, Italy, and Japan. It is also well to keep in mind, first, that the generalizations emphasize the repetitive features of the economic changes that take place during business cycles; second, that they merely express strong tendencies toward repetition—not invariant rules of behavior. Diversity and individuality are no less characteristic of business cycles than the family resemblance among them, and this fact inevitably complicates the task of understanding the nature and causes of business cycles. Fortunately, there is less uncertainty about the broad processes that typically generate business cycles than about the specific causes of this or that cyclical episode.

The cumulative process of expansion

The continual transformation of the economy during a business cycle, which we have just reviewed, indicates that once the forces of recovery have taken hold, they will cumulate in strength. In other words, the expansion will spread out over the economic system, gather momentum, and for a time become a self-reinforcing process.

The proximate impulse to expansion may come from an increase of spending by business firms, consumers, or the government, or it may originate outside the domestic economy. The source or sources of the expansive impulse will be considered later. For the moment, let us assume merely that the economy is jarred out of its depressed level by an appreciable rise in the volume of newly initiated construction. A chain of familiar consequences will then be set in motion. Contractors will hire additional labor, disburse larger sums in wages, place larger orders for materials, supplies, and equipment with dealers or manufacturers, and finance at least a part of their rising outlays from new bank loans. The employment of labor on construction sites will at first increase only a little but after a few weeks or months—as the sequence of technical operations permits—more rapidly. Sales by retail shops and service establishments that cater to consumers will follow suit; for most construction workers will soon spend all or part of their larger income, and some will even feel encouraged to buy on the installment plan. The impact of the additional spending by contractors and their workmen will be spotty and uneven, but the effects will gradually spread out. Although some dealers or manufacturers will be content to meet the enlarged demand by drawing down their inventories, others will want to maintain inventories at their current level, and still others will seek to expand them in order better to accommodate a rising volume of sales. Here and there, therefore, there will be a stimulus to production not only of services and of goods made to specification but also of staples that are normally carried in stock.

In response to larger construction spending, the rough balance between expanding and contracting enterprises that had previously ruled in the economy will thus be tipped, albeit irregularly, toward expansion. As firms revise their production schedules upward, they also will often increase their purchases from other firms, give fuller work to their present employees, perhaps recall some former employees or hire new ones, but in any event disburse larger sums in wages. Thus, each expanding center of production will stimulate activity elsewhere, including lending by the banks, in everwidening circles. The spread of expansion from these centers will serve to check or counteract spirals of contraction that meanwhile are being generated at other points. With the scope of the expansion gradually becoming wider, retailers will be more prone to place orders with their suppliers in quantities that exceed their current sales, wholesalers and manufacturers will behave similarly, working hours will lengthen here and there, the work force will grow in an increasing number of firms and in the aggregate, and so too will income disbursements and sales to consumers.

We have supposed thus far that the higher volume of newly initiated construction will merely be maintained. In fact, construction work will tend to grow and so too will the activity of those making all sorts of machinery and equipment. Business firms, viewed in the mass, will still be operating well below capacity; but some firms—and their number is now increasing—will be operating at or close to full capacity. Moreover, as production rises, the profits of these firms, and indeed of business generally, will tend to improve. For a time, service enterprises—shops, theaters, buses, airlines, etc.—can handle more customers without adding appreciably, if at all, to the aggregate hours worked by their employees. That is much less likely to happen in manufacturing and other commodity-producing establishments. However, since these enterprises also rely heavily on overhead types of labor, their labor requirements per unit of output will tend to fall as output expands, thus reinforcing increases of productivity stemming from improvements of organization or technology. Experience shows that the swiftest advances of output per man-hour typically occur in the early stages of a business-cycle expansion and that they then usually outweigh such increases as may occur in wage rates. The result is that unit labor costs of production tend to decline rather sharply, at least for a few months. Depreciation charges per unit of output will also be falling. Meanwhile, such increases as occur in other cost items are as yet apt to be quite moderate, and they can frequently be offset by advancing selling prices. Hence, an increasing number of firms will find that their profit margins are rising handsomely and, since their volume of business is also growing, that their total profits are rising still more. With business profits and consumer incomes improving on a wide front, with shortages of capacity looming more frequently, with delivery periods lengthening, and with interest rates, machinery and equipment prices and construction costs still relatively favorable, it is only natural that contracts and orders for investment goods should rise briskly. Investment expenditures will follow suit, though with an irregular lag and diminished amplitude.

Moreover, as the expansion spreads, it generates in more people a feeling of confidence about the economic future—a mood that may gradually change from optimism to exuberance. As people become more optimistic, they respond more strongly to such increases of sales, prices, or profits as keep occurring. In other words, a given increase of sales, prices, or profits evokes a larger business response. An advance of prices, whether in commodity markets, business salesrooms, or on the stock exchange, is now more apt to encourage expectations that prices will go still higher. Increases of sales, improvements of profits, and delays of deliveries are similarly projected. In this sort of environment, dishoarding and borrowing become easier to rationalize and buying rises briskly all around. Many firms, fearing that they may not get all of the supplies they will soon need, begin bunching their orders more heavily and some actually order more than they expect to get. Not a few investors who had previously postponed action on attractive projects because the time did not seem right, now decide to go ahead. The new spirit of enterprise fosters more new projects that are related loosely, if at all, to the specific shortages of facilities that keep arising. More business firms brush up their long-range plans for expansion or modernization. More promoters push projects to exploit new products or techniques. More new firms are organized to share in the growing markets. More legislatures authorize improvements worthy of an era of prosperity. More families decide to buy a new automobile, to refurnish their home, or to build or buy a new house. Thus, the widening scope of expansion and the improved outlook that goes with it foster both investment and consumption, with advances of the one reinforcing the other in a cumulative process.

Even an adverse development, such as a strike in a major industry or a deliberate effort to reduce inventories of some major product, may now be taken in stride. At an early stage of the expansion, any such reversal of fortune could have sufficed to terminate it. Now, in view of the high level of business and consumer optimism and the large backlog of outstanding commitments for capital goods, a brief inventory adjustment is merely apt to bring a pause to the growth of aggregate economic activity; once this adjustment is completed the economy can resume its advance in spirited fashion.

Gathering forces of recession

And yet, as history so plainly teaches, a general expansion of economic activity sometimes lasts only a year and rarely lasts more than three or four years. Why does not the process of expansion continue indefinitely? And if the expansion must end, why is it not followed by a high plateau of economic activity instead of a decline? A partial answer to these questions can sometimes be found in disturbances that originate outside the mainstream of the domestic economy—such as political developments that threaten radical changes in property rights, or a drastic cut of military expenditures at the end of a war, or a major crisis abroad. Developments of this nature are entirely capable of cutting short an expansion that otherwise would have continued. However, experience strongly suggests that even in the absence of serious external disturbances the course of aggregate activity will in time be reversed by restrictive forces that gradually but insistently come into play as a result of the expansion process itself.

First, as the expansion continues, the slack in the economy is taken up and reduced. Although improvements of technology and new installations keep adding to the capacity of the nation’s workshops, production generally rises still faster; hence, idle or excess capacity diminishes in a growing majority of the nation’s businesses. Although the nation’s labor force keeps growing, jobs increase faster; hence, unemployment declines. Although the reserves of the banking system may be expanding, bank loans and investments generate deposits at a faster rate; hence, the ratio of reserves to deposits keeps falling. Although producers of metals and other materials and supplies respond to the brisk demand by raising production schedules, they are frequently unable to move quickly enough; hence, deliveries stretch out or become less dependable. The pecuniary expression of the mounting shortages is a general rise of prices—of labor, credit, raw materials, intermediate products, and finished goods; but that is not all. The shortages are real and their physical expression is a narrower scope of the expansion itself. Rising sales by a particular firm or industry still release forces of physical expansion elsewhere, but their effects are blunted since more and more businessmen must now contend with bottlenecks. Once labor is in short supply in a community, an increase of employment by one firm must often result in some reduction of employment elsewhere in the same community. Once this or that material is in short supply, some firms must get along with less than they need or wait longer for deliveries. Once the banking system stops expanding credit or materially reduces its rate of expansion, any new loans to some firms will affect adversely the ability of other firms to get the credit they need. Instances of this sort multiply as the economy moves toward full employment. At some point, therefore, the scope of the expansion stops widening and begins to narrow. Although aggregate activity is still growing, it can no longer maintain its initial rapid pace.

Second, the advance of prosperity tends to raise unit costs of production and therefore threatens profit margins—unless selling prices rise sufficiently. Taking the business system as a whole, much the largest item in costs and one which businessmen watch with the greatest care is labor—more precisely, the cost of labor per unit of output. This cost depends, first, on the hourly wage of labor and, second, on output per man-hour. Both tend to rise as the expansion progresses, but at unequal rates. The price of labor moves sluggishly in the early part of the expansion, but advances of wages tend to become more frequent and larger as competition for labor increases and trade unions take advantage of improved market conditions. Increasing resort to overtime work at premium rates of pay accentuates the rise in the average price of labor, and so too does the faster upgrading of workers. On the other hand, output per man-hour, which improved sharply early in the expansion, tends to increase more gradually as the expansion lengthens, and it may also decline before the expansion is over. To be sure, improvements in organization and technology continue to be made at a thousand points at this as at every stage of the business cycle. However, their effectiveness in raising productivity is offset by developments that increasingly grow out of prosperity—such as a decline in the average quality of newly hired labor, fatigue of both workers and their managers, restlessness among workers and rapid turnover of labor, the need to put some obsolete plants or equipment back into use, the need to operate some highly efficient plants beyond their optimum capacity, and the need or wish to add liberally—once substantial increases of business have occurred—to indirect or overhead types of labor. Thus, as the expansion of aggregate activity continues, increases of productivity tend to diminish or even vanish, while the price of labor not only rises but tends to rise faster than productivity. The result is that unit labor costs of production tend to move up persistently.

Third, the increases of construction costs, equipment prices, and interest rates that are generated by the expansion process gradually become of more serious concern to the investing community. After all, a rise in long-term interest rates tends to reduce the value of existing capital goods at the very time that it raises the carrying charges on new investments. Higher costs of new capital goods likewise serve to raise fixed charges. For a time, optimistic expectations concerning the earnings stream from new investment projects overpower the restraining influence of higher costs of capital goods or of higher interest rates, but they will not do so indefinitely. A firm that expects to earn 20 per cent annually from a new project can overlook a modest rise of construction costs or interest rates, especially when it plans to finance the investment from retained earnings or depreciation reserves. Not all investors, however, are in such a fortunate position. Home builders, in particular, are sensitive to a rise of construction and financing costs, partly because their activities are largely financed by borrowing and partly because interest charges are a very considerable fraction of the total cost of operating a dwelling. Experience shows that contracts for residential construction typically turn down before commitments for any other major category of investment. Business orders for machinery and equipment, as well as contracts for new factories, commercial buildings, and public utility plants still keep rising for a time. These types of investment are more responsive to prospective demand than to conditions of supply; but as the expansion of economic activity becomes more intense, they too begin to feel the pressure of rising costs. In deciding to invest in a particular project, a business firm may have given little heed to recent increases in costs. That decision, however, must still be followed by another, namely, whether to get the project under way now or later. Investors know that they will have the new plant or equipment on their hands for a long time and that their annual carrying charges will depend on the cost of the new capital goods, if not also on the rate of interest. They have got along thus far without the desired investment, and they will have to manage in any event without it for some months or years. If, therefore, they expect costs to be appreciably lower a year or so from now, they may well bide their time. Such postponements in placing orders and contracts become more frequent even as business decisions to invest continue to accumulate.

The rise in construction, equipment, and financing costs during an expansion impinges so broadly on the investing class that it would eventually check the investment boom even if prosperity were diffused uniformly over the economic community. However, this is not the case, and the uneven spread of profits is still another major development that impedes the continuance of expansion. At every stage of the business cycle there are bound to be some firms whose profits are declining or whose losses are increasing. But these firms are not a steady fraction of the business population, and there are cogent reasons for expecting their numbers to increase as the expansion of aggregate activity stretches out. To protect profit margins, selling prices must rise sufficiently over the entire range of business enterprise to offset higher unit costs of production. Since business conditions are good, many firms can and do raise prices that much or more. But there are always some firms that find it hard to advance selling prices, and their number tends to grow at an advanced stage of the expansion. In some industries, sales have recently been pushed with such vigor that the markets for their products are approaching saturation at existing prices. In other industries, exaggerated notions concerning the volume of sales that could be made at a good profit have led to overstocking or overbuilding, so that prices come under pressure. Errors of this type occur at all times, but they are likely to be bunched when enthusiasm has infected a large and widening circle of businessmen. In still other cases, business custom, long-term contracts, or governmental regulation make it difficult or inexpedient to raise selling prices. Of course, firms that cannot advance selling prices will try all the harder to resist increases in costs, but such efforts meet with limited success at a time of extensive shortages. With the rise in unit costs of production continuing across the business front, more and more firms therefore find that their profit margins are becoming narrower, thus offsetting the influence on profits of rising sales or reinforcing the influence of declining sales—instances of which now become more numerous. We thus find in experience, as we should expect, that after a business expansion has run for some time, the proportion of firms enjoying rising profits begins to shrink, although profits of business in the aggregate still continue to advance.

