Wages

I. THEORY

BIBLIOGRAPHY

II. STRUCTURE

BIBLIOGRAPHY

III. SYSTEMS OF PAYMENT

BIBLIOGRAPHY

IV. WAGE AND HOUR LEGISLATION

BIBLIOGRAPHY

V. FRINGE BENEFITS

BIBLIOGRAPHY

I. TheoryAllan M. Cartter

II. StructureMelvin W. Reder

III. Systems of PaymentRobert B. McKersie

IV. Wage and Hour LegislationN. Arnold Tolles

V. Fringe BenefitsPeter Henle

I. THEORY

The majority of adults in industrial societies participate in economic life as wage or salary earners; thus, the principles determining the level of remuneration for labor are among the most interesting aspects of economic theory. At the same time, however, because the exchange of services for pay involves a personal relationship between employer and employee, wage determination is complicated by a variety of human actions and responses and by various institutional arrangements for reaching and enforcing contractual agreements. Economic exchanges involving commodities are essentially impersonal; economic exchanges involving labor services have more direct and far-reaching implications for human relationships and personal welfare. These relationships have evolved through the centuries, and man’s explanation of the determinants of economic rewards has evolved with them.

In preindustrial society the role and position of the laborer were commonly prescribed by custom or law—whether the economic relationship was one of master and slave, lord and serf, or craftsman and apprentice. With few exceptions, economic units were small, and despite the rigid pattern of social structure the master-employer had certain obligations and responsibilities toward those dependent upon him. With the rise of modern industrial society, the size of representative economic units expanded markedly, and work relationships became increasingly depersonalized. Largely as a result of the economies of large-scale production, new institutional forms have evolved to carry on the labor and entrepreneurial functions. The corporation and the trade union are modern social inventions representing employer and employee, and wage contracts in Western industrial societies are frequently arrived at through collective bargaining or through arbitration by third parties.

In preindustrial society wages were determined largely by custom and tradition, external factors permitting. The external constraints in the short run were the vagaries of crop successes or failures, pestilence, and war. In the long run, the margin of subsistence set a lower limit to wage rates ( or their premarket equivalent), and the available food supply set an upper limit. In most preindustrial societies in the past, as is true in many parts of Asia, Africa, and Latin America today, population growth tended to keep pace with or outstrip the expansion of food supplies, and wages for the great bulk of the population remained close to the minimum subsistence level.

In the early years of commercial and industrial societies, market wage systems developed, and work relationships became increasingly impersonal. A more legalistic contractual attitude toward employment was associated with the gradual withdrawal of a sense of obligation by employer, church, or state for the welfare of the individual worker and the resulting creation of a highly (although often imperfectly) competitive labor market. Coincident with this growth of a more market-oriented economy came corresponding social and political freedoms that had not been present in the earlier tradition-bound societies, and the rate of economic growth began to exceed population growth. For those who quickly benefited from the rapid growth of economic opportunities, such as skilled workers, shopkeepers, entrepreneurs, and professional men, this meant great social progress; for those who remained among the less skilled and were entirely dependent upon market forces over which they had little control, the appeal of alternative systems that would restrict or destroy the free market economy was frequently enhanced.

If by outward appearances the nineteenth century was characterized by impersonal market forces in the determination of wages and prices, the twentieth century has appeared to be a century of regulated or administered prices. Wage settlements are more frequently made by agreement between collective agencies, sometimes under the duress of public or government pressure, and occasionally within the framework of governmental regulation of wages. To the casual observer it may seem that wages are no longer determined by market forces, but by men; that money wages are no longer the result of the interplay of many economic variables, but the initial datum to which the economy must adapt; and that political forces have replaced economic forces in wage determination. A review of wage theory, however, suggests that, despite these outward appearances, market forces still determine the level of real wages (i.e., wages measured in terms of purchasing power) for employed members of the labor force. The same economic forces are present in a free market economy, a government-regulated economy, or an economy with complete state ownership of property and capital. Arbitrarily determined moneywage rates cannot determine realwages or lead to the most efficient levels of employment in various lines of activity, and the principles of contemporary wage theory are applicable to a centralized economy seeking national guides for resource allocation as well as to a decentralized market economy.

Historically, there have been four broad schools of thought concerning the determination of wages. These are usually referred to as the subsistence theoryof the late eighteenth and early nineteenth centuries, the wages-fund theoryof the mid-nineteenth century, the marginal productivity theoryof the very late nineteenth and the twentieth centuries, and the bargaining power theoryof the twentieth century. Each has emphasized a different aspect of wage determination, although the contemporary marginal productivity theory might be said to include each of the others. Broadly interpreted, none of the other three is incorrect—each is merely incomplete.

Classical wage theory. The subsistence theory of wages is attributed largely to the French physio-cratic school, particularly to Turgot and Quesnay. Later variations of it are found in Say, Adam Smith, Malthus, Ricardo, Torrens, and Marx, among others. Essentially this was a labor supply theory, assuming certain “natural” tendencies in population growth and stressing the replacement cost of the labor force. As most clearly articulated by Malthus and Ricardo, the theory held that population tends to grow more rapidly then the supply of food, keeping the bulk of the population at or near subsistence. The subsistence level of wages, termed the “natural” wage by Ricardo, effectively sets the long-run limit on population. The market wage, if it rises above the natural level, encourages larger families (or a higher survival rate). This gradually expands labor supply and depresses the market wage rate. If the market rate were below the subsistence level, families would delay having children (or fewer would survive to adulthood), and the market wage would again tend to seek the natural level. This theory, termed the “iron law of wages,” was a pessimistic view of future economic development but not an entirely unrealistic one as judged by the experience of previous centuries. The standard of living of the common laborer had improved temporarily after the great plague swept Europe in the fourteenth century but had again declined to near subsistence and remained there until the nineteenth century.

The later adherents to the subsistence-wage doctrine—Torrens, Ricardo, and Marx—allowed for some variation in the “customary” minimum standards of living. Thus the minimum level in Europe in the first half of the nineteenth century was, in most countries and in most years, well above the bare subsistence level. By the third quarter of the century, when J. S. Mill was writing, it was evident that living standards in England, France, and the United States were not only considerably above subsistence but steadily rising, thus holding out a more hopeful picture of the future.

Among the major political economists in the classical tradition, Marx alone predicted increasing misery for the working class. This view was predicated on the observed gradual disappearance of the yeoman class, as the machine displaced the master craftsman, and on the presumed enlargement of the working class living at the subsistence margin. To Marx this was a distinguishing feature of industrial capitalism, the contrast being the more marked because it was accompanied by a rapid accumulation of capital by the entrepreneurial class. Thus, the industrial society automatically produced a “reserve army of the unemployed” through technological displacement. Marx did not live long enough to see clearly the creation of a new and expanding middle class or to observe the steadily rising living standards enjoyed by the laboring classes under modern industrial conditions. He was a keener critic of contemporary social conditions than he was a prophet.

The wages-fund theory of the mid-nineteenth century was a logical successor to the subsistence theory, for it assumed the same economic properties except for the inevitable adjustment of population to economic conditions. Whereas the earlier writers had focused their attention primarily on the supply side of the market, the later writers tried to explain the determinants of labor demand. As expressed by J. S. Mill, the theory held that at any one time there was a fixed fund from which wages could be advanced to labor in anticipation of the production and sale of marketable goods. This stock of capital, referred to as a “wages fund,” might grow over time as a nation accumulated wealth through savings and reinvestment. At any moment, however, the fund must be divided by the number of laborers; the larger the labor force, the lower the wage. Attempts by workers to raise their wages could only be successful at the cost of unemployment or the reduction of the wages of other workers. Although Mill held out more hope than had Malthus or Ricardo that rising aspirations would retard population growth and thus help to raise wage levels, he placed primary emphasis upon the accumulation of wealth as a means of enlarging the wages fund and raising wage levels.

The weakness of the early classical wage theories lay in assuming a relatively fixed demand for labor, fixed either by available food supplies or by a wages fund. Particularly in the latter case, the present stockof capital was emphasized, rather than the flowof commodities that could be produced with available resources, including labor. Both theories are more nearly applicable to a pre-industrial society without the means of credit creation; in such societies they have some applicability even today.

Marginal productivity theory. The marginal productivity theory of wages was independently developed by a number of economists near the close of the nineteenth century. Although von Thiinen in Germany was an early precursor of this school, Wicksell in Sweden, Walras in France, Jevons, Wicksteed, and Marshall in England, Barone in Italy, and J. B. Clark in the United States all contributed to the rapid emergence of the theory between 1880 and 1900.

Strictly speaking, marginal productivity is not a theoryof wages, but a principle concerning the nature of the demand for factors of production. It is a special application of the law of diminishing returns, expressing the direct functional relationship between the quantity of a factor employed and its product. Thus, in a simple two-factor model, if the quantity of capital is held constant and increments of labor are added, the total product will increase at a decreasing rate. The application of this principle to the determination of wages is that if an employer is acting so as to maximize his net returns, he will continue to employ more labor as long as the marginal product of labor exceeds the wage cost of the added laborers. Equilibrium for the individual firm occurs where the wage is just equal to labor’s marginal product, and similarly for other factors of production.

J. B. Clark (1899) constructed the most elaborate theory around the marginal productivity principle. Assuming a fixed (inelastic) supply of labor (i.e., constant population, level of skills, etc.), conditions of perfect competition, and long-run general equilibrium in a static state, he indicated that marginal productivity would determine a unique level of real wages. Each factor of production would receive a return just equal to its marginal product, and the sum of factor returns would just exhaust the total product. The latter condition is satisfied if the production function is a homogeneous function of the first degree (Wicksteed 1894), or if each firm is producing at minimum unit cost under conditions of competitive equilibrium (Walras 1874-1877; Wicksell 1893; Hicks 1932, appendix).

Clark’s theory was expressed in real terms for the economy as a whole; for the individual firm under perfect competition, equilibrium would be achieved at that level of employment where the wage was equal to the value of the marginal product. The theory of pricing under conditions of imperfect competition, which developed in the 1930s, states that the equilibrium rule for a single employer is to equate the marginal wage cost (which under conditions of monopsony may be greater than the wage paid to the marginal laborer) with labor’s marginal revenue product (E. H. Chamber-lin 1933; Joan Robinson 1933).

The marginal productivity theory as a complete and determinate theory of wages in the real world is subject to many criticisms. First, the supply of labor is not absolutely fixed even in the short run, although some debate continues as to whether the supply curve is positively sloped or backward bending. In the long run the supply of labor, as the early classical writers maintained, is at least partly dependent upon the level of real wages. Similarly, the supply of other factors of production—most notably capital—is not entirely independent of the level of wages, for changes in wage rates affect the level of investment both through their impact on profit rates and on the level of savings out of wage earners’ incomes.

At the level of the individual firm, difficulties arise in applying the theory because of imperfections in knowledge on the part of both workers and employers, indivisibilities due to technical factors, immobility, and marked imperfections in competition both in product and labor markets. The marginal productivity principle is still a valuable analytic tool, but a large number of variations must be developed to handle cases with different combinations of supply and demand conditions under varying degrees of competition (Dunlop 1944).

The analysis of the short-run demand for labor by the firm has grown more complicated as business and industry have become more complex. Technical factors increasingly dictate inflexible factor proportions in the short run; the influence of relative factor prices appears more in the design of new plants and in the character of innovation than it does in short-run adjustments of employment. Similarly, the administrative organization of the production process frequently protects certain jobs or groups of employees from the market through internal promotion, seniority rights, the accumulation of benefits, etc. Just as the military ordinarily “hires” only privates and lieutenants and may depart substantially from market norms in the pay of sergeants and colonels, so too the large modern corporation is partially veiled from market forces. Increasing attention in recent years has been paid to “wage structure” and “job clusters,” because investigators recognize that in an age of increasing differentiation of labor skills it is difficult to give empirical content to the concept of thewage rate for a homogeneous labor unit (International Economic Association 1957; Reynolds 1951).

It has often been said that the marginal productivity concept provides the framework for a theory of wages for the economy as a whole and for a theory of employment for the individual firm; that is to say, the aggregate supply of labor is relatively fixed (at least for periods of less than a generation), and marginal productivity therefore “determines” the level of real wages for the economy as a whole. The individual firm, however, faces a much more elastic labor-supply situation, and within a fairly narrow range, wage rates are “determined” by certain external constraints—e.g., by the prevailing wage in the local labor market, by what other firms in the same industry are currently paying, etc. For the firm, therefore, variations in the demand for labor are more likely to result in changes in employment than in major wage adjustments.

Recent developments. At the level of the firm and the industry, much greater attention has been paid in recent years to the determinants of supply in the presence of a trade union. The early literature treated this as a classic case of monopsony, and the theory of bilateral monopoly pricing was applied to the situation of a strong trade union bargaining with a large employer. This assumed, however, that the union pursued a course analogous to the business firm and, as a monopolistic “seller” of labor, attempted to maximize returns over and above supply costs. Since the union does not incur the costs of supplying labor and the ordinary supply schedule reflects the reservation prices of its various individual members, this approach has been largely discarded as being an inadequate explanation of trade union behavior (Dunlop 1944; Ross 1948).

Impressed with the power of trade unions and the indeterminateness of the bilateral monopoly solution, some writers (e.g., John Davidson 1898; Maurice Dobb 1928) attempted to explain wage determination on the basis of the relative bargaining power of employers and employees. A bargaining power theory of wages is obviously an incomplete theory, for if bargaining power were the only factor, much greater variation in wage levels would be apparent. However, with the development of mathematical game theory in recent years, there has been renewed interest in bargaining models.

The Danish economist Zeuthen (1930) was perhaps the first to develop a sophisticated bargaining model applicable to economic situations. Later mathematical variants (Pen 1952; Shackle 1957; Cross 1965) and nonmathematical variants (Schelling 1956; Chamberlain 1955) have come closer to a description of economic processes in the real world. The advantage of such models is twofold : first, they illuminate the strategy of bargaining, presenting it as a many-dimensional problem; second, they provide a solution which, at least in theory, is determinate. Unlike the earlier advocates of a bargaining-power theory of wages, recent contributors attempt to develop a bargaining model within the framework of the more traditional supply and demand analysis (Pen 1950; Cartter 1959).

Traditional neoclassical wage theory is at its best in describing the determinants of the real level of wages for the economy as a whole, and it has proved useful—if somewhat less well adapted—in explaining money wages and employment behavior at the level of the firm. The weakest link has been the aggregative theory of money wages. Prior to Keynes, it was assumed that the real wage solution for the economy as a whole could be simply converted into money terms. Keynes questioned the symmetry of this conversion, stressing a “money illusion” on the part of wage earners and the possibility of an underemployment equilibrium. Two general approaches have been made to the problem within the Keynesian framework. One is to analyze the equilibrium level of real output, attained by the intersection of aggregate demand and supply functions as the money wage level shifts, with given assumptions about the labor-productivity function. In such a model only slight variations in the parameters can produce an aggregate money-supply function for labor which is positively sloped (the classical assumption), negatively sloped (the underconsumptionist assumption), or nearly vertical (the Keynesian assumption). (See Weintraub 1958.)