These developments—the narrowing scope of expansion as full employment is approached, the rise of unit labor costs, the rise of financing costs, the rising cost of new capital goods, the spread of these cost increases across the economy, and the shrinkage in the proportion of business firms experiencing rising profits—tend gradually to undermine the expansion of investment. Prominent among the first to reduce investment commitments are the firms whose fortunes are waning. Their curtailments spread doubt among businessmen whose profits are still rising, many of whom have also become concerned about prospective profits or have come to feel that construction and financing costs will recede before long from the abnormal level to which they have been pushed by prosperity. These attitudes and responses are likely to be reflected in some weakening of stock exchange prices, which in turn will stir fresh doubts. With investment commitments declining, but actual expenditures still rising, backlogs of unfilled orders for capital goods and of uncompleted contracts for business construction must sooner or later turn down. Meanwhile, uncompleted contracts for residential construction have, in all probability, already been declining for some time. The decline in these several backlogs induces reductions in orders for raw materials and parts, and the reduced pressure on suppliers in turn serves to stabilize, if not lower, prices. Since many of the consumer trades can now also count on faster deliveries, the orders placed with their suppliers are likely to turn down as well. These changes reinforce efforts to adjust inventories that have already been induced at numerous points by the narrower scope of expansion and the reduced rate of growth of aggregate activity. For all these reasons, while inventories on hand still keep rising, investment in inventories begins declining. In view of the smaller backlogs, business expenditures on fixed capital will themselves gradually move to a lower level a little later. Public expenditures may still rise, but they are unlikely to do so on a sufficient scale to offset the declines of private investment. The growth of consumer spending, therefore, is retarded, if it does not actually stop. As these adjustments proceed, the balance between expanding and contracting economic activities tips steadily toward contraction. The need for overtime is much reduced, unemployment begins to rise, aggregate production soon turns down—in short, a business recession gets under way.

The process of contraction

The course of a typical recession is well known. A decline of production is accompanied by a reduction in the number of jobs, besides a reduced work week for many. The flow of incomes to individuals, therefore, tends to decline, and consumer spending—at least for expensive durable goods—follows suit. Retailers and wholesalers are now more apt to place orders for merchandise that are below the level of their respective sales. Many manufacturers, in their turn, also attempt to reduce their inventories. Taking the economy as a whole, the broad result of these efforts is that production declines more than sales, and that inventory investment not only declines but is soon succeeded by liquidation. Meanwhile, quoted prices of many commodities, especially of raw materials, tend to soften, and discounts or concessions from list prices become more numerous and larger. Wage rates, however, are generally maintained and actually rise here and there. Even when they decline somewhat, unit costs of production still tend to rise, perhaps sharply, because it takes time before overhead costs, including the employment of indirect types of labor, can be adjusted to the lower volume of business. Many firms that are already experiencing lower profit margins therefore find that they must put up with still lower margins, while others first begin to feel the profit squeeze. With sales more often than not also declining, an increasing majority of businesses now experience falling profits, bankruptcies become more frequent, business profits in the aggregate—which probably began shrinking before sales did—decline further, and stock exchange prices extend their fall as well. In view of these developments, many businessmen and consumers, even if they are not actually poorer, become more concerned about the future. New business commitments for investment in fixed capital therefore tend to become less numerous, and—unless forces of recovery soon come into play—investment expenditures of this type as well as outlays on consumer durables will extend their decline, which is as yet modest, and reinforce the contraction process.

As a decline in one sector reacts on another, the economy may begin spiraling downward on a scale that outruns the magnitudes that we ordinarily associate with recession. The likelihood that a depression will develop depends on numerous factors—among them, the scale of speculation during the preceding phase of prosperity, the extent to which credit was permitted to grow, whether or not the quality of credit suffered significant deterioration, whether any major markets became temporarily saturated, how much excess capacity had been created before the recession started, whether and in what degree the balance of international payments has become adverse, the organization of the financial system and its ability to withstand shocks, the shape of political developments, and the aptness and scale of monetary actions and other governmental efforts, if any, to stem the economic decline. If the onset of the contraction is marked by a financial crisis or if one develops somewhat later, there is a substantial probability that the decline of aggregate activity will prove severe and perhaps abnormally long as well. For when businessmen and their bankers begin to scramble for liquidity, both trade credit and bank credit will decline and so too will the money supply; commodity prices at wholesale and retail will slump and wage rates decline, while interest rates for a time rise sharply; confidence will become impaired and many investment projects will be abandoned instead of merely being postponed; business losses and bankruptcies will multiply; more workers will earn less or become totally unemployed; and, since spells of unemployment also lengthen, more and more families will deplete their savings and be forced to reduce their spending drastically. Even if the shift from expansion to contraction is made gradually, untoward disturbances originating outside the economy may still strike with great force and transform a mild contraction into a depression.

Forces of progress and recovery

Normally, however, a contraction in aggregate activity does not lead to depression. A contraction is not a mirror image of expansion, as it might well be if the business cycle were merely an oscillation. A contraction does not usually cumulate and feed on itself in the manner of an expansion. Normally, many progressive developments continue, and some even become stronger, during the contraction phase of the business cycle; in other words, the forces making for contraction are powerfully counteracted by forces of growth that limit the degree to which it can cumulate.

What are these forces of growth? First, businessmen and consumers in a modern nation are accustomed to seeking and to expecting economic improvement. This optimistic state of mind generally continues during a contraction, provided its dimensions remain moderate. Investment opportunities, connected with new technology or market strategy, always keep arising in the minds of imaginative and resourceful men. Not a few of these opportunities are acted on promptly in spite of the recession. Second, most people are extremely reluctant to give up the standard of living that they have managed to attain, and in any event they cannot quickly readjust family expenditures. Hence, consumer spending is well maintained in the face of declines of income that are judged to be temporary. Third, the pitch of both interfirm and interindustry competition becomes more intense during a recession. Unlike investment commitments, which are at their highest level before aggregate activity turns down, the bunching of installations of new plant and equipment is likely to be heaviest when the recession is well under way. The newer facilities typically serve new products or permit lower costs of production of old products. Many progressive enterprises are therefore able to extend their markets even when business as a whole is falling off. Firms that suffer from shifts of demand or from an outworn technology may have managed to limp along or even do reasonably well when activity was brisk. Now, finding that competitors are penetrating their markets on a scale that threatens survival, the hard-pressed firms are more likely to move with energy to modernize their plant, acquire new equipment, improve their products, try out new marketing strategies, and eliminate waste. Meanwhile, vigorous businesses whose plants are operating at or close to optimum capacity do not stand still. Not a few of them anticipate a large expansion of sales when the dull season is over, and therefore undertake additions or improvements to their plant and equipment. Fourth, a nation’s resources normally continue to grow even during a recession. Since the population is still growing, the stabilizing force of consumption is reinforced. Since the number of business firms is still increasing, the formation of new businesses contributes, although at a reduced rate, to the demand for capital goods. Since the stock of housing, consumer durables, and industrial facilities is still expanding, a large market is assured for repairs, improvements, and replacements, although there is undoubtedly some postponing of this type of expenditure. Fifth, public efforts to promote economic growth and the general welfare are customary in a well-governed nation. These efforts may not always be wise or geared closely to the business cycle, but neither are they confined to times of prosperity. On the contrary, they are more likely to come during recessions—especially in recent times when full employment has become an increasingly firm objective of the public policy of nations.

The progressive forces that operate during recessions serve as a brake on the cumulative process of contraction. True, aggregate activity falls below the level reached at the peak of prosperity. The decline, however, is usually of moderate proportions. Not only that, but sales decline much less in the aggregate than production and the level of sales soon becomes higher than that of production. For a while, the liquidation of inventories proceeds at an increasing rate, but this cannot continue. To handle the volume of business on hand, especially if sales stabilize or decline very gradually, manufacturers and distributors must soon slow down, if not halt, the decline of their inventories. Taking the economic system as a whole, once inventory disinvestment declines more rapidly than the decline of sales, production must begin rising. Of course, a recovery of production will be preceded by an increase of orders, and an early upturn of orders is precisely what occurs when dealers and manufacturers take steps to slow down appreciably the decline of their inventories.

While business firms keep bringing inventories into better alignment with their sales, other developments that grow out of the recession also favor an early recovery. Since the reserves of commercial banks tend to pile up again, reserve ratios improve. Hence, interest rates decline and credit becomes more readily available. The effects of easy credit are likely to be felt most promptly by smaller businesses and the homebuilding industry, but they tend to ramify as banks put their reserves to use. When the demand for loans is still deficient, banks seek out customers energetically. At the same time, they augment their investments in bonds, thereby strengthening the bond market and stimulating a renewed interest in preferred stocks and giltedged common stocks. Meanwhile, numerous readjustments in the nation’s workshops serve to lower unit costs of production. In view of the decline of aggregate demand, wage rates often stop rising and sometimes decline a little, over-time operations become less frequent, not a few of the less efficient enterprises go out of business, production is increasingly concentrated in the most modern plants and on the best equipment, many of the less efficient workers are let go, the ranks of the overhead types of labor are thinned here and there, and workers generally become more attentive to their duties. These changes reinforce the improvements of organization and technology which always occur in a progressive economy and which are often speeded up during a recession, in response to the keener competition that develops at such a time. Of course, the beneficial changes in the costs of production of individual businesses are frequently offset or nullified by declining selling prices. However, once the adjustments of inventories have made good headway, commodity prices tend to stabilize. Hence, more and more firms are apt to find that their profit margins begin improving. With the prospect of profits brightening, interest rates declining, and costs of capital goods lower, some of the numerous investment projects that had previously been postponed are now revived and they supplement the new crop of active projects. As these developments become stronger, the decline of investment commitments ceases, new firms are established in larger numbers, orders and contracts for investment goods turn up, inventory disinvestment continues to ebb, and a recovery of aggregate production and employment soon gets under way.

Thus, corrective forces released by the recession combine with the more persistent forces of growth to bring the contraction of aggregate activity to a halt. Typically, the process works fairly speedily and the contraction is over in about a year or a year and a half. However, as previously noted, a contraction sometimes develops into a spiraling depression. When that happens, declining investment in fixed capital supplants inventory disinvestment as the principal drag on the economy. Worse still, the stubborn human trait of optimism begins to give way, so that a mere readjustment of inventories may bring only an abortive recovery. Once many men begin to lose faith in themselves or in the institutions of their society, full recovery may need to wait on substantial innovations or an actual reduction in the stock of fixed capital, unless powerful external influences come into play— such as a reorganization of the monetary system, massive governmental expenditures, or a sudden increase of exports on account of foreign developments. Fortunately, no industrial country has suffered a spiraling depression since World War II, and the likelihood of such a development—as will be noted later—has been greatly reduced.

Differences among business cycles

The preceding sketch of the nature and causes of business cycles has stressed typical behavior. Yet no business cycle of actual experience corresponds precisely to our sketch, and some cycles bear only a faint resemblance to it. What history discloses is a succession of business cycles that differ considerably in length, in the intensity of their phases, in the industrial and financial developments that gain prominence during their course, and in their geographic scope. In American experience, for example, while expansions have normally run longer than contractions, there is no peacetime expansion on record before 1960 that lasted as long as the decline from 1873 to 1879. Industrial production has typically fluctuated over a wider range than industrial prices, but the opposite is true of several business cycles associated with wars. Interest rates have commonly risen during expansions of aggregate activity, but they continued to decline during almost the entire expansion from 1933 to 1937. Broad indexes of wholesale prices have generally declined during contractions of activity, but they failed to do so during the recession of 1890–1891 or 1957–1958. Contracts and orders for investment goods have typically moved up before total production or employment in the recovery process, but they did not do so at the upturns of 1914 or 1933. Declining stock prices have frequently signaled the approach of a recession, but the stock market crash of 1929 came after aggregate activity had already turned down. Some economic declines, such as those of 1887–1888 and 1926–1927, were merely pauses in the growth of the domestic economy. Others, such as the depression of 1920–1921, attained international scope, while the depression of the 1930s became a world-wide upheaval of catastrophic proportions.

In view of these and countless other variations among business cycles, the causes of any particular cycle are always in some or large degree peculiar to it. One prolific source of cyclical variation in the United States, as elsewhere, is found in the behavior of money, foreign trade, and the balance of payments. For example, good harvests in 1879, when crops abroad were poor, stimulated large exports of grain at favorable prices, thereby improving farmers’ incomes, enlarging the business of shippers, inducing an inflow of gold, and otherwise speeding economic recovery. In 1891 and 1892, fear that political agitation for free silver would result in abandonment of the gold standard led to domestic hoarding of gold, to massive gold shipments abroad, and finally to a financial crisis in the spring of 1893. The expansion of 1891–1893, therefore, developed nothing like the vigor suggested by our account of the cumulative process of expansion. The outbreak of war in Europe in 1914 soon caused a sharp upsurge in American exports, thereby checking a contraction in aggregate activity that otherwise might have dragged on. To cite one more illustration, the expansion of 1958–1960 proved incomplete, in large part because of the restrictive monetary and fiscal policies that were undertaken by the government to curb inflationary pressures and to prevent further deterioration in the balance of international payments.

Business-cycle movements often spread from one country to another and sometimes engulf almost the whole world economy. Foreign trade, commodity prices, stock prices, and interest rates play a vital role in this process of transmission, both directly and through their influence on business psychology. The economies of most commercial nations are far more closely tied to the course of foreign trade and investment than is the economy of the United States. In view of the large role of foreign trade in small countries like the Netherlands or Norway, conditions abroad can have a decisive influence on domestic prosperity. Even in a larger country like Great Britain, an improvement of exports has not infrequently been the immediate cause of economic recovery. However, as the economic activity of a nation expands, its imports also tend to rise, partly because of a larger need for foreign raw materials and partly because of larger purchases abroad of equipment and consumer products. Meanwhile, since domestic markets keep improving, some firms find it more profitable or more convenient to cultivate home trade than to push exports. If, in response to the upswing of activity, domestic costs and prices advance more rapidly than prices charged by foreign enterprises, exports will probably suffer and monetary reserves—whether of gold or foreign currencies—will tend to diminish. A restriction of credit often follows, because under a regime of stable exchange rates the state of a country’s balance of payments and the size of its monetary reserves and borrowing facilities may leave little room for an independent financial policy. This pattern of developments has become familiar to the nations of western Europe and to Japan.

Many nations of Latin America, Asia, and Africa derive their foreign exchange mainly from the export of one or at most a few raw materials, supplemented by investments made in these countries by foreigners or, perhaps, by gifts from abroad. But the prices of internationally traded raw materials tend to fluctuate widely, in part because of variations in the state of demand in the industrial countries. These price fluctuations often have a critical bearing on the ability of the raw-material producing nations to acquire from abroad the capital goods and supplies needed to develop their economies.

Not only are the economies of different nations tied together, but as various theories of long waves or major cycles have sought to suggest, no business-cycle movement can be understood solely in terms of what happened during that phase or the one just preceding it. Thus, the American contractions of 1923–1924 and 1926–1927 were merely minor interruptions of a great onrush of economic activity from 1921 to 1929. The period began with a rapid increase of production, was followed by a stretch of slower growth, and ended on a note of reacceleration. Financial activities followed a different and more hectic course. Emerging as an international creditor after the war, the United States played its new role with exuberance. Through 1924 the volume of foreign loans was substantial, yet the loans were on the whole of sound quality—as attested by later experience. The next few years witnessed a further expansion of foreign loans and a sharp deterioration of their quality. The speculative craze expressed itself also in other financial areas, notably in the real estate market and superlatively in the stock market. Consumer credit shared in the general upsurge and made possible a huge expansion in the output of durable consumer goods during the 1920s, not only absolutely but also relative to total output, thus adding a new hazard to economic stability. The financial situation was also made vulnerable by the great pyramiding of international credits that developed under the gold exchange standard. Governmental policies in the United States after 1929, which brought on tax increases and—worse still— tolerated the destruction of a third of the nation’s money supply, cannot escape a very large part of the responsibility for the Great Depression; but neither financial developments abroad nor the course of policy, private and public, in the decade prior to the depression can go blameless.

Progress toward economic stability

Besides such differences among business cycles as we have noted, which largely reflect episodic influences, there are other differences of a more persistent kind. Just as the business cycle itself emerged gradually in the course of economic evolution, so many of its features have undergone changes as the economy has continued to evolve.

The structure of a nation’s economy and its institutions inevitably leave their stamp on the character of its cyclical fluctuations. Thus, after the introduction of the Federal Reserve System, the fluctuations of short-term interest rates in the United States became narrower, while the lag of long-term interest rates during recoveries and recessions became shorter and of late has virtually vanished. With the growth of trade unions and increasing resort to long-term labor contracts, wage rates have become less responsive to cyclical contractions of activity. More important still, the precise relations among the movements of production, employment, and personal income have kept changing as the structure of the American economy and its institutions have evolved. During the early decades of the nineteenth century, when agriculture was the dominant occupation, occasional declines in the nation’s total volume of production, whether large or small, had little effect on the number of jobs and sometimes had slight influence even on the flow of money incomes. Later, as wage-jobs gained rapidly in importance, the movements of employment and personal income fell into step with production. In recent times, how-ever, numerous changes in the structure of the American economy have served powerfully to reduce the impact of a cyclical decline of production on the lives and fortunes of individuals.

Important among these changes is the vast expansion of government, the greatly increased role of the income tax in public revenues, the shift of income tax collection to a pay-as-you-go basis, the rapid growth of unemployment insurance and other programs of social security, the growing frequency and scale of private pensions, the spread of business corporations, and their increasing pursuit of stable dividend policies. As a result of these and related developments, the movement of personal income is no longer closely linked to the fluctuations of production. For example, in the course of the recession of 1957–1958, the physical volume of industrial production fell 14 per cent and of total production nearly 5 per cent. In the early decades of this century, aggregate personal income would have responded decisively to such a decline in production. This time government receipts and expenditures offset the drop in the flow of income from production, first, because much less was collected in taxes from corporations and individuals, second, because the amount of unemployment insurance and other social security payments rose. Corporations in turn reacted to the decline in profits by reducing their savings rather than the flow of dividends or pensions to individuals. In the end, the aggregate of personal incomes, whether before or after taxes, declined less than 1 per cent, and in the case of after-tax incomes even this decline was over before the recession ended.

Major structural changes have also occurred in the sphere of employment. Manufacturing, mining, construction, and freight transportation are the cyclically volatile industries; but their relative importance as providers of jobs has been gradually declining in recent decades, while that of the more stable service industries has been increasing. In addition, the proportion of people who work as managers, engineers, scientists, accountants, secretaries, salesmen, or in kindred “white-collar” occupations has been steadily rising. Much of this type of employment is of an overhead character and therefore less responsive to the business cycle than are the jobs of machine operators, craftsmen, truck drivers, laborers, and others in the “blue-collar” category. It appears, therefore, that changes in the structure of the labor force have of late been loosening the links which, over a considerable part of economic history, tied the short-run movements of total employment in the United States rather firmly to the movements of production. We can no longer suppose, moreover, when employment falls during a recession, that there will be a corresponding decline in the number of people receiving an income. On the contrary, as a result of the widening sweep of social security programs, the number of income recipients actually increased during each recession of the postwar period.

These developments have left an imprint on the behavior of consumer spending in recent business cycles. First, consumers have maintained their spending at a high level even after business activity had been declining for some months, so that the cumulative process of contraction has been curbed. Second, retail trade has tended to turn up before production or employment, instead of lagging during the recovery stage as it did in earlier times. Thus, consumer spending has emerged as one of the active factors in arresting recession and hastening recovery. Of course, if the fluctuations of production had been larger in the postwar period, the impact of recessions on the lives of working people would have been greater. On the other hand, the more stable behavior of personal income and consumption has itself been a major reason why recent contractions of activity have been brief and of only moderate intensity.

Many other factors have contributed to this result. The need to overhaul the financial system became clear during the 1930s and led to numerous reforms, among them the development of the long-term amortized mortgage, the regulation of stock exchanges, the insurance of mortgages, the creation of a secondary market for mortgages, the insurance of savings and loan accounts, and—most important of all—the insurance of bank deposits. These financial reforms have served to prevent crises or the propagation of fear. Even more basic has been the change in political attitudes that emerged during the 1930s and which the Congress later articulated in the Employment Act of 1946. It is now generally agreed that mass unemployment is intolerable under modern conditions and that the federal government has a continuing responsibility to promote a high and rising level of employment and production. In recent times, there-fore, the business cycle has no longer run a free course, and this fact has figured prominently in the plans of businessmen as well as consumers. The general expectation of the postwar period has been that the government would move with some vigor to check any recession that developed, and that its monetary, fiscal, and regulatory actions would contribute to that objective. By and large, this confidence has been justified by events. Not only has monetary policy in the main been shaped with a view to promoting stable prosperity, but fiscal policy—which previously had been handicapped by the convention of annually balanced budgets— has lately also been guided by the state of the economy. Business firms too have been paying closer attention to the business cycle. There is evidence, in particular, that inventories are being better managed and that this is helping to moderate the cyclical swings in production. On the other hand, governmental policies have often served to intensify inflationary expectations or pressures, and this has become a recurring problem.

The nations of western Europe have also experienced structural changes in the postwar period that, on balance, have worked in a stabilizing direction. White-collar occupations have gained in importance, and so too have systems of social security and of tax collection on a pay-as-you-go basis. Some countries, especially Sweden, achieved notable success with contracyclical policies well before the United States. Of late, all of western Europe has been striving energetically and ingeniously to promote economic expansion and full employment, and these efforts have been attended by great success. Even before World War II, the business cycle was a milder type of fluctuation in western Europe than in the United States, and the difference has persisted in the postwar period. Indeed, the main problem facing European nations in recent years has not been unemployment but rather the difficulties caused by inflation and balance-of-payments disequilibria. Japan has also been struggling with this problem.

It would, nevertheless, be premature to conclude that the older hazards of the business cycle belong to the past. True, the business cycle has become milder as a result of a favorable conjuncture of structural changes and of both better and wider understanding of the requirements of business-cycle policy. Certainly, there is increasing recognition of the desirability of preventing recessions, rather than merely acting to moderate them once they occur. However, the forces that tend to generate cyclical movements have not vanished in western Europe or Japan any more than in the United States. It is possible that in the future a “recession” will mean merely a reduced rate of growth of aggregate activity instead of an actual and sustained decline, but there is as yet insufficient ground for believing that economic developments will generally conform to this model in the near future. Hence, the wise course for economists is to continue basic research on the nature and causes of business cycles, to remain watchful of developments that seem likely to bring on a slump in activity, and to extend the search for acceptable pathways to prosperity without inflation.

Arthur F. Burns

[See alsoEmployment and unemployment; Fiscal policy; Income and employment theory; Monetary policy. Other relevant material may be found in the biographies ofHansen>; Keynes, John Maynard; Mitchell.]

BIBLIOGRAPHY

Abramovitz, Moses 1950 Inventories and Business Cycles, With Special Reference to Manufacturers’ Inventories. New York: National Bureau of Economic Research.

Aftalion, Albert 1913 Les crises périodiques de surproduction. 2 vols. Paris: Rivière. → Volume 1: Les variations périodiques des prix et des revenus: Les théories dominantes. Volume 2: Les mouvements périodiques de la production: Essai d’une théorie.

American Economic Association 1944 Readings in Business Cycle Theory. Edited by Gottfried Haberler et al. Philadelphia: Blakiston. → Includes an article by Kondratieff. A bibliography appears on pages 443–487.

American Economic Association 1965 Readings in Business Cycles. Edited by Robert A. Gordon and Lawrence R. Klein. Homewood, III.: Irwin.

Burns, Arthur F. 1954 The Frontiers of Economic Knowledge: Essays. Published for the National Bureau of Economic Research. Princeton Univ. Press.

Burns, Arthur F.; and Mitchell, Wesley C. 1946 Measuring Business Cycles. New York: National Bureau of Economic Research.

Clark, John J.; and Cohen, Morris (editors) 1963 Business Fluctuations, Growth and Economic Stabilization: A Reader. New York: Random House. → A bibliography appears on pages 623–669.