The second approach is through an aggregate distribution model that attempts to view the interrelationship of the share of income going to labor and the determination of an equilibrium level of total income. Such models suggest that the share of total income going to labor is closely interdependent with spending and savings decisions in other sectors of the economy, profit distributions, the degree of monopoly, and related factors (Boulding 1951; Kalecki 1954; Bronfenbrenner 1956; Cartter 1959; Kaldor 1956). Thus, the over-all share of income going to labor is not uniquely determined by wage decisions of employers and employees, for these decisions may or may not be compatible with other sets of decisions made by consumers and investors. The wage share is represented in some contemporary growth models as a dependent variable, and attempts by trade unions to fix this share artificially through wage agreements appear as futile as attempts to set the real wage in the static Keynesian model.

Empirical studies. Empirical studies of the effect of unionism on wages have been made for several countries, although the subject has been most fully treated in the United States. These studies fall into two classes: those focusing on the effect of unions upon relative wage rates and those concentrating on the aggregate wage share. The former attempt to assess the impact of the union within the firm or industry, either by examining the historical path of wages in organized and unorganized sectors or by a cross-sectional view at a single point in time. It is difficult to obtain conclusive evidence from which one can generalize, since the union impact apparently differs depending upon many other factors—market structure in both labor and product markets, centralization of authority within the union, the phase of the business cycle, etc. The most defensible conclusions, appropriate to advanced industrial countries, are that some (although not all) unions do obtain a relative wage advantage for their members; that wage advantages are usually obtained early in the life of the union and that union wages do not cumulatively diverge from general wage trends; that union wage advantages are most commonly obtained during periods of recession or of price stability and are frequently eroded during periods of rapid inflation; and that the typical magnitude of the relative wage advantage of successful unions is in the range of 10-15 per cent during periods of stable prices (Ross 1948; Rees 1962; Lewis 1963).

The second approach, analyzing trends in labor’s share of aggregate income, provides less conclusive evidence of the power of organized labor. In most industrial countries, particularly since World War ii, labor’s share has drifted upward, but the increase does not seem closely correlated with the rise of union power. A disaggregated view suggests that labor’s share of income is actually declining in many heavily organized sectors of the economy and rising in the nonunion sectors. This conclusion, however, reflects other compensating changes—the impact of further mechanization and the substitution of capital for labor in heavy-industry sectors and the rapid wage gains made by unorganized lesser-skilled workers in trade and services (which are normally associated with a prolonged period of prosperity). Unfortunately, comparable national income data are not easily available for periods prior to the 1930s, and the evolution of corporate business and other structural changes in the economy obscure a clearer view (Creamer 1950; Levinson 1951; Phelps Brown 1962; Clark Kerr in Taylor & Pierson 1957).

Earlier studies by Douglas (1934) for the United States and New South Wales applied a production function that gave distributional results with a close correspondence to historical data for periods between 1890 and 1922. The apparent constancy of labor’s share at close to 75 per cent for the United States was consistent with a long-term elasticity of labor demand of - 4.0 and a demand elasticity of - 1.33 for capital. Later studies have stressed the great variability of labor’s share in particular employment sectors, both cyclically and over long periods of time. Thirty years ago it appeared that the “laws of production” explained the functional distribution of income. Today it might be more correct to say that the over-all share of income going to labor is determined not in the labor market alone but also by the general determinants of the national income.

The focus of interest in the labor market has changed several times over the last fifty years. Wage theory in the first quarter of the century was essentially normative, dealing with perfectly competitive models of the world. In the 1920s and 1930s interest shifted to the structure and role of trade unions and to the institutional aspects of the labor market. Then attention was increasingly focused on the determination of the level of wages under a variety of degrees of market imperfection at the level of the firm, industry, or nation. Recent interest has shifted to the dynamics of the labor market—to the role of wages in aggregate growth models and to the play of veiled market forces in bargaining models. Current wage theory stresses relative wage rates and their movement over time. The interest of economists in the process of economic development—and in the problems of economies at all stages of development—has helped to make wage theory somewhat less parochial and limited in its application only to Western industrial societies than was formerly the case. It has, however, redirected emphasis away from the firm and to an aggregative view of economic variables.

Allan M. Cartter

[See alsoIncome distribution, article onfunctional share;Labor unions, article oninfluence on wages.]

BIBLIOGRAPHY

Boulding, Kenneth E. (1951) 1956 Wages as a Share in the National Income. Pages 123-148 in Institute on the Structure of the Labor Market, American University, Washington, D.C., 1950, The Impact of the Union: Eight Economic Theorists Evaluate the Labor Union Movement. Edited by David McCord Wright. New York: Kelley & Millman.

Bowen, William G. 1960 The Wage-Price Issue: A Theoretical Analysis.Princeton Univ. Press.

Bronfenbrenner, Martin S. 1956 A Contribution to the Aggregative Theory of Wages. Journal of Political Economy64:459-469.

Brown, E. H. Phelps 1962 The Economics of Labor.New Haven: Yale Univ. Press.

Cartter, Allan M. 1959 Theory of Wages and Employment.Homewood, 111.: Irwin.

Chamberlain, Neil W. 1955 A General Theory of Economic Process.New York: Wiley.

Chamberlin, Edward H. (1933) 1956 The Theory of Monopolistic Competition: A Re-orientation of the Theory of Value.7th ed. Harvard Economic Studies, Vol. 38. Cambridge, Mass.: Harvard Univ. Press.

Clark, John Bates (1899) 1902 The Distribution of Wealth: A Theory of Wages, Interest and Profits.New York and London: Macmillan.

Correa, Hector 1963 The Economics of Human Resources.Amsterdam: North Holland Publishing.

Creamer, Daniel B. 1950 Behavior of Wage Rates During Business Cycles.Occasional Paper No. 34. New York: National Bureau of Economic Research.

Cross, John G. 1965 A Theory of the Bargaining Process. American EconomicReview 55:67-94.

Davidson, John 1898 The Bargain Theory of Wages: A Critical Development From the Historic Theories.New York: Putnam.

Dobb, Maurice (1928) 1948 Wages.3d ed., rev. New York: Pitman.

Douglas, Paul H. (1934)1957 The Theory of Wages.New York: Kelley.

Dunlop, John T. (1944) 1950 Wage Determination Under Trade Unions.New York: Kelley.

Guillebaud, Claude W. 1960 Wage Determination and Wages Policy.Welwyn (England): Nisbet.

Hicks, J. R. (1932) 1964 The Theory of Wages.2d ed. New York: St. Martins.

International Economic Association 1957 The Theory of Wage Determination: Proceedings.Edited by John T. Dunlop. London: Macmillan.

Kaldor, Nicholas 1956 Alternative Theories of Distribution. Review of Economic Studies23, no. 2:83-100.

Kalecki, Michael (1954) 1965 Theory of Economic Dynamics: An Essay on Cyclical and Long-run Changes in Capitalist Economy.2d ed., rev. London: Allen & Unwin.

Kerr, Clark 1954 Trade-unionism and Distributive Shares. American Economic Review44:279-316.

Levinson, Harold M. 1951 Unionism, Wage Trends and Income Distribution, 1914-1947.Ann Arbor: Univ. of Michigan, School of Business Administration, Bureau of Business Research.

Lewis, H. Gregg 1963 Unionism and Relative Wages in the United States: An Empirical Inquiry.Univ. of Chicago Press.

Pen, Jan (1950) 1959 The Wage Rate Under Collective Bargaining. Cambridge, Mass.: Harvard Univ. Press. → First published as De loonvorming in de moderne volkshuishouding.

Pen, Jan 1952 A General Theory of Bargaining. American Economic Review42:24-42.

Perlman, Richard (editor) 1964 Wage Determination: Market or Power Forces?Boston: Heath.

Reder, Melvin W. 1952 The Theory of Union Wage Policy. Review of Economics and Statistics34:34-55.

Rees, Albert 1962 The Economics of Trade Unions.Univ. of Chicago Press.

Reynolds, Lloyd G. 1951 The Structure of Labor Markets: Wages and Labor Mobility in Theory and Practice.New York: Harper.

Reynolds, Lloyd G.; and Taft, Cynthia H. 1956 The Evolution of Wage Structure.Yale Studies in Economics, Vol. 6. New Haven: Yale Univ. Press.

Robertson, D. J. 1961 The Economics of Wages and the Distribution of Income.London: Macmillan.

Robinson, Joan 1933 The Economics of Imperfect Competition.London: Macmillan.

Ross, Arthur M. (1948) 1956 Trade Union Wage Policy.Berkeley: Univ. of California Press.

Rothschild, Kurt W. (1954) 1956 The Theory of Wages.Oxford: Blackwell.

Schelling, Thomas C. 1956 An Essay on Bargaining. American Economic Review46:281-306.

Schultz, Henry 1929 Marginal Productivity and the General Pricing Process. Journal of Political Economy37:505-551.

Shackle, G. L. S. 1957 The Nature of the Bargaining Process. Pages 292-314 in International Economic Association, The Theory of Wage Determination: Proceedings of a Conference Held by the International Economic Association.Edited by John T. Dunlop. London: Macmillan; New York: St. Martins.

Stigler, George J. 1941 Production and Distribution Theories: The Formative Period.New York: Macmillan.

Taylor, George W.; and Pierson, Frank C. (editors) 1957 New Concepts in Wage Determination.New York: McGraw-Hill.

Universities-National Bureau Committee for Economic Research, Conference, Princeton, N.J., 1960 1962 Aspects of Labor Economics.Princeton Univ. Press.

Walras, LÉon (1874-1877) 1954 Elements of Pure Economics: Or, the Theory of Social Wealth.Translated by William Jaffé. Homewood, 111.: Irwin; London: Allen & Unwin. → First published as Éléments d’économie politique pure.

Weintraub, Sidney 1958 An Approach to the Theory of Income Distribution.Philadelphia: Chilton.

Weintraub, Sidney 1963 Some Aspects of Wage Theory and Policy.Philadelphia: Chilton.

Wicksell, Knut (1893)1954 Value, Capital and Rent.New York: Rinehart. → First published as Über Wert, Kapital und Rente nach den neueren nationalökono-mischen Theorien.

Wicksteed, Philip Henry (1894) 1932 An Essay on the Co-ordination of the Laws of Distribution.Series of Reprints of Scarce Works on Political Economy, No. 12. London School of Economics and Political Science.

Zeuthen, Frederik 1930 Problems of Monopoly and Economic Warfare.London: Routledge.

II. STRUCTURE

Wage structure is a generic term for the set of interrelations among a collection of wage rates. It is frequently used in contrast to the concept of wage level, which refers to an average of the collection. Thus we may speak of changes in the level of wages in the United States, referring to movements in, say, the median hourly wage paid to all wage earners, as contrasted with changes in one or more wage structures, referring to movements of one wage rate relative to others. Correlative with the concept of wage structure is that of wage differential—some measure of the difference between elements of a wage structure. In short, a wage structure is the relation among two or more wages related to one another in some systematic way.

Obviously, there are many wage structures that might be of interest: within a given country or region there is an occupational structure, an industrial structure, a structure by size of firms, etc. Among a set of countries or regions, wage structures are also found: for example, the average hourly earnings of factory workers in the states of the United States form a geographical wage structure. In this article only a few of the more important types of wage structure can be considered.

The idea of a wage structure, although not the name, is as old as the wage concept itself; it is almost as old as intellectual concern with economic problems. However, few theoretical tools have been specifically designed to analyze wage structures. One outstanding exception is the concept of non-competing groups,associated with the names of J. S. Mill and J. E. Cairns, which pertains primarily to occupational wage differentials. Another is the distinction between equalizing and nonequalizing wage differentials, expounded by Adam Smith.

From one point of view, a wage structure is simply a kind of price structure, as are the structure of interest rates and the structure of commodity prices among, say, finished goods, semifinished goods, and raw materials. Economists are concerned with the behavior of these price structures for a variety of reasons. A change in the price of one raw material or one type of labor vis-a-vis another may alter the relative rates of production of different commodities or the terms and direction of international trade; it may also change the distribution of income among the sellers of the different commodities or services whose prices constitute the structure in question.

In the following discussion, we shall mean by “wages” the average straight-time hourly earnings (during a given week) of a worker, unless the contrary is explicitly stated. Hourly earnings are not always the wage concept best suited to a particular analysis of wage structure. However, they are the most commonly used measure because of their availability. For the purpose of comparing labor costs to employers at different locations, the data wanted is average hourly employee compensation including fringe benefits.To study the inter-temporal behavior of wages viewed as the “price of labor,” one needs average straight-time hourly compensation (including fringe benefits) so as to abstract from the effect of varying amounts of overtime (usually paid at premium rates).

Discussions of the economic welfare of different occupational groups usually refer to earnings over a lifetime rather than in a “typical” hour; for this purpose the differential impact of unemployment and length of working life are very important. Refinement of existing measures of wage structure would greatly improve present-day analysis, but this must await the slow accumulation of better statistical information.

Theoretical considerations

Although few economists would deny that a wage structure is one species of price structure, many contend that wage differentials are not subject to the laws of the market (sometimes called the laws of supply and demand) to the same degree as others or are not subject to them at all. To treat a wage structure as subject to these laws is to assume (1) that each of the wage elements within the structure behaves as though it were determined in a (purely) competitive market and (2) that deviations of the observed wage elements from their equilibrium values either are negligible or can be attributed to lags in the process of adjustment to departures from equilibrium.

Each of these assumptions is pregnant with factual implications. The first assumption implies that labor unions and the pricing policies of firms have only a negligible influence on wage differentials; either they have no appreciable effect upon wages or they cause all wages to vary proportionately. The first assumption also implies that departures from economic rationality, or changes in the objectives of buyers and sellers of labor services, do not affect relative wages. Yet a further implication is that variations in “quality” among similar units of labor service are negligible so that the labor quantities (e.g., man hours of labor) corresponding to any element of the wage structure may be treated as different service units of the same factor of production.

The second assumption implies that if the supply of any kind of labor (the quantity corresponding to an element in the structure) exceeds the demand, its wage will fall, and vice versa. This is to say that each element of the wage structure conforms to the laws of the market. Obviously the process of adjustment to a discrepancy between demand and supply takes time; the behavior of the wage during the “adjustment period” reflects the temporary effect of forces other than those of excess supply. The length of adjustment periods and the behavior of wages within them reflect the strength of “nonmarket” forces in wage determination.

No one pretends that any of the intellectual constructions used in applying the laws of the market to wage structures is descriptive of the real world. All of them, to a greater or lesser degree, simplify reality. Whether they constitute more useful simplifications than those posited by alternative theories is the important and controversial issue.

Institutional factors. In the analysis of wages, as in other applications of price theory, some economists believe the best explanation of empirical phenomena is to be found by applying competitive price theory as a first approximation and then making allowance for the operation of “other forces,” which are assumed to be of secondary importance. Other economists feel strongly that this approach involves a misplacing of emphasis. While the latter usually concede that market forces are one of a number of factors bearing upon wage determination, they do not believe that they always merit primacy. Instead, they stress a variety of other forces, varying in importance with time and place, that might loosely be called “institutional.” Among the more important of these are (1) trade union wage policy and collective bargaining; (2) the attitudes of employers, and of the community generally, toward “wage relativities”; and (3) governmental wage policies. Let us consider each of these briefly.