Friedman, Milton; and Schwartz, Anna J. 1963 A Monetary History of the United States: 1867–1960. National Bureau of Economic Research, Studies in Business Cycles, No. 12. Princeton Univ. Press.

Haberler, Gottfried (1937) 1958 Prosperity and Depression: A Theoretical Analysis of Cyclical Movements. 4th ed., rev. & enl. Harvard Economic Studies, Vol. 105. Cambridge, Mass.: Harvard Univ. Press; London: Allen & Unwin.

Hansen, Alvin H. (1951) 1964 Business Cycles and National Income. Enl. ed. New York: Norton. → A bibliography appears on pages 699–710.

Hicks, John R. (1950) 1956 A Contribution to the Theory of the Trade Cycle. Oxford: Clarendon.

Johns Hopkins University, Department of Political Economy 1957 Business Fluctuations. Economic Library Selections Series 2, No. 4.

Juglar, ClÉment (1862) 1889 Des crises commerciales et de leur retour périodique en France, en Angleterre et aux É;tats-Unis. 2d ed. Paris: Guillaumin. → Partially translated as A Brief History of Panics and Their Periodical Occurrence in the United States; published by Putnam in 1916.

Keynes, John Maynard 1936 The General Theory of Employment, Interest and Money. London: Macmillan. → A paperback edition was published in 1965 by Harcourt.

Kuznets, Simon 1961 Capital in the American Economy: Its Formation and Financing. National Bureau of Economic Research, Studies in Capital Formation and Financing, No. 9. Princeton Univ. Press. → See especially pages 316–388.

Mitchell, Wesley C. 1913 Business Cycles. Berkeley: Univ. of California Press. → Part 3 was reprinted by the University of California Press in 1959 as Business Cycles and Their Causes.

Mitchell, Wesley C. 1927 Business Cycles: The Problem and Its Setting. New York: National Bureau of Economic Research.

Moore, Geoffrey H. (editor) 1961 Business Cycle Indicators. 2 vols. National Bureau of Economic Research, Studies in Business Cycles, No. 10. Princeton Univ. Press. → See Volume 1, pages 736–744, for a list of business-cycle reports by the National Bureau of Economic Research, which has led for many years in this field of study.

Persons, Warren M. 1919 Indices of Business Conditions. Review of Economics and Statistics 1:5–107.

Pigou, Arthur C. (1927)1929 Industrial Fluctuations. 2d ed. London: Macmillan.

Robertson, Dennis H. (1915) 1948 A Study of Industrial Fluctuations. London School of Economics and Political Science Series of Reprints of Scarce Works on Political Economy, No. 8. London: Aldwych.

Schumpeter, Joseph A. 1939 Business Cycles: A Theoretical, Historical, and Statistical Analysis of the Capitalist Process. 2 vols. New York and London: McGraw-Hill. → An abridged version was published in 1964.

Thorp, Willard L. 1926 Business Annals. New York: National Bureau of Economic Research.

Tinbergen, JAN 1938–1939 Statistical Testing of Business-cycle Theories. 2 vols. Geneva: League of Nations, Economic Intelligence Service. → Volume 1: A Method and Its Application to Investment Activity. Volume 2: Business Cycles in the United States of America: 1919–1932.

Tugan-Baranovskii, Mikhail I. (1894) 1913 Les crises industrielles en Angleterre. 2d ed. Paris: Giard & Brière. → First published as Promyshlennye krizisy v sovremennoi Anglii, ikh prichiny i vliianie na narodnuiu zhizn’.

Universities-National Bureau Committee for Economic Research 1956 Policies to Combat Depression. Princeton Univ. Press. → A report of the National Bureau of Economic Research, Special Conference Series, No. 7.

Wicksell, Knut (1898) 1936 Interest and Prices: A Study of the Causes Regulating the Value of Money. London: Macmillan. → First published as Geldzins und Güterpreise.

II MATHEMATICAL MODELS

A mathematical model of business cycles is not necessarily a special kind of business cycle theory as far as economic content is concerned. The mathematical formulation is an instrument for organizing our factual knowledge and our hypotheses. For this purpose, mathematical tools may be not only useful, but indispensable. Use of these tools may produce fruitful theories that could not have been discovered by verbal reasoning, and a precise mathematical formulation may serve to verify or reject previous theories set forth in a loose, verbal form and to clear the way for more systematic empirical studies.

It is not easy to date the origin of mathematical business cycle models. Fragments of such models may be found in even classical economic theory. However, it is probably fair to say that the development of explicit and complete mathematical business cycle models does not date further back than the early 1930s (see Frisch 1933; Kalecki 1935; Tinbergen 1935). These first models were of a highly macroeconomic type, involving only a few key variables to characterize the economic system. Subsequently, a very large number and variety of such models have been developed (see, for example, Samuelson 1939; Metzler 1941; Hicks 1950; Goodwin 1951). A good survey of some of these models is found in Allen’s textbook on mathematical economics (1957, pp. 209–280).

More detailed models, involving a large number of economic variables, have also been developed by Tinbergen (1938–1939), Klein (1950), and others. The purpose of these detailed models has been not only to furnish a more detailed theoretical explanation of business cycles but also to pave the way for verification and measurement by means of principles of statistical inference. Electronic computers play an increasingly important role in this kind of business cycle research.

General features

Facts and data concerning the ups and downs of business activity constitute a bewildering mass of information. Any attempt to write “the whole story of what happens” during booms and depressions is not only hopeless but also rather unrewarding as far as gaining real understanding is concerned. Somehow one has to look for principles of systematic classification and for simplifying ideas of simulation that can help to reduce the number of things to be taken into account. With this in mind, what are the general features of the dynamic process we call business cycles?

Apart from some relatively crude theories that explain business cycles as something “coming from the outside” (sunspot theories or the like), all theories of business cycles focus attention on the idea that what we observe is a result of human decision and action. The driving force is the prospect of profit or economic advantage, of one kind or another. The release, strength, and direction of such forces can be regarded as reactions to a system of signals that guide the economic activities of the various individuals or groups. These signals are prices of goods and services or other data that enter into the calculations of economic gains or losses for each decision unit.

The forces thus released are counteracted by various elements of inertia and friction due, in part, to human hesitation and slowness and to constraints set by nature or by rigid institutions.

There is also an intricate network of “feedbacks,” with the characteristic property that the feedback line from the activity of one decision group usually connects with the signal system guiding some other decision unit. Clearly, if elements of inertia and delay are present in such a system, a continued process of adjustment of some kind is almost unavoidable.

If we view the process of booms and depressions through the framework just described, the analogy with models of force and motion in mechanical engineering and related fields becomes striking. To utilize the idea of such analogies while at the same time being on guard against stretching the analogy too far is one of the main principles of mathematical model building in the field of business cycle analysis.

The notion of dynamic equilibrium. Economists have long had a deeply rooted feeling that a “normal” situation in business activity is a state of affairs where motion is absent (except, perhaps, for some kind of trend). Strangely enough, the use of mathematics to set up systems of general market equilibrium may have strengthened the hold that this notion of normality has on so many economists. Given the wide acceptance of this notion, it is understandable that ups and downs in business activity are often looked upon as “deviations from normal,” as imperfections of the market, as unforeseeable and unwanted exceptions to the rule. But such ideas are not particularly fruitful as a basis for understanding business cycles.

The point is, of course, that in a stationary situation, such as the equilibrium of the Walrasian system, the forces in operation are not zero. They are in fact very strong, but they happen to balance at zero motion. This is, however, a very special case of a balance of forces. The more general case is a balance brought about by sustained motion of certain elements of the economy. This explains why it is indeed possible to represent the process of change in business activity by means of mathematical equations based on the principle of forces in balance. The economic forces may be in balance at various rates of motion of the economic magnitudes involved, and this general idea of dynamic equilibrium is fundamental in mathematical business cycle theories.

Explanation of turning points. A central problem in business cycle theory has, of course, been to explain the turning points, i.e., to explain why expansion should turn into contraction, and vice versa. One of the major contributions of the mathematical approach to business cycle theory has been to demonstrate that the explanation of turning points is no more difficult than the explanation of any other phase of cyclical movements. Employing the notion of dynamic equilibrium, one can simply say that the relative strength of the economic forces in operation at any time will determine whether the motion necessary for balance will be up or down.

The reason why the mathematical approach is superior to a verbal analysis is obvious. By verbal reasoning it is simple enough to enumerate the various economic forces involved in a process of development, but it is often difficult, if not impossible, to determine the direction of the motion resulting from the relative strengths of the various forces.

Effects of learning—irreversibility. One objection to business cycle theories in the form of rigid mathematical models has been that they lead to a monotonous recurrence of booms and depressions of the same kind, while in fact “history never repeats itself.” Certain mathematical models are indeed open to this criticism, but others are not.

One of the remedies for this deficiency is the explicit introduction of elements of learning into the model. For example, the pattern of consumers’ demand may gradually change as a consequence of accumulating experience, or the way in which producers form their expectations (their basis for action) may gradually change as a result of their comparing past expectations with realizations. In recent years, more and more attention has been given to such elements as necessary parts of mathematical business cycle models (cf. Goodwin 1951).

While it is possible to introduce elements of irreversibility into a model in this way, it should be realized that a model must be based on the assumption that there are some aspects of economic development that repeat themselves. Otherwise, no theory, mathematical or verbal, is feasible.

Main types of models

By classifying the various business cycle models according to the principles involved rather than according to the particular economic variables dealt with, it is possible to group the models in a two-by-two table. First, we consider whether the principal active forces responsible for motion are assumed to come from the outside or are assumed to be endogenous parts of the economic system itself. The first type of model is sometimes called an open model, and the second type is called closed. Second, for each of these two types of models, we consider whether the cycles are produced because the driving force is itself cyclical (“forced oscillations”) or because of the particular way in which the economic system responds to the stimulating forces (“free oscillations”).

These principles of classification are helpful, even though a really comprehensive business cycle model may contain elements that would place it in all four categories simultaneously (cf. Samuelson 1947, pp. 335–349).

Some explicit examples will illustrate many of the points discussed above.

The cobweb model with external forces. Let x^d(t) be the demand for product x at time t, and let p(t ) be the price of x at time t. Assume that

Let x⁸(t) be the supply of x at time t, and assume that

where Έ is positive and v(t) is some external force that independently influences x⁸(t). For example, v(t) may be some weather factor or perhaps some influence from another economic sector that is independent of the one considered here. For market clearance at time t, x^d(t) must equal x⁸(t). Let x(t) be the quantity of x at which the market clears at time t. From (1) and (2) and the market-clearance condition, it will generally be possible to derive

The usual shape of supply and demand curves would imply that the first derivative of the function G is negative.

If v(t) were a constant, independent of t, this model would be the usual textbook case of the “cobweb” (Allen 1957, pp. 2–6). In such a model, there could be business cycles of period 2Έ, which would either eventually die out or go on vigorously forever.

Now let us consider the effect of changes in v(t). If x(t) oscillated but tended toward some constant when v(t) was constant, a change in v(t) to a new level would generally set the variable x(t ) in motion again; and x(t) would go on oscillating for some time, even if v(t) were to remain constant at the new level. In other words, the driving force v(t) need not itself oscillate systematically in order to generate oscillations in x(t). It is sufficient that v(t) change occasionally, perhaps in a quite irregular manner.

If v(t) should have a cycle of its own, this would, of course, have certain consequences for the resulting time shape of x(t). But x(t) would, in addition, have cyclical properties that are not present in v(t) , but are a consequence of the functional form G and of the lag Έ. In this case, the model is one of “free” oscillations with an external driving force.

Consider now the special case where Έ is equal to zero. From (3) it can be seen that in this case x(t) can be expressed directly as a function of v(t) , assuming (3) permits such a solution. Thus, x(t) could not move except when v(t) is in motion. If v(t ) had a cyclical nature, these cycles would, in some manner or other, be reflected in x(t) as “forced” oscillations.

Investment cycles—a closed model. Let C(t) denote consumption, I(t) net investment, and Y(t) income at time t in a closed economy. Then we have

Consider a simple “Keynesian” consumption function:

or, as an alternative, a dynamic version:

Suppose that for some reason there are outside forces causing independent oscillations in the rate of investment I(t). Then, in the case of the consumption function (5a), we should have forced oscillations in C(t) and Y(t). If the consumption function were (5b) instead of (5a), consumption and income could be subject to both free oscillations and forced oscillations.

The model above would be called open, because it does not “explain” the behavior of investment. The idea of the acceleration principle can be used to close the model. Let us first consider a very simple version of this idea.