Trade union policy.From time to time, unions announce specific objectives of wage policy. For example, in the late 1930s and the 1940s the Swedish unions announced a “solidaristic” wage policy designed to raise the wages of the lower-paid workers relative to those of workers (initially) better paid by adjusting the terms of collective agreements. Frequently, industrial unions in the United States ask for, say, a ten-cent per hour increase for all workers, which implies larger percentage increases for the lower-paid workers. At other times they demand equal percentage increases for all workers or even special allowances for highly paid craftsmen.

Unions sometimes insist that incentive plans embody certain characteristics but in other cases resist them altogether; either position may affect the wage structure of workers employed under a union contract. The desire of each union to do as well as other unions having members in adjacent areas or doing similar work has been much emphasized by students of wages. This desire has allegedly led to “wage patterns” that exert great pressure on participants in collective bargaining to imitate the wage behavior of other bargainers within the same “orbit of coercive comparison” or on the same “wage contour.”

A further consequence sometimes attributed to union wage policy is the reduction of “personal differentials” —differences in wages paid to workers apparently doing the same job. Personal differentials sometimes reflect differential effort, as employers contend, and sometimes favoritism, as unions allege. While unionism cannot eliminate such differentials, it can make it harder for the employer to distribute them arbitrarily.

Employer and community attitudes.It has been observed in many countries that some employers pay more than others for the same type of work. Although the higher-paying firms tend to get the cream of the labor supply, it is often contended that this is not the sole—or even the principal—reason for a “high wage” policy. The same writers deny that achievement of better morale and higher productivity, which are associated with high wages, is the primary motivation. Rather, they argue, the social conscience of those employers who are able to pay more than the market requires frequently drives them to do so as a contribution to the well-being of their employees and the stability of the community. Sometimes the spur of conscience alone generates a high wage policy; in other cases there may be considerable social and even governmental pressure. But whatever the reason, differential sensitivity to the demands of social conscience and differential ability to respond to such demands lead to wage differentials among firms that cannot be explained entirely by the laws of the market.

Governmental policies.Governmental wage policies also exert a significant influence on a nation’s wage structure through a number of channels: minimum wage laws, family allowances, national wage agreements imposed by governmental pressure, the wage policies of public enterprises, social security laws, etc.

The role of policy—union, employer, and governmental—in wage determination is so obvious and so prevalent that it is hard for many observers, particularly those not trained in economics, to believe that wage structure can reflect anything else. This is especially so because wage policies are often the subject of heated public controversy. Market forces, on the other hand, work anonymously and even pseudonymously, and are therefore easily overlooked. No competent economist would deny that wage policies are influenced by market pressures. However, those who emphasize the importance of the institutional arrangements and attitudes tend to regard market forces as only one among many relevant influences and to portray the basic characteristics of a wage structure as being more or less independent of considerations of cost minimization.

To exemplify, although exaggeratedly, the opposing views, consider an administered salary schedule such as the set of salaries for the various positions in a civil service. An extreme “institution-alist” might seek to explain the interrelations among the various salary levels solely in terms of the decisions of civil service administrators. On the other hand, an extreme “price theorist” would contend that the administrators’ decisions affected only the relation of the salary scale to the various job titles. If the salary attached to a job title was lower than what the market required to obtain a person with the requisite skills, the job would tend either to go unfilled or to be improperly performed by the inadequate personnel obtainable at the civil service scale. The job’s duties, although not necessarily its title, would tend to move “up the scale” to a level where the salary was adequate to attract competent personnel. The reverse process would occur, although even more slowly, if a job title were “overpriced” relative to the market. Ultimately the salaries paid to given individuals will correspond to the market’s evaluation of their capacities; all that the decisions of the civil service administration can accomplish is to determine their job titles or ranks.

But although the long-run tendency of market forces is in this direction, few students contend that these forces completely obliterate the effect of institutional predilections, especially those reflected in governmental policy. If a government desires wage structures with certain characteristics and persistently bends its will to achieve them, it can thwart market forces indefinitely. (If it succeeds, there will be further predictable consequences, but that is another matter.)

In short, despite continuing debate over the importance of market and institutional forces in wage determination, the disagreement is primarily one of emphasis. Most students agree that institutional factors are at their best in explaining short-run variations in wage structure, while market forces become increasingly dominant in explaining long-term trends or large interregional or international differences.

Occupational wage structure

It is convenient to treat the occupational wage structure of a given place as the “fundamental” structure and build upon it in explaining the other structures—industrial, racial, etc. However, this is not to deny that the various wage structures are interdependent.

Equalizing differentials. It is apparent to the most casual observer that individuals in certain occupations typically earn more than those in others, whether earnings are measured per hour, per year, or per lifetime. One cause of such differences is that some jobs are less pleasant than others—that is, the work is more fatiguing or more hazardous, is located in unpleasant physical surroundings or an unusually inaccessible area, or has low prestige in the community.

To fill occupations or jobs that are generally disfavored in the community tends to require extra compensation relative to others involving the same “degree” of skill or training. This is sometimes expressed by saying that the net advantages of jobs requiring the same degree of skill and training tend to be equal. It follows that relatively unpleasant jobs require a compensatingor equalizing differentialin wages to balance the nonpecuniary dissatisfactions and equalize the net advantages. Examples of equalizing wage differentials are the premiums frequently paid for working night shifts (shift differentials) and for work in unpleasant places and the common practice in job evaluation schemes of awarding points that raise wages for the unpleasantness or danger of a job.

If occupational wage structures primarily reflected equalizing differentials, we would expect the most pleasant jobs to be the lowest paid. Notoriously this has not been the case. The highest-paid jobs (executive positions, the professions) also have the best working conditions and the most prestige. This is partly because persons qualifying for such jobs prefer to have part of their differential earnings paid as on-the-job comforts. Also, on-the-job amenities often appreciably enhance productivity and hence are desired by employers. In short, the differential reward of superior occupations is nonequalizing.

Noncompeting groups and training costs. Why then do these differential rewards persist? Why doesn’t competition eliminate them? The usual answer is that “workers” fall into noncompeting groups.The labor market for, say, nuclear physicists is quite distinct from that for janitors; no matter how high the wage of the physicist, his employer cannot substitute a janitor to reduce labor costs. This is common sense, but the economics that underlie it are not always well understood. Since the highest-paid occupations are generally filled by highly educated persons, it is often asserted that doctors and lawyers “deserve” high incomes because of the time and expense incurred in their training. However, whatever their deserts, generations of highly educated but impecunious students of the classics, dead languages, etc., bear witness that education of itself is no guarantee of material prosperity. In order for an occupation to yield a high income to its typical practitioner, there must be an effective demand for its practice.

The costs of training, in time, trouble, and money, do affect the income from an occupation, but only indirectly by acting as a check to the supply of practitioners. If the net advantages, including earnings, of any occupation fall below those of others requiring the same degree of training and natural ability, it will cease to attract its current share of new recruits. This will lead to a relative decline in the number of its practitioners that will persist until net advantages equalize or the occupation becomes extinct. This tendency exercises a powerful restraint upon divergence among average earnings in different but competitive occupations. However, it is a slow-acting restraint that permits sharp fluctuations in relative earnings for substantial periods of time; its full power can be observed only over fairly long periods, say, 15-25 years. During periods of “temporary” disturbance, wage earners in the unfavored occupations tend to suffer unemployment as well as relatively declining hourly wages.