Let K(t) be the (physical) amount of capital present in the economy at time t, and let K*(t) denote the amount of capital that producers would like to have at that time. If these two amounts of capital are equal, producers would be satisfied. If, on the other hand, K*(t) is larger than K(t), the demand for new capital per unit of time would be unlimited, i.e., producers would be willing to buy any rate of investment that could be supplied. Assuming that there is a capacity limit on total production in the economy, there would be an upper limit, a “ceiling,” on the amount of output of capital goods. In other words, the rate of investment would be restricted on the supply side. If, instead, K*(t) is below K(t), there would be no demand for new capital goods, not even for replacement purposes. In this situation it is, therefore, demand that determines investment, and the rate of depreciation establishes a (negative) floor under which demand cannot fall (unless capital is purposely destroyed).

The model now “explains” investment, provided we know the determinants of the desired capital stock, K*(t), and how existing capital depreciates. But if this knowledge is lacking, the model is still an open model and the question is how to close it. The simple assumptions that have been introduced for this purpose (cf. Allen 1957, pp. 242–247) are that the desired capital stock is a function of total net output; more specifically, that K*(t) and Y(t) are proportional and that there is a constant rate of depreciation.

Under these assumptions, it is easy to indicate the characteristic properties that the model would have. Suppose that consumption is given by (5a) and that the capacity to produce investment goods is sufficiently high for the amount of capital to reach and to exceed the amount of desired capital. After the desired amount of capital has been reached, output obviously must fall below capacity. But as output falls, so does the desired amount of capital, and this decline cannot stop until gross output of investment goods is zero. This would then lead to a minimum level of net output and thus to a minimum level of desired capital. Only after the existing amount of capital has been worn down below the minimum desired level could there be any demand for new capital goods. But when that situation eventually occurs, output must increase again. The amount of desired capital must then increase, and output will again reach full capacity. Thus, we have a closed model, with free and maintained oscillations.

This version of the model is, however, unsatisfactory in several respects. First, an explanation of why the desired amount of capital should be a function of output is needed. Second, there is some question as to whether it is safe to assume that the level of desired capital will ever actually be reached. And third, there is some question as to whether the closed model is not a somewhat artificial product, obtained by neglecting such things as wage policy and monetary policy.

It may be of interest to indicate briefly how some of these defects could be remedied. Let X(t) denote total gross output, and let depreciation be equal to 8∆K(t), where 8 is a constant. We then have

Assume that X(t ) is the output of a “classical” production function:

where N(t) is employment and where complementarity is assumed to exist between inputs N and K. Let the (real) wage rate, w(t), be an increasing function of employment:

Finally, let r(t) be the rate of interest.

Consider two different situations:

Situation 1. Suppose, tentatively, that there is no limit on X (t) from the demand side. Then we may assume that employment is determined by setting the marginal productivity of labor equal to the wage rate, provided total wages are below current revenues. If the marginal productivity of capital is greater than or equal to r + ∆, the demand for X will in fact be unlimited, because of the demand for an increased amount of capital.

Situation 2. Suppose, tentatively, that output, X(t), is limited by effective demand, the role of producers being simply that of producing to order. Then employment follows from (7), with K(t) given, provided total wages according to (8) are below current revenues. If the corresponding marginal productivity of capital is less than or equal to r + ΄, demand for X(t) will in fact be limited and will be equal to consumers’ demand, because producers will not want any new capital.

Whether in this model there will be a switching back and forth between the two situations, similar to that in the simpler model previously discussed, depends in an essential way on policy concerning the rate of interest. If situation 1 exists but is about to break down, lowering the rate of interest could prolong the situation. It should be noted that if situation 2 is allowed to occur, the reduction of the rate of interest that could then get us back to situation 1 would generally be much greater than the reduction that would be sufficient to maintain a situation 1 already in existence.

This model has a great deal of flexibility and can be extended to include technical progress, rachet effects in consumers’ demand, and so on.

Trygve Haavelmo

[Directly related are the entriesEconometric models, aggregate; Statics and dynamics in economics.]

BIBLIOGRAPHY

Allen, R. G. D. (1957)1963 Mathematical Economics. 2d ed. New York: St. Martins; London: Macmillan.

Frisch, Ragnar (1933) 1965 Propagation Problems and Impulse Problems in Dynamic Economics. Pages 155–185 in American Economic Association, Readings in Business Cycles. Edited by R. A. Gordon and L. R. Klein. Homewood, III.: Irwin.

Goodwin, R. M. 1951 The Nonlinear Accelerator and the Persistence of Business Cycles. Econometrica 19:1–17.

Hicks, John R. 1950 A Contribution to the Theory of the Trade Cycle. Oxford: Clarendon.

Kalecki, M. 1935 A Macrodynamic Theory of Business Cycles. Econometrica 3:327–344.

Klein, Lawrence R. 1950 Economic Fluctuations in the United States: 1921–1941. New York: Wiley. M

Etzler, Lloyd A. (1941) 1965 The Nature and Stability of Inventory Cycles. Pages 100–129 in American Economic Association, Readings in Business Cycles. Edited by R. A. Gordon and L. R. Klein. Homewood, III.: Irwin.

Samuelson, Paul A. (1939) 1944 Interactions Between the Multiplier Analysis and the Principle of Acceleration. Pages 261–269 in American Economic Association, Readings in Business Cycle Theory. Edited by Gottfried Haberler. Philadelphia: Blakiston. → First published in Volume 21 of the Review of Economic Statistics.

Samuelson, Paul A. (1947) 1958 Foundations of Economic Analysis. Harvard Economic Studies, Vol. 80. Cambridge, Mass.: Harvard Univ. Press. → A paperback edition was published in 1965 by Atheneum.

Tinbergen, Jan 1935 Annual Survey: Quantitative Business Cycle Theory. Econometrica 3:241–308.

Tinbergen, Jan 1938–1939 Statistical Testing of Business-cycle Theories. 2 vols. Geneva: League of Nations, Economic Intelligence Service. → Volume 1: A Method and Its Application to Investment Activity. Volume 2: Business Cycles in the United States of America: 1919–1932.

Business Cycles

The business cycle is the periodic but irregular up-and-down movement in economic activity, measured by fluctuations in real gross domestic product (GDP) and other macroeconomic variables. A business cycle is typically characterized by four phases—recession, recovery, growth, and decline—that repeat themselves over time. Economists note, however, that complete business cycles vary in length. The duration of business cycles can be anywhere from about two to twelve years, with most cycles averaging six years in length. Some business analysts use the business cycle model and terminology to study and explain fluctuations in business inventory and other individual elements of corporate operations. But the term "business cycle" is still primarily associated with larger (industry-wide, regional, national, or even international) business trends.

STAGES OF A BUSINESS CYCLE

Recession

A recession—also sometimes referred to as a trough—is a period of reduced economic activity in which levels of buying, selling, production, and employment typically diminish. This is the most unwelcome stage of the business cycle for business owners and consumers alike. A particularly severe recession is known as a depression.

Recovery

Also known as an upturn, the recovery stage of the business cycle is the point at which the economy "troughs" out and starts working its way up to better financial footing.

Growth

Economic growth is in essence a period of sustained expansion. Hallmarks of this part of the business cycle include increased consumer confidence, which translates into higher levels of business activity. Because the economy tends to operate at or near full capacity during periods of prosperity, growth periods are generally accompanied by inflationary pressures.

Decline

Also referred to as a contraction or downturn, a decline basically marks the end of the period of growth in the business cycle. Declines are characterized by decreased levels of consumer purchases (especially of durable goods) and, subsequently, reduced production by businesses.

FACTORS THAT SHAPE BUSINESS CYCLES

For centuries, economists in both the United States and Europe regarded economic downturns as "diseases" that had to be treated; it followed, then, that economies characterized by growth and affluence were regarded as "healthy" economies. By the end of the 19th century, however, many economists had begun to recognize that economies were cyclical by their very nature, and studies increasingly turned to determining which factors were primarily responsible for shaping the direction and disposition of national, regional, and industry-specific economies. Today, economists, corporate executives, and business owners cite several factors as particularly important in shaping the complexion of business environments.

Volatility of Investment Spending

Variations in investment spending is one of the important factors in business cycles. Investment spending is considered the most volatile component of the aggregate or total demand (it varies much more from year to year than the largest component of the aggregate demand, the consumption spending), and empirical studies by economists have revealed that the volatility of the investment component is an important factor in explaining business cycles in the United States. According to these studies, increases in investment spur a subsequent increase in aggregate demand, leading to economic expansion. Decreases in investment have the opposite effect. Indeed, economists can point to several points in American history in which the importance of investment spending was made quite evident. The Great Depression, for instance, was caused by a collapse in investment spending in the aftermath of the stock market crash of 1929. Similarly, the prosperity of the late 1950s was attributed to a capital goods boom.

There are several reasons for the volatility that can often be seen in investment spending. One generic reason is the pace at which investment accelerates in response to upward trends in sales. This linkage, which is called the acceleration principle by economists, can be briefly explained as follows. Suppose a firm is operating at full capacity. When sales of its goods increase, output will have to be increased by increasing plant capacity through further investment. As a result, changes in sales result in magnified percentage changes in investment expenditures. This accelerates the pace of economic expansion, which generates greater income in the economy, leading to further increases in sales. Thus, once the expansion starts, the pace of investment spending accelerates. In more concrete terms, the response of the investment spending is related to the rate at which sales are increasing. In general, if an increase in sales is expanding, investment spending rises, and if an increase in sales has peaked and is beginning to slow, investment spending falls. Thus, the pace of investment spending is influenced by changes in the rate of sales.

Momentum

Many economists cite a certain "follow-the-leader" mentality in consumer spending. In situations where consumer confidence is high and people adopt more free-spending habits, other customers are deemed to be more likely to increase their spending as well. Conversely, downturns in spending tend to be imitated as well.

Technological Innovations

Technological innovations can have an acute impact on business cycles. Indeed, technological breakthroughs in communication, transportation, manufacturing, and other operational areas can have a ripple effect throughout an industry or an economy. Technological innovations may relate to production and use of a new product or production of an existing product using a new process. The video imaging and personal computer industries, for instance, have undergone immense technological innovations in recent years, and the latter industry in particular has had a pronounced impact on the business operations of countless organizations. However, technological innovations—and consequent increases in investment—take place at irregular intervals. Fluctuating investments, due to variations in the pace of technological innovations, lead to business fluctuations in the economy.

There are many reasons why the pace of technological innovation varies. Major innovations do not occur every day. Nor do they take place at a constant rate. Chance factors greatly influence the timing of major innovations, as well as the number of innovations in a particular year. Economists consider the variations in technological innovation as random (with no systematic pattern). Thus, irregularity in the pace of innovations in new products or processes becomes a source of business fluctuations.

Variations in Inventories

Variations in inventories—expansion and contraction in the level of inventories of goods kept by businesses—also contribute to business cycles. Inventories are the stocks of goods firms keep on hand to meet demand for their products. How do variations in the level of inventories trigger changes in a business cycle? Usually, during a business downturn, firms let their inventories decline. As inventories dwindle, businesses eventually use down their inventories to the point where they are short. This, in turn, starts an increase in inventory levels as companies begin to produce more than is sold, leading to an economic expansion. This expansion continues as long as the rate of increase in sales holds up and producers continue to increase inventories at the preceding rate. However, as the rate of increase in sales slows, firms begin to cut back on their inventory accumulation. The subsequent reduction in inventory investment dampens the economic expansion, and eventually causes an economic downturn. The process then repeats itself all over again. It should be noted that while variations in inventory levels impact overall rates of economic growth, the resulting business cycles are not really long. The business cycles generated by fluctuations in inventories are called minor or short business cycles. These periods, which usually last about two to four years, are sometimes also called inventory cycles.

Fluctuations in Government Spending

Variations in government spending are yet another source of business fluctuations. This may appear to be an unlikely source, as the government is widely considered to be a stabilizing force in the economy rather than a source of economic fluctuations or instability. Nevertheless, government spending has been a major destabilizing force on several occasions, especially during and after wars. Government spending increased by an enormous amount during World War II, leading to an economic expansion that continued for several years after the war. Government spending also increased, though to a smaller extent compared to World War II, during the Korean and Vietnam Wars. These also led to economic expansions. However, government spending not only contributes to economic expansions, but economic contractions as well. In fact, the recession of 1953–54 was caused by the reduction in government spending after the Korean War ended. More recently, the end of the Cold War resulted in a reduction in defense spending by the United States that had a pronounced impact on certain defense-dependent industries and geographic regions.