The preceding remarks suggest that the well-paid occupations require relatively much training and education. Conceptually, it is possible to assign a money value to the various cost elements of training for a “typical” practitioner of an occupation (e.g., tuition, cost of books, foregone earnings). This value may be termed the investment necessary for producing a typical member of an occupation, that is, his long-run supply price. From this point of view, nonequalizing differences in earnings among occupations reflect unequal returns to given amounts of investment in training. This approach has recently attracted much interest among economists, but there has not yet been sufficient empirical research to appraise its effectiveness in explaining occupational wage differences.

But no matter how well differential investment explains (nonequalizing) differences in occupational earnings, it should not be forgotten that the capital market for investment in human capacities is extremely “imperfect.” It is very much easier for some prospective trainees for a (relatively) “capital intensive” occupation to obtain the necessary funds, through loans or gifts, than for others. The favored individuals are those whose families or friends can either provide the capital themselves or induce others to do so. The importance of personal connections has tended to decline in Western countries as government and institutionally supported scholarships and loans have become more readily available. However, these are still largely confined to the more able students. Among the mediocre, the children of well-to-do parents retain a substantial advantage in seeking training. Moreover, the children of well-educated parents are generally more successful in academic pursuits, which gives them a great advantage in “open” competition for scholarships and other sources of assistance. The children of relatively uneducated persons labor under a substantial handicap in academic competition with the offspring of those better schooled; that is, they are less desirable “sites” for investment.

The relative level of earnings in an occupation is normally independent of the state of relative demand in the long run.However, even in the long run it is possible for the level of demand to affect relative occupational earnings, if taste for engaging in a particular occupation or native capacity for doing so is distributed very unevenly among the population so that the long-run supply price for the occupation increases with the relative number of practitioners. For example, the relative earnings of concert violinists of virtuoso quality might be an increasing function of the demand (at given prices) for live performances. But, except for unusual and narrowly defined occupations, this is a factor of secondary importance in the long run, although not in the short. During comparatively short time periods within which prior commitments inhibit interoccupational movement, net advantages of particular occupations vary sharply with the state of current demand. In these short periods, cost of training plays but a minor role in affecting the interoccupational wage structure.

Union and governmental policies. The above remarks follow the tradition of neoclassical price theory in stressing market forces as explanatory factors for occupational wage structures. However, relative occupational earnings reflect more than the laws of supply and demand. Union behavior is often considered important in determining relative occupational net advantages. Occupations in which there are strong unions may enjoy high wages because the unions set wages above “competitive levels” and protect them by restricting labor supply. This can—and does—happen. However, the ability of a union to control labor supply and maintain Wages above the competitive level is strongly conditioned by the state of the labor market itself. It is debatable whether the independent impact of unions (the effect separate from the correlation of unionism with favorable market conditions) upon occupational wage differences is a majordeterminant of occupational wage structure.

To some extent, a similar comment applies to the ability of governments to affect relative occupational wages. The ability of governments to maintain their control of wage rates, and the conditions both of labor supply and demand, gives them much greater power over wage structures than any union. However, it is not always possible to separate the impact of governmental and union policies. In some countries collective agreements negotiated by unions are “extended” by law to apply to all firms within an industry, unionized or not. But in most cases it is the power of the government rather than the bargaining strength of the union that is the major source of resistance to market forces.

Skill differentials. Most statistics pertaining to occupational wage differentials refer to “skill differentials”—the difference between the straight-time hourly wages paid to common manual labor and those paid to skilled manual workers, such as electricians and machinists. The definition of “common labor” must be culturally specific; what is a common “skill” in an advanced society (e.g., ability to drive a small truck) may be a special skill in a more primitive one. The concept of skill and its measurement are not in a satisfactory state, but the theoretical problems involved are too intricate for a brief discussion.

If we select two occupations to mirror skill differentials, they will exhibit a kind of “life cycle.” At the “birth” of the special skill (when it first comes into demand) it will be scarce and its prac-tioners correspondingly highly paid relative both to common labor and to other skilled jobs. These high rewards will attract trainees who will eventually bring down the earnings to the general level of skilled workers. In time, technical progress will make the skill obsolete, and its earnings will fall below the general level attained by comparably skilled workers, so that the number of practitioners will tend to decline.

Secular trends.The hypothesis of a life cycle in specific skill differentials provides one possible explanation for a secular movement in skill differentials generally. Suppose that communities in early stages of economic growth tend to have many occupations in the early phases of these life cycles, and conversely for advanced economies. Then, as an economy ages, relative weight would shift from “young” to “mature” skills, reducing the average of all skill differentials. However, it is as yet unknown whether there is any relation between the stage of an economy’s development and the life cycle phase of its “typical” skill differential.

A different explanation of secularly declining skill margins stresses the trend toward equality of opportunity for occupational training and general education in advanced countries. A century ago skilled workers were markedly different from unskilled workers in training, in literacy, and sometimes in race or ethnic origin. Abilities and opportunities for “picking up” skills on the job or through after-work training were more limited than now, and consequently the possibility of substituting less for more skilled workers was less than at present. The situation in contemporary underdeveloped countries suggests that in labor force structure “ontogeny recapitulates phylogeny.”

The trend toward greater similarity of education, at least among manual workers, is not the only factor working to reduce the margin for skill. The limitations or prohibition on the use of child labor and of manual labor by females have also reduced the supply of relatively cheap substitutes for adult males in manual occupations. This has tended to eliminate from the labor force the least productive and poorest-paid group of unskilled workers, thereby increasing the wages of their close competitors (unskilled adult males) relative to those of the more skilled. In the United States the restriction of immigration in the early 1920s also reduced the supply of unskilled labor relative to skilled.

Minimum-wage laws have also tended to narrow skill margins secularly within the sector of the economy to which they apply because they apply mainly to unskilled, low-wage jobs. Their differential effect may sometimes be offset by the pressure of unions, which is generally more effective in the skilled trades; however, this is debatable. It must be noted that the scope of legislation establishing minimum wages, maximum hours, and the like is usually confined to the “industrial” sector of an economy and to employers who hire more than a prescribed number of employees. Such laws usually do not apply to agriculture or small business— especially retail trade.

The tendency for skill margins to contract secularly is probably most marked in those sectors of the economy where minimum-wage and related labor standards are effectively enforced. This is because in retail trade and service, agriculture, etc., especially where self-employed and family workers predominate, there are opportunities for substituting low-wage females and juveniles that cannot (because of legislation or social custom) be utilized elsewhere. Available evidence is certainly not inconsistent with this statement. However, the paucity and poor quality of the data make it imprudent to go beyond saying that the secular tendency for skill margins to decline is less clearly observed in that part of the economy excluded from effective coverage of minimum-wage regulation than elsewhere.

The long-run decline in the skill margin in advanced countries has not occurred slowly and steadily. Instead, the skill margin appears to have remained constant for relatively long periods of time and then to have declined sharply within a very few years. These sharp declines have occurred mainly during periods of extreme labor shortage, especially during World War I and World War n. During major depressions, when there was much unemployment, widening of skill margins has sometimes been observed, but there are notable exceptions. For example, there does not appear to have been much, if any, widening of skill margins during the great depression of the early 1930s. (See Table 1.) The observed skill margin is not responsive to minor variations in business activity.

International comparisons.Among manual workers, in the early 1960s, the margin for skill was a little greater in the United States than in most countries of western Europe and the British Commonwealth; in the United States the ratio of skilled to unskilled hourly earnings was about 4 to 3, as compared with about 5 to 4 in other developed

economies (see Tables 2a-2dfor data on some of the other developed economies). Within the United States the skill margin was highest in the South and lowest on the Pacific Coast; this partially reflects racial and city-size differentials (see Table 3 for other regional comparisons). The ranking of specific occupations would appear from scanty data to be similar among advanced countries.