Politically Generated Business Cycles

Many economists have hypothesized that business cycles are the result of the politically motivated use of macroeconomic policies (monetary and fiscal policies) that are designed to serve the interest of politicians running for re-election. The theory of political business cycles is predicated on the belief that elected officials (the president, members of Congress, governors, etc.) have a tendency to engineer expansionary macroeconomic policies in order to aid their re-election efforts.

Monetary Policies

Variations in the nation's monetary policies, independent of changes induced by political pressures, are an important influence in business cycles as well. Use of fiscal policy—increased government spending and/or tax cuts—is the most common way of boosting aggregate demand, causing an economic expansion. The Central Bank, in the case of the United States, the Federal Reserve Bank, has two legislated goals—price stability and full employment. Its role in monetary policy is a key to managing business cycles and has an important impact on consumer and investor confidence as well.

Fluctuations in Exports and Imports

The difference between exports and imports is the net foreign demand for goods and services, also called net exports. Because net exports are a component of the aggregate demand in the economy, variations in exports and imports can lead to business fluctuations as well. There are many reasons for variations in exports and imports over time. Growth in the gross domestic product of an economy is the most important determinant of its demand for imported goods—as people's incomes grow, their appetite for additional goods and services, including goods produced abroad, increases. The opposite holds when foreign economies are growing—growth in incomes in foreign countries also leads to an increased demand for imported goods by the residents of these countries. This, in turn, causes U.S. exports to grow. Currency exchange rates can also have a dramatic impact on international trade—and hence, domestic business cycles—as well.

BUSINESS CYCLE VARIANTS, STAGFLATION AND THE JOBLESS RECOVERY

Business cycles are difficult to anticipate accurately, in part because of the number of variables involved in large economic systems. Nonetheless, the importance of tracking and understanding business cycles has lead to a great deal of study of the subject and knowledge about the subject. It was as a result somewhat surprising when, in the 1970s, the nation found itself stuck in a period of seemingly contradictory economic conditions, slow economic growth and rising inflation. The condition was named stagflation and paralyzed the U.S. economy from the mid-1970s through the early 1980s.

Another somewhat unexpected business cycle phenomenon has occurred in the early 2000s. It is what has come to be known as the "jobless recovery." According to the National Bureau of Economic Research's Business Cycle Dating Committee, in a late 2003 report, "the most recent economic peak occurred in March 2001, ending a record-long expansion that began in 1991. The most recent trough occurred in November 2001, inaugurating an expansion." The problem with the expansion has been that it has not included a rise in employment or real personal income, something seen in all previous recoveries.

The reasons for the jobless recovery are not fully understood but are the cause of much debate within the economic and political circles. Within this debate there are four leading explanations that analysts have given for the jobless recovery. According to a study published in Economic Perspectives in the summer of 2004, these four explanations are:

• An imbalance in labor available by sector.
• The emergence of just-in-time hiring practices.
• The rising cost of health care benefits.
• Rapidly increasing productivity not being off-set by aggregate demand.
• Only time and further analysis will show which of these factors, or which combination of factors explains the advent of a jobless recovery. Neil Shister, editorial director of the World Trade summarizes a discussion of the jobless recovery this way, "The culprit is ourselves. We have become dramatically more productive." This assessment suggests that much more will need to be understood about modern business cycles before we can again anticipate them and plan for their effects on the economy generally.

KEYS TO SUCCESSFUL BUSINESS CYCLE MANAGEMENT

Small business owners can take several steps to help ensure that their establishments weather business cycles with a minimum of uncertainty and damage. The concept of cycle management is earning adherents who agree that strategies that work at the bottom of a cycle need to be adopted as much as those which work at the top of a cycle. While there is no definitive formula for every company, the approaches generally emphasize a long-term view focused on a company's core strengths and stressing the need to plan with greater discretion at all times. Essentially, efforts are made to adjust a company's operations in such a manner that it maintains an even keel through the ups and downs of a business cycle.

Specific tips for managing business cycle downturns include the following:

• Flexibility—Having a flexible business plan allows for development times that span the entire cycle and includes various recession-resistant funding structures.
• Long-term Planning—Consultants encourage small businesses to adopt a moderate stance in their long-range forecasting.
• Attention to Customers—This can be an especially important factor for businesses seeking to emerge from an economic downturn. Maintaining close relations and open communication with customers is a tough discipline to maintain in good times, but it is especially crucial coming out of bad times. Customers are the best gauges of when a company is likely to begin recovering from an economic slowdown.
• Objectivity—Small business owners need to maintain a high level of objectivity when riding business cycles. Operational decisions based on hopes and desires rather than a sober examination of the facts can devastate a business, especially in economic down periods.
• Study—Timing any action for an upturn is tricky. The consequences of getting the timing wrong, of being early or late, can be serious. How, then, does a company strike the right balance between being early or late? Listening to economists, politicians, and media to get a sense of what is happening is useful. The best route, however, is to avoid trying to predict the upturn. Instead, listen to your customers and know your own response-time requirements.

BIBLIOGRAPHY

Aaronson, Daniel, and Ellen R. Rissman; Daniel G. Sullivan. "Assessing the Jobless Recovery." Economic Perspectives. Summer 2004.

Arnold, Lutz G. Business Cycle Theory. Oxford University Press, 2002.

Bonamici, Kate. "Why You Shouldn't be Scared of Stagflation." Fortune. 31 October 2005.

Hall, Robert, and Martin Feldstein. The NBER's Business-Cycle Dating Procedures. National Bureau of Economic Research, 21 October 2003.

Hendrix, Craig, and Jan Amonette. "It's Time to Determine Your E-Business Cycle." Indianapolis Business Journal. 8 May 2000.

Marshall, Randi F. "Is Stagflation Back?" Newsday. 29 April 2005.

Nardi Spiller, Christina. The Dynamics of the Price Structure and the Business Cycle. Business & Economics, August 2003.

Shister, Neil. "Global Trade and the 'Jobless Recovery'." World Trade. October 2004.

Walsh, Max. "Goldilocks and the Business Cycle." The Bulletin with Newsweek. 7 December 1999.

                              Hillstrom, Northern Lights

                                updated by Magee, ECDI

BUSINESS CYCLES

BUSINESS CYCLES are the irregular fluctuations in aggregate economic activity observed in all developed market economies. Aggregate economic activity is measured by real gross domestic product (GDP), the sum weighted by market prices, of all goods and services produced in an economy. Comparisons of real GDP across years are adjusted for changes in the average price level (inflation). A business cycle contraction or recession is commonly defined as at least two successive three-month periods (quarters) in which real GDP falls. A business cycle then contains some period in which real GDP grows followed by at least half a year in which real GDP falls. Some business cycles are longer than others. As Table 1 shows, most contractions last for less than a year, with real GDP falling by 1 to 6 percent. Expansions are more variable, though most last from two to six years.

Business cycles date from at least colonial times. The data for the colonial period are limited; thus, it is more difficult to date cycles precisely. When the United States was primarily agricultural, fluctuations in climate exerted a strong influence on economic cycles. While business cycles are defined in terms of GDP, a number of other economic variables tend to move in concert with GDP. Aggregate consumption expenditures rise and fall with GDP. Investment does too, but it tends to rise much faster than GDP during expansions and to fall much faster in recessions. The trade balance increases as GDP falls and vice versa, and imports in particular tend to rise during expansions and fall during contractions. Interest rates, notably short-term interest rates, tend to rise in expansions and fall in contractions. The aggregate price level often moves up and down with GDP, as do profits.

Aggregate employment also rises during expansions and falls during contractions, usually fluctuating less than GDP. The unemployment rate rarely rises by more than 4 percent in a recession, in part because the average hours worked per worker rises and falls with the cycle. Also, firms tend to "hoard" some workers during recessions to obviate the need to rehire as the expansion begins. Consequently, output per worker falls during recessions.

Other economic variables sometimes move in concert with GDP. The "leading indicators," which tend to precede changes in GDP, include capacity utilization by industry, construction starts or construction plans, orders received for capital equipment, new business formation, new bond and equity issues, and business expectations. Studying these along with the aggregate variables, analysts attempt to forecast cycles, which is especially difficult when predicting the timing of turning points between expansion and contraction. While severe contractions affect every sector of the economy, milder contractions are observed only in some sectors, and employment continues to rise in about a quarter of industries.

Different Types of Cycles

Scholars in the early post–World War II period often distinguished "growth cycles," in which contractions were defined as a decline in the rate of GDP growth, from the less frequent business cycles, in which contractions were defined as decreases in GDP. Some held out hope that the business cycle could be replaced with less severe growth cycles.

While the chronology of business cycles produced by the National Bureau of Economic Research (NBER) (Table 1) is widely accepted, different definitions of the term "contraction" could easily combine or subdivide particular cycles. Indeed, the NBER does not adhere strictly to the definition of a recession as two successive quarters of GDP decline. Instead, a committee determines, by looking at movements in variables, when turning points have occurred. Some scholars have criticized this committee's decisions.

Economists from time to time have hypothesized the existence of at least three other economic cycles, including a shorter Kitchin or inventory cycle, identified by Joseph Kitchin in 1923, of about forty months in length; a Kuznets cycle, suggested by Simon Kuznets in 1958, of fifteen to twenty-five years in duration; and a Kondratiev or Long Wave cycle, popularized by Nikolai Kondratiev in 1922, of fifty to sixty years in duration. However, none of these is accepted as widely as the business cycle, which bears a strong similarity to the cycle identified by Clément Juglar in the 1860s.

Debate regarding Long Waves has been intense. If these exist, only a handful would have occurred in the modern era, and the major wars complicate interpretation of the historical record. Moreover, explaining fairly regular fluctuations of a half-century duration is arguably a more difficult theoretical task than explaining business cycles. Analyzing the same variables, most scholars of Long Waves emphasize the interplay among technological and economic phenomena. While the existence of fairly regular Long Waves is disputed both empirically and theoretically, the historical record clearly shows periods of more than one business cycle in length in which economic growth and employment were high, such as the 1950s and 1960s, and other periods of more than a business cycle in length in which economic growth was sluggish at best and unemployment was high, such as the 1930s or the 1970s and 1980s. Such periods likely require a different type of explanation. An understanding of the causes of economic growth in general should in turn inform the understanding of business cycles, because fluctuations would not likely be seen in a world without growth.

Table 1

Table 1

The fact that growth rates are higher in some periods than others poses difficulties for the empirical evaluation of business cycles. Ascertaining the severity of the business cycle requires knowing the growth rate around which cyclical fluctuations occur. But observation of a change in GDP from one year to the next conflates the effect of the trend growth rate and the effect of the cycle. Thus, analysts use complex and controversial statistical techniques to distinguish trends from cycles. This task would increase in complexity if economists accepted the existence of more than one type of cycle.

The existence of natural seasonal fluctuations in economic activity, associated with climatic changes and the bunching of purchases around holidays such as Christmas, adds another complication. Economists prefer to look at "seasonally adjusted" figures when evaluating economic performance. Has the change from month to month been greater or less than is usually observed between those two months? But as the economy evolves, so does the desirable seasonal adjustment.

Causes of Business Cycles

Economists have long debated the causes of business cycles, especially since the Great Depression. At that time, macroeconomics, the study of aggregate economic activity, emerged. Economists increasingly have recognized, however, that an understanding of business cycles requires microeconomic foundations, that is, an under-standing of how the interaction of individuals and firms in the markets for goods and services, finance, and labor generates business cycles.

Theories of business cycles can be divided into two broad categories. The first argues that cycles are exogenous or due to a variety of shocks. These shocks stimulate either economic expansion or contraction. The second argues that cycles are endogenous or self-generated by the market economy. Theoretical debates often have an ideological tinge influenced by scholarly attitudes toward the market economy and the desirability of government interference. Nevertheless, after decades of often heated debate, economists widely recognize that one right answer is not likely. Different forces have differential impacts on different cycles, and both exogenous and endogenous arguments have some explanatory power.

Exogenous theories emphasize a variety of shocks. Some speak of political shocks; for instance, politicians may encourage economic expansion just before elections. Shocks to the prices of important raw materials, such as oil, are often mentioned at least with respect to particular cycles.

More commonly, scholars argue that increases in the money supply encourage expansions and that central banks, fearing inflation, then restrict the money supply and trigger a contraction. To be sure, the supply of money does tend to rise and fall through cycles, but the debate concerns whether this is in large part a result of or a cause of cycles. Central banks, such as the Federal Reserve Bank, are not the sole influences on the money supply, which is affected also by the level of economic activity and the behavior of individual banks.