In underdeveloped countries skill margins are generally wider than in the more advanced countries. This can be explained, at least roughly, on the hypothesis that as of 1960, labor supply conditions in underdeveloped economies were similar to those in now advanced countries fifty to one hundred years earlier. However, skill margins show extensive variation from one underdeveloped area to another, and occasionally appear to be quite small. Where this is the case, the data usually refer to a small sector of the economy where minimum-wage and related standards are enforced;

between the wages paid the unskilled workers within this “protected” sector and the earnings of the much larger number of comparable workers in agriculture, peddling, and odd jobs—the “disguised unemployed”—there is a huge gap.

White-collar and professional workers.The above remarks on skill differentials refer to the difference in earnings among manual workers. Differences in earnings among white-collar and professional workers are subject to similar forces, but their manifestations differ. Earnings data for these occupations are much less plentiful than for manual workers and refer mainly to the period since 1939. These data do not show a marked decline in the relative earnings of nonsalaried professionals (in the United States) from 1939 to the early 1950s. Nor has the relative advantage (in terms of annual income) associated with more education shown an unambiguous reduction since 1939. However, there are so many problems, both of fact and interpretation, in the literature relating income differences to education that it is necessary to reserve judgment on its findings.

Color and sex differentials. One aspect of occupational wage structures that has attracted much

attention is the association of occupation with other social characteristics. For example, Negroes are disproportionately represented in the low-paying and otherwise undesirable occupations. In part, this is because of their fewer years of schooling, but it also reflects discrimination against them in the labor market. Even within given occupations Negroes earn less than whites, and their relative income disadvantage appears to be greater within those occupations requiring extensive schooling. [See Discrimination, economic]

The labor market role of women is somewhat analogous to that of Negroes and other unfavored groups. Certain characteristics generally ascribed to them mark them as labor market inferiors. As in the case of Negroes, the ascription involves a combination of superstition and fact that inhibits rational use of female productive capacity. Women are clearly discriminated against as candidates for positions in which they give orders to men and sometimes in jobs where they must directly compete with men.

Situations in which women are paid less than men for doing the same job are the targets of legislation and trade union demands for “equal pay for equal work.” However, the most important reason for the relatively low earnings of women is their virtual absence from certain industries and occupations and their concentration in others. The sectors where women are concentrated typically pay relatively low wages—for example, garment and textile manufacturing, sales work in retail stores, stenographic and related clerical jobs, and elementary-school teaching.

Whether the low wages paid in these sectors reflect labor market discrimination against women within those sectorsis not clear. The fact that women are able to fill these jobs increases the supply of persons available for them. Conversely, there are many jobs in heavy industry (e.g., those requiring great physical strength) that women cannot fill. Hence, the limited range of factory jobs for which women are suited tends to lower the supply price and worsen the terms of employment

on those jobs for which they can qualify. This accounts for much, but not all, of the observed inferiority of women as wage earners.

However, it is likely that because the supply price of female labor is lower than that of male labor, employers tend to define job titles and duties so as to create special low-paid “female” jobs. Sometimes this involves unequal pay for equal work, which is labor market “discrimination” but there is no reliable information as to the relative importance of the discriminatory component of sexual wage differentials.

Size of firm differentials. Students generally agree that large firms, except when they are unprofitable, tend to pay higher wages on what seems to be the same job. This aspect of wage structure has been observed in several countries, particularly the United States and Japan (see Tables 4a and 4b). It is likely that firms paying high wages obtain superior personnel, but some writers contend that this is not the main reason why they do so. It has been suggested that labor union and/or govern-

mental pressure forces large firms to share their profits with their workers. However, this contention is plausible only where large firms are unusually profitable. It also has been suggested that large firms must pay higher wages in order to attract persons from a larger geographical area, that is, to offset the inconvenience of a longer journey to work. The interpretation of wage differentials within occupations among firms in a given industry is at present a disputed issue.

Geographical differentials. One reason why wages differ from place to place is that the occupational composition of the labor force varies. Another is that the industrial composition or sexual or racial mixes of the labor forces differ. Thus, differences in earnings per worker among nations, regions, states, etc., should not be interpreted as necessarily reflecting different prices for the same productive service. But “pure” geographical differentials probably do exist, and the following remarks refer to them.

For concreteness, define a “pure” geographical differential as a difference in average straight-time hourly compensation between two places in the same industry and occupation for a person of the same age, sex, race, and union status. A pure geographical wage differential might arise because in one country, state, or city, A, the supply of capital and other complementary factors was greater than in another, B. This would tend to make the marginal productivity, and the wage, higher in A. But, migration costs ignored, this difference could persist only if the workers in B refused to move to A or if A refused to accept them. One would not expect such a difference to persist if A and B were within the same country or within an area of free migration, except where the differential is of an equalizing nature. For example, an equalizing geographical differential (for comparable work) would arise where wages were lower in A than in B because A had a less pleasant climate.

Price theory implies that any pure, nonequalizing, geographical differentials are caused by barriers to migration or are purely temporary. Temporary differentials may arise because of different conditions of labor demand in different places. If there is a pure, nonequalizing differential between A and B, favoring A, labor will move from B to A, other things being equal. But because human migration is a slow process, the differential may persist for a long time. Thus, within the United States there have been long-standing earnings differentials in favor of the Pacific Coast and adverse to the Southern states vis-à-vis the rest of the nation. These differentials have been reduced by net migration to the Pacific Coast and away from the South. However, because of birth-rate differentials and related factors, these geographical differentials still exist. (Certainly, not all of the measured geographical differential is “pure,” but part of it probably is.)

One species of geographical differential is the city-size differential. Money wages are higher in large than in small cities, except for the very largest cities, and in small cities than on farms. In part, such money wage differences offset differences in living costs, and, to a large extent, they also reflect differences in occupational structure. However, it seems very likely that there is a pure city-size differential.

Industry differentials. Competitive price theory implies that in the long run all industries will pay the same wage to workers in a given occupation in a given location, except for those differentials necessary to equalize net advantages. This means that interindustry wage differences should reflect only differences in the occupational mix and equalizing differences on account of location. Unquestionably differences in occupation and location are a major source of existing industry differentials. In general, the high-wage industries in manufacturing require physical strength and some mechanical aptitude and hence employ relatively few women. They are typically in “heavy” industry, where firms are relatively large, and tend to be located in large cities. Thus, automobile and steel manufacturing are high-wage industries, while textiles and garment manufacturing are low-wage industries. (See Table 5.)

These statements apply to the long run. At any given date, rapidly expanding industries will tend to create temporary shortages in those occupations

and locations on whose labor supply they draw heavily. This will tend to make their wage rates higher than those paid in other industries. Consequently, “new” industries tend to pay higher wages than the “average,” other things being equal, and old ones to pay less.

This interpretation of the interindustry wage structure is disputed by economists who contend that monopoly power in the product market gives certain industries the ability to pay more for their labor and that union and governmental pressure forces these gains to be shared with employees. This view is prevalent among economists and policy makers who are concerned with the possible inflationary effects of “key bargains” between large firms and big unions.

It is likely that at any given time an industry’s current profit position and the strength of the unions with which it deals do affect the movement in its wage level relative to that in other industries. But it is not so clear that these forces can appreciably influence relative wage levels for any prolonged period. They could do so in principle, but there is little or no uncontroverted evidence that they have done so in fact.

Since 1965 there has been a great increase in knowledge of the separate effects of such variables as industry, city size, skill level, education, race, and sex on hourly wages in the United States. This is due to the availability of details on individuals obtainable from the 1-in-1,000 sample of the 1960 census. These data have already been utilized in separate studies by Fuchs (1967) and by Weiss (1966). Further studies on these and similar data are currently in progress, and within a decade there should be a great increase both in our knowledge of the details of wage structure and our understanding of their interrelationships.