During the Great Depression and again later, some economists pointed to technological shocks. If, as seems to be the case, innovation does not occur evenly through time, then investment, consumption, and employment decisions would be expected to vary through time as a result. One problem that plagues this analysis is the difficulty of measuring innovation. Moreover, different innovations likely have different effects on different sectors. The development of new products likely has a positive effect on employment, while the development of better ways of producing existing products likely has a negative effect on employment.

Theories of business cycles must grapple with two opposing questions: Why is economic activity not stable, and how is complete chaos avoided (why do both contractions and expansions always end)? The common presupposition is that equilibrating mechanisms take the economy back toward the trend growth rate but are sluggish in operation. Examples of equilibrating mechanisms include the tendency of firms to increase production as inventories fall, the tendency of people to buy more as prices fall, or the tendency of firms to hire more as wages fall. Another rarely discussed possibility is that shocks of opposing effects may hit the economy. Most of the time these shocks are roughly balanced, and thus cycles are not too severe. Occasionally, as during the Great Depression, shocks are unbalanced and produce lengthy expansions or contractions.

Exogenous theories need only posit some set of shocks and usually some imperfect equilibrating mechanisms. Endogenous theories must argue both for equilibrating mechanisms and for nonequilibrating mechanisms that take the economy away from its trend growth rate. One early example was the general theory of John Maynard Keynes in 1936. Keynes noted that any expenditure has a multiplier effect, as the person receiving the money in turn spends it and so on. He also recognized an accelerator effect in that any attempt to increase output would require a much greater increase in the rate of investment. The multiplier-accelerator mechanism would cause any positive or negative growth impulse to be magnified and the economy to move further away from the trend growth rate. Writing during the Great Depression, Keynes was skeptical that any equilibrating mechanisms were strong enough always to reverse a contraction. Subsequently, the followers of Keynes stressed that inflexibility in wages and prices can cause an economy to move away from the trend growth rate.

A variety of other endogenous approaches is possible. Banks may naturally increase and decrease credit through time. Businesses may habitually overinvest as they fight for market share and then cut back in the face of over-capacity. Businesses may also saturate markets for consumer durables and inevitably experience a sudden drop in demand for such goods, which in turn induces them to reduce production.

Theories of business cycles strive to explain not only movements in GDP but in those variables that tend to move in concert with GDP. Some theories posit that unemployment is largely voluntary. Workers adjust work decisions in response to changes in real wages, or perhaps only to perceptions of such changes if they are fooled by changes in price levels. Other theories stress the involuntary nature of unemployment. During contractions many individuals cannot find work, at least not at anything approaching previously available wage rates. The evidence from unemployed individuals seems to support the latter position, though unemployment rates are constructed on surveys that tend not to ask the unemployed what sort of wage offer they seek.

All theories of business cycles face the problem of the role of expectations. No doubt expectations influence important economic variables, notably business investment and consumer durable purchase decisions. But how are expectations formed? Do they respond primarily to movements in economic variables, and if so, do they respond in a predictable fashion?

Effects of Business Cycles

As noted, business cycles affect variables such as employment, profits, hours of work, and often prices and wages. They thus have a significant impact on people's lives. In severe contractions a sizable proportion of the population loses its income. This was especially true before the creation of unemployment insurance and welfare, when the unemployed depended on charity. Unemployment in turn can affect a variety of noneconomic variables, including decisions regarding marriage and having children; mental health; attitudes toward the wider society, including the potential for civil disorder; and voting patterns, giving politicians a greater chance of reelection during times of economic expansion.

With the advantage of hindsight, economists know that all contractions end, and most end fairly quickly. It is thus all too easy to downplay the effects of cycles. Families may be unable to either borrow or save enough in advance to avoid serious hardship during a contraction, and fears of a future recession may cause families to forgo investments in houses and cars. Moreover, individuals who come of age during a serious recession may find their entire lives affected, as during the next expansion prospective employers may eschew those who have been long unemployed.

Changes in Business Cycles

Since business cycles involve an interaction among several economic and likely several noneconomic variables, some changes in the character of business cycles, including average duration, severity of fluctuations, or impact upon different sectors of the economy, should occur as an economy develops. Some have argued for the existence of a "new economy," in which the application of information technology would lessen the severity of cycles, yet most economists have been skeptical.

Considerable debate has focused on this question: Have business cycle fluctuations become less severe than they were before World War I? The debate has hinged primarily on the estimation of movements in GDP before such statistics were collected by the government. Most economists accept that business cycles involve longer expansions and shorter and shallower recessions than did those before World War I.

Most scholars attribute the longer expansions and shorter recessions primarily to government initiatives. The establishment of automatic stabilizers, such as unemployment insurance, have ensured that workers do not lose their entire incomes and thus their ability to spend when they lose their jobs in a recession. In addition, the government and the Federal Reserve have striven to adjust spending, taxation, and the money supply to reduce the severity of cycles. By putting more or less money in people's hands, they aim to increase or decrease the level of economic activity. Some economists worry that the government, due to a limited ability to predict cycles plus the time required to actually adjust spending or tax decisions, as often as not worsens cycles by, say, increasing spending after a contraction has already ended. Another concern is that taxpayers, faced with an increase in government debt, may reduce their own spending to save in anticipation of future tax increases. Nevertheless, substantial empirical evidence indicates that changes in government spending and taxation do affect the level of economic activity.

Other possible explanations of increased economic stability include the lesser incidence of financial panics, which were an important component of nineteenth-century recessions, due in large part to deposit insurance, introduced in 1934; increased flexibility of wages and prices, important for equilibrating mechanisms; management of inventories so firms do not build them up at the start of a downturn, then slash production to compensate; increased importance of the service sector, which tends to be less volatile than industry because many goods are purchased irregularly; and increased business confidence that downturns will be short.

Analysis of Particular Cycles

The discussion above suggests that different theoretical approaches have different explanatory power with respect to particular cycles. No approach should be ignored in studying any cycle. An obvious danger is to ignore endogenous arguments in favor of unique exogenous shocks when analyzing a particular cycle. It is possible though that exogenous shocks loom larger in the more severe cycles that attract most historical attention.

Banking panics were a common characteristic of recessions as late as the Great Depression. The resulting bank failures surely exacerbated contractionary tendencies, for people could not spend money they had lost. The question is how great this contractionary tendency was relative to the size of particular recessions. Likewise, stock market crashes can have a contractionary impact; these dramatic events may receive more attention than warranted by their economic impact. More generally, changes in interest rates and money supply often are associated with cycles and are attributed an important causal role. The 1929 crash has often been blamed for inducing the Great Depression, although the crash of 1987 was not associated with a serious economic downturn.

Sudden increases in raw material prices, such as occurred with copper during the early days of electrification in 1907 or oil in the 1970s, likely played some role in inducing recessions. Electrification, the assembly line, the automobile, and the television are among a number of major technological innovations that almost certainly had some impact on the level of economic activity and employment. As prices of many products rose, consumers and investors reduced their purchases. Saturated markets for houses, cars, and other durables have been observed during contractions in the 1930s and the 1970s. New products usually, but not always, cause increased investment and employment, while new production processes generally cause employment to fall.

Since the Great Depression, governments have taken an active role in trying to affect economic activity by adjusting the level of taxes and spending. Central banks too have tried to affect economic activity through adjustments to the money supply or interest rates. Even before the Great Depression, major government expenditures, such as during war or for the development of transport infrastructure, would have had some expansionary effect. As noted above, economists debate whether governments thus alleviate cycles or instead make these worse by increasing spending during expansions or reducing spending in recessions.

BIBLIOGRAPHY

Berry, Brian J. L. Long-Wave Rhythms in Economic Development and Political Behavior. Baltimore: Johns Hopkins University Press, 1991.

Burns, Arthur F., and Wesley C. Mitchell. Measuring Business Cycles. New York: National Bureau of Economic Research, 1946. A classic work by two leading scholars of business cycles of the time.

Diebold, Francis X., and Glenn D. Rudebusch. Business Cycles: Durations, Dynamics, and Forecasting. Princeton, N.J.: Princeton University Press, 1999. Discusses the statistical analysis of business cycles, and compares pre–World War I and post–World War II cycles.

Friedman, Milton, and Anna J. Schwartz. A Monetary History of the United States, 1867–1960. Princeton, N.J.: Princeton University Press, 1963. Classic argument for the importance of changes in monetary variables in generating cycles; unusually relies on examinations of particular cycles rather than on statistical analysis across several cycles.

Goldstein, Joshua S. Long Cycles: Prosperity and War in the Modern Age. New Haven, Conn.: Yale University Press, 1988.

Hall, Thomas E. Business Cycles: The Nature and Causes of Economic Fluctuations. New York: Praeger, 1990. A rare combination of a theoretical survey and application to selected twentieth-century cycles.

Keynes, John Maynard. The General Theory of Employment, Interest, and Money. London: Macmillan, 1936. Generally recognized as having spawned the field of macroeconomics.

National Bureau of Economic Research. "U.S. Business Cycle Expansions and Contractions." Available at http://www.nber.org/cycles.html.

Ralf, Kirsten. Business Cycles: Market Structure and Market Inter-action. New York: Physica-Verlag, 2000. Surveys modern theories, with an emphasis on microeconomic foundations.

Schumpeter, Joseph A. Business Cycles. New York: McGraw-Hill, 1939. Classic argument for the existence of four cycles of differing average durations.

Solomou, Solomos. Phases of Economic Growth, 1850–1973: Kondratieff Waves and Kuznets Swings. New York: Cambridge University Press, 1987. Argues for the existence of Kuznets cycles.

Zarnowitz, Victor. Business Cycles: Theory, History, Indicators, and Forecasting. Chicago: University of Chicago Press, 1992. Discussion of the evolution of the measurement of business cycles by a scholar long connected with the NBER.

RickSzostak

See alsoBanking ; Economic Indicators ; Economics ; Federal Reserve System ; Financial Panics ; Great Depression ; Keynesianism ; National Bureau of Economic Research ; Unemployment .

Recession

DETERMINING THE TIMING OF RECESSIONS

CHANGES IN THE SEVERITY OF RECESSIONS

WHAT CAUSES RECESSIONS?

BIBLIOGRAPHY

A recession is the declining phase of a business cycle, when seasonally adjusted output falls significantly and unemployment increases, though by no means in every industry. A recession should not be confused with periods of low output and high unemployment. Early in a recession output, though falling, is still above its trend value. Similarly, shortly after the recession ends, output is still below its prerecession level. Recessions should not be confused with depressions. While economic historians sometimes use this term to refer to conditions in the early 1890s, most American economists use it only in connection with the Great Depression, which started in August 1929 and encompassed one catastrophic recession until March 1933. This was followed by a recovery that was succeeded by another sharp recession from May 1937 to June 1938.

DETERMINING THE TIMING OF RECESSIONS

There are three major ways of determining when a recession has occurred. A popular one is two or more successive quarters of declining GDP (gross domestic product). This is simplistic. Suppose quarterly GDP grows at the following rates: –2.0 percent, 0.1 percent, –2.5 percent. Under this definition that would not be a recession. Moreover, since official GDP figures are issued only quarterly in the United States, it cannot provide the months of turning points, without relying on unofficial estimates. And looking only at a single variable, GDP, is risky because the U.S. Department of Commerce revises, sometimes substantially, its initial GDP estimates.

Economists therefore use more complex procedures. By far the most widely accepted is that of the National Bureau of Economic Research (NBER), a private research organization. As it explained in an October 21, 2003, press release titled “The NBER’s Business-Cycle Dating Procedure” it defines a recession as “a significant decline in economic activity spread across the economy, lasting more than a few months, normally visible in real GDP, real income, employment, industrial production, and wholesale-retail sales.” It determines when a recession started or ended by looking primarily at four statistical series, the most important of which is quarterly real GDP, supplemented by data that allow it to gauge monthly GDP. The other series it uses are real personal income minus transfer payments, employment, industrial production, and real wholesale and retail sales. Primarily because the initially issued data are subject to substantial revisions, it identifies the cyclical peaks and troughs only with substantial lags; for example it did not identify the November 2001 trough until July 2002. The NBER has also identified peaks and troughs back to 1857. To do so it had to rely on a large number of less comprehensive series, and on contemporaneous descriptions of business conditions. The table shows the NBER’s turning points.

Although widely used, the NBER’s estimation of turning points has been criticized for being subjective or based on arbitrary rules. Take for example the following hypothetical growth rates –0.1, 0.1, –2.0, –1.8, –0.9. Did the recession start with the 0.1 percent decline, or only with the 2.0 decline? Thus, Christina Romer estimated alternative turning points in her 1999 article “Changes in Business Cycles: Evidence and Explanations.” The largest discrepancy between the NBER and Romer’s estimates, five months, occurred in the 1982 trough. Another problem is the NBER’s delay in announcing turning points. Hence some economists developed methods that, using the growth rate of GDP in successive quarters, allow them to quickly determine in an objective way the likelihood that the NBER will subsequently declare that a turning point has occurred.

Following the NBER’s lead turning point chronologies have also been developed for other countries. There is some tendency for recessions to be synchronized in various countries since a recession in one country lowers its demand for the products of other countries, thus “exporting” part of its recession.

CHANGES IN THE SEVERITY OF RECESSIONS

There is much debate about whether U.S. recessions have been milder in the post–World War II era. What is clear is that recessions have become substantially shorter since 1983-1984, a phenomenon called “the great moderation.” The reasons for this are controversial. One hypothesis is that computerization has reduced inventory investment. That matters because about half of business-cycle fluctuations consist of fluctuating in inventories. However, the timing does not fit: Computerization has been a steady process, while the great moderation started abruptly. Another hypothesis points to other institutional changes: the larger role of government with its automatic stabilizers; the greater availability of consumer credit, which allows the unemployed to maintain consumption; and the shift from manufacturing to the more stable service sector. But these changes are not large enough to explain the great moderation. A third hypothesis points to the greatly increased efficiency of monetary policy. Finally, some economists attribute the moderation to good luck; that is, to the shocks to the economy being smaller since 1983. Several of these explanations may well be part of the story. In any case, unlike the fluctuations that the NBER measures, fluctuations in the unemployment rate have not become shorter. This is possible because the peaks and troughs in unemployment lag those of GDP.

Given that recent U.S. recessions have been moderate, how damaging are they? Some economists argue that relative to changes in the trend rate of growth they are unimportant. Imagine that the trend rate of growth falls from 3 percent to 2.75 percent. After thirty years income has grown by only 86 percent, instead of by 126 percent as it would have at the 3 percent growth rate, a 40 percent difference. Compare that to a recession that lowers GDP by 5 percent for a year. However, happiness studies have shown that above a certain threshold long-run per capita income has little effect on happiness, while temporary unemployment has a substantial effect. Moreover, recessions could reduce the trend rate of growth.

WHAT CAUSES RECESSIONS?

Explaining why initial declines in income are followed by subsequent declines is relatively easy. As their incomes fall consumers spend less, and falling sales also lower the incentive to invest. Moreover, if prices fall (or rise less than expected) the real interest rate exceeds the rate borrowers expected, and that may drive them into bankruptcy; a more salient danger in the nineteenth century when prices declined much more in recessions than they do now.

There is much less consensus on what causes the initial downturn. One should distinguish two types of explanations. The first considers recessions as endogenous (related to) to capitalism, with expansions generating forces that terminate them. The other considers recessions to be due to exogenous, or outside, shocks. An old-fashioned example of the former is underconsumption theory. A more modern version is that expansions generate overconfidence that leads to excessive borrowing, risky lending, and lax financial practices that make firms vulnerable to failure.

Modern theories stress exogenous shocks. Monetarist theory, which for major U.S. recessions has much empirical support, focuses on declines in the growth rate of money due mainly to bank failures (until 1934) and since 1921 to inept monetary policy. It faces the problem that the growth rate of money may be endogenous rather than causal. Another theory argues that recessions are due to negative shocks to productivity that lower equilibrium real wage, which workers, who do not know this, refuse to accept. Still another theory claims that the government adopts expansionary policies prior to elections, and subsequently offsets them with restrictive policies.

More than one theory may be correct, because recessions differ. The 1914 recession was not preceded by a monetary shock, the 1921 recession was. The 1945 recession was due to end of wartime pressures for high output, while the 1973-1975 recession followed a doubling of oil prices to which the U.S. Federal Reserve Board initially responded with a restrictive policy, and the 1981 recession probably resulted from a major shift in monetary policy. The 1929 stock market crash may perhaps have played a significant role in the severity of the 1929-1933 recession, while subsequent crashes did much less damage.

SEE ALSO Business Cycles, Political; Business Cycles, Real; Depression, Economic; Economic Crises; Great Depression; Stagflation; Stagnation; Underconsumption

BIBLIOGRAPHY

Friedman, Milton, and Anna Schwartz.1963. A Monetary History of the United States, 1867-1960. Princeton, NJ: Princeton University Press.

Hall, Robert, et al. 2003. The NBER’s Business-Cycle Dating Procedure. The National Bureau of Economic Research, October 21.

McAdams, Peter. 2003. US, Japan and the Euro-Area: Comparing Business-Cycle Features, European Central Bank Working Paper 283.

Romer, Christina. 1999. Changes in Business Cycles: Evidence and Explanations. Journal of Economic Perspectives 13 (2): 23-44.

Sill, Keith. 2004. What Accounts for the Postwar Decline in Economic Volatility? Business Review (Q1): 23-31.

Thomas Mayer

Recession

What It Means

A recession is a downturn in the economy. Economists have two ways of identifying when a recession is occurring. According to the most precise definition, a recession is a decline in a country’s gross domestic product, or GDP (the total value of all goods and services produced within that country in a specific time period), for two or more successive quarters (in the financial world, each year is commonly broken down into four three-month periods called quarters). For practical purposes, however, most economists agree that a recession is best defined more loosely as an extended period of decreased economic activity marked by the following characteristics: high unemployment rates (a measure of the number of people who want to work but do not have jobs), a decline in the profits made by corporations, and a decrease in the amount of money people are investing in the stock market (a central location, such as the New York Stock Exchange, where people buy stocks, which are shares of ownership in corporations, in order to receive shares in their profits).

Typically recessions can last anywhere from 6 to 18 months. (Before the Great Depression, which lasted from 1929 until about 1938 in the United States, every economic setback was considered a recession. After the Great Depression economists used the word depression to characterize a particularly long and severe recession.) During a recession interest rates (the fees that bank customers pay to borrow money) tend to fall. Low interest rates can then offer a way out of the recession. As bank loans become cheaper, more people are likely to apply for mortgages (loans for homes), and more corporations are likely to apply for loans to expand their business. These loans put more money into circulation, stimulating the economy in a way that creates more jobs, more spending, and more corporate profit. Since World War II the average recession in the United States has lasted 11 months.

Economists agree that a recession is a normal and inevitable part of the business cycle, which consists of five stages. The first stage is expansion or growth in the economy, which occurs when more people invest in the stock market and buy homes. During an expansion, unemployment rates go down. The second stage, the high point of the expansion, is called the peak. The peak is followed by the contraction or recession stage. The fourth stage is the low point, or trough, which is followed by the recovery stage, during which the economy begins to regain its strength.

When Did It Begin

The first major downturn in the U.S. economy was called the Panic of 1819 (which lasted from 1819 until 1824). Most economists trace the beginning of this recession all the way back to the War of 1812, a three-year conflict with Great Britain that ended in 1815. Most of the battles took place in the Great Lakes area of the United States (on U.S. soil) and in what are now the provinces of Ontario and Quebec in Canada. In order to finance the war, the American government borrowed money from U.S. banks, which in turn produced more bank notes (money) to help fund the war. More banks opened and produced still more notes, causing a boom (or upturn) in the economy. During the boom investors and families borrowed money to purchase land and to improve farms. Some banks produced false notes to keep up with borrowers’ demands.

At the time paper money was backed by silver, which means that a bank note represented a certain amount of silver stored at the bank. Banks realized that they did not have the proper backing for the amount of paper money and loans they were distributing, and they decided to stop producing so many bank notes and issuing credit. Investors responded to this contraction by bringing their notes to the bank and attempting to cash them in for silver. Many banks could not honor customers’ requests (when too many people demand their money at once, it is called a run on the bank) and were forced to close, further escalating the runs on banks and worsening the recession. Though the crisis struck the entire nation, it was most severe in Philadelphia, where unemployment reached 75 percent. Most economists agree that the Panic of 1819 was a case of a young country’s first encounter with the difficult challenges of the business cycle.

More Detailed Information

The terms recession and depression are the words most commonly used when describing an economic slowdown. Economists do not agree on a precise standard for distinguishing between a recession and a depression, however. In the most general sense they do agree that a depression is a more severe and longer lasting economic decline than a recession, but there is no consensus about exactly when a recession should be considered a depression. A common joke among accountants, economists, and finance professionals says, “A recession is when a friend loses his job. A depression is when you lose your job”; in other words, a person only recognizes the severity of an economic crisis when he or she experiences that crisis directly.

Distinguishing between a recession and a depression may seem like a trivial concern, but the differences between the two types of financial crisis are important for members of a society to recognize. This is so because recovery from an economic decline depends in a large part on consumer confidence. If investors believe in the value of the stocks they are purchasing, they will continue to purchase stocks and thereby keep the economy strong. If, however, any decline in the economy is immediately labeled a depression by the press in newspapers and on the news, investor panic is likely to ensue. Nervous investors often try to unload their stocks. When there is rampant selling and limited buying on the stock market, stock prices fall, and real crisis occurs. In other words, misrepresenting the severity of an economic downturn can cause that decline to intensify.

Some economists claim that the most accurate way to distinguish between a recession and a depression is to examine the actual decline in the GDP during an economic downturn. According to these economists, the country is in a depression when the GDP declines by 10 percent or more. Anything less than a 10-percent decline qualifies as a recession. Using this definition, the United States has not come close to experiencing a depression since the infamous Great Depression. During that period there were actually two separate depressions. In the first, which occurred between 1929 and 1933, the GDP declined by 33 percent. In the second depression, which lasted from 1937 to 1939, the GDP fell by 18 percent. Nothing resembling either of these depressions has happened to the U.S. economy since.

The fiscal disaster that came the closest to being labeled a depression occurred in 1973, when the members of OAPEC (Organization of Arab Petroleum Exporting Countries) refused to ship petroleum to supporters of Israel, a list of nations that included the United States, Japan, and most of the countries in Western Europe. In the United States imports of Arab oil slipped from 1.2 million barrels a day to 19,000 barrels a day. The price of the oil quadrupled, and the cost of gas at the pump rose from $.39 per gallon to $.55 per gallon. As a result of the crisis, the value of stocks on the New York Stock Exchange dropped $97 billion, and the GDP fell by 4.9 percent. Though these certainly qualified as what many people would call hard times, the crisis was a recession rather than a depression.

Though the United States has not experienced a depression since the 1930s, there have been some bad days on the stock market, and at the time people feared that a depression could be coming. One of these days was October 17, 1987, a day commonly referred to as Black Monday. The stock market endured the second-largest one-day drop in its history on that day (September 17, 2001, was the largest), and by the end of October 1987, the U.S. stock market had fallen by 22.6 percent. Other countries, such as Australia, China, and Canada, experienced similar losses that day. Economists do not agree on the causes of the sharp decline, but the crisis was short lived in large part because the media nurtured popular faith in the economy and urged investors not to dump their stocks.

Recent Trends

The recession in the United States that began in 2001 is believed to have had three causes: the widespread failure of dot-com (technology) businesses; the September 11, 2001, terrorist attacks on the World Trade Center and the Pentagon; and two well-publicized cases of large-scale accounting fraud. The widespread failure of technology businesses (what has been called the bursting of the dot-com bubble) began in March 2000, when stock prices for such leading technology firms as Cisco, IBM, and Dell began to fall. Most economists claim that these and other technology companies suffered sizeable losses in the stock market because they borrowed too much start-up money in the mid- and late 1990s and tried to expand the Internet-based sections of their businesses without sufficient planning.

The U.S. stock market also experienced significant losses in the aftermath of the September 11, 2001, terrorist attacks. The New York Stock Exchange, NASDAQ, and the American Stock Exchange closed on the day of the attacks and did not reopen until September 17, 2001. It was longest shutdown of trading in the United States since the Great Depression. When the market reopened, American stocks dropped $1.2 trillion in value in the first week of trading. In addition to the loss of the two World Trade Center towers, many other businesses in lower Manhattan suffered structural damage and the loss of crucial files and records. Shortly after the attacks Enron Corporation, a leading energy company based in Houston, Texas, was charged with accounting fraud. While the press was covering the scandal in November and December 2001, Enron’s stock dropped from $90 per share to $.30 per share. Enron filed for bankruptcy in December 2001. The following year WorldCom, a telecommunications company that later became MCI, also declared bankruptcy, largely because of reports that they, too, were guilty of accounting fraud.