Thus far, the discussion in this article has referred primarily to the industrialized sector of the economy. Agriculture, in all countries, stands at or near the bottom of the interindustry wage hierarchy. In part, this is because of the large amount of common labor in its input mix. But it is also due to the fact that as a result of relatively high birth rates and a low income elasticity of demand for foodstuffs, rural areas are net exporters of labor. Most economies are in a state of lagging adjustment to a disequilibrium between the net advantages of rural and urban employment, and agriculture is located so as to take advantage of the relatively low-wage rural labor market. Other industries, particularly food processing, that can be efficiently carried on in small towns and rural areas similarly benefit from this locational advantage.

Small towns, small firms, contact with agriculture, employment of females and youths, and impermanent employment relations are part of a low-wage syndrome. Low-wage industries tend to exhibit these characteristics more than others. Not surprisingly, such industries usually lie outside the mesh of statutory minimum-wage requirements, and their workers are less prone to be unionized than the labor force as a whole.

Melvin W. Reder

BIBLIOGRAPHY

Becker, Gary S. 1957 Economics of Discrimination.Univ. of Chicago Press.

Becker, Gary S. 1964 Human Capital: A Theoretical and Empirical Analysis With Special Reference to Education.New York: National Bureau of Economic Research.

Brown, E. H. Phelps 1962 The Economics of Labor.New Haven: Yale Univ. Press. → See especially Chapter 5.

Brown, E. H. Phelps; and Browne, M. H. 1962 Earnings in Industries of the United Kingdom: 1948-1959. Economic Journal72:517-549.

Dunlop, John T. (editor) (1957) 1959 The Theory of Wage Determination.New York: St. Martins. → See especially Part 4.

Fuchs, Victor R. 1967 Differentials in Hourly Earnings by Region and City Size, 1959.New York: National Bureau of Economic Research.

Gilman, Harry J. 1965 Economic Discrimination and Unemployment. American Economic Review55:1077-1096.

GÜnter, H. 1964 Changes in Occupational Wage Differentials. International Labour Review89:136-155.

Keat, Paul S. 1960 Long-run Changes in Occupational Wage Structure: 1900-1956. Journal of Political Economy68:584-600.

Kendrick, John W. 1961 Productivity Trends in the United States.National Bureau of Economic Research, General Series, No. 71. Princeton Univ. Press.

Kerr, Clark 1954 The Balkanization of Labor Markets. Pages 92-110 in Labor Mobility and Economic Opportunity: Essays,by E. Wight Bakke et al. New York: Wiley.

Lebergott, Stanley 1947 Wage Structures. Review of Economic Statistics29:274-285.

Lewis, W. Arthur 1954 Economic Development With Unlimited Supplies of Labor. Manchester School of Economic and Social Studies22:139-191.

Miller, Herman P. 1966 Income Distribution in the United States.1960 Census Monograph Series. Washington: Government Printing Office. → See especially Chapter 6.

Ober, Harry 1948 Occupational Wage Differentials: 1907-1947. U.S. Bureau of Labor Statistics, Monthly Labor Review67:127-134.

Occupational Wage Relationships in Metropolitan Areas: 1961-1962. 1963 U.S. Bureau of Labor Statistics, Monthly Labor Review86:1426-1431.

Reder, Melvin W. 1955 The Theory of Occupational Wage Differentials. American Economic Review45:833-852.

Reder, Melvin W. 1962 Wage Differentials: Theory and Measurement. Pages 257-317 in Universities-National Bureau Committee for Economic Research, Conference, Princeton, N.J., 1960, Aspects of Labor Economics.Princeton Univ. Press. → Contains six pages of comment by Donald E. Cullen.

Reynolds, Lloyd G.; and Taft, Cynthia H. 1956 The Evolution of Wage Structure.New Haven: Yale Univ. Press.

Schultz, Theodore W. (editor) 1962 Investment in Human Beings. Journal of Political Economy70 (Supplement), no. 5, part 2. → Papers presented at a conference called by the Universities-National Bureau Committee for Economic Research.

Taira, Koji 1966 Wage Differentials in Developing Countries: A Survey of Findings. International Labour Review93:281-301.

U.S. Bureau of Labor Statistics 1966 Employment and Earnings Statistics for United States 1909-1965. Washington: Government Printing Office.

U.S. Department of Commerce, Bureau of the Census 1960 Historical Statistics of the United States: Colonial Times to 1957.Washington: Government Printing Office.

Weiss, L. W. 1966 Concentration and Labor Earnings. American Economic Review56:96-117.

Woods, H. D.; and Ostry, Sylvia 1963 Labour Policy and Labour Economics in Canada.New York: St. Martins.

Wootton, Barbara (Adam) 1955 Social Foundations of Wage Policy: A Study of Contemporary British Wage and Salary Structure.New York: Norton.

Woytinsky, Wladimir S. 1953 Employment and Wages in the United States. New York: Twentieth Century Fund. → See especially Part 4.

III. SYSTEMS OF PAYMENT

A system of wage payment prescribes a method for both measuring achievement and allocating rewards.Existing systems of wage payment present important variations and alternatives: achievement can be defined as extra output (piecework), as cost reduction (Scanlon Plan), as profit improvement (profit sharing), or as meeting predetermined norms (measured daywork). Similarly, rewards can be allocated to individuals on a daily basis (piecework), to plant-wide groups on a monthly basis (Scanlon Plan), to company-wide groups in deferred compensation at yearly intervals (profit sharing), or to employees on a time basis without direct reference to output (measured daywork).

Four systems have been chosen for analysis— they range between those relying primarily on direct financial inducements, as in output incentives, and those using nonfinancial techniques for motivating employees, as in measured daywork. Cost reduction and profit sharing fall somewhere in between.

Output incentives. The use of output incentives dates back to the beginning of the factory system. Prices were established for component operations, and the worker was paid on the basis of his output. For example, in the early days of the steel industry, the foreman acted as a subcontractor and the piecework price served as the method of reimbursement.

In many industries, payment by output has become highly developed. These are the so-called piecework industries, such as men’s and women’s apparel, hosiery, and shoes. In these industries the grade system has developed, wherein the price of the finished product determines the value of the labor content. Piecework incentives help to standardize unit labor costs in these highly competitive industries.

Except for the piecework industries, most output incentives today are of the standard-hour variety. Production standards are established by industrial engineering analysis. Much of the work measurement is done by actual time study, although the use of predetermined and standard data is gaining in popularity.

In the United States, output incentives cover about 30 per cent of the production and maintenance work force (Lewis 1960). The extent of coverage for other countries runs higher: in England, over 40 per cent (Robertson 1960, p. 92); in much of Europe, about 50-60 per cent (Guilbert 1960); in Scandinavia about 75 per cent (Sellie 1961, p. 18); in the U.S.S.R. about 75 per cent (Campbell 1960, p. 136); in Communist China the vast majority of workers are paid on this basis (Hoffmann 1964, p. 96).

The industries using output incentives are distinguished by several characteristics (McKersie et al. 1964; Guilbert 1960).

(1) Labor costs represent a high percentage of the value added by manufacture. For example, the cigarette industry has a labor cost factor of 12 per cent and incentive coverage of only 3 per cent, while the footwear industry has a labor cost factor of 50 per cent and incentive coverage of 71 per cent.

(2) The plant is characterized by a low state of technology. For example, the hardware industry is characterized by considerable handwork and has an incentive coverage of 44 per cent, while the canning industry is highly mechanized and has an incentive coverage of only 12 per cent. In a study of the steel industry in Europe, it was found that the portion of remuneration related to output declined as the state of technology increased (Lutz & Willener 1960). Interestingly, the steel industry in the United States continues to use output incentives in the presence of advanced technology. The purpose of these incentives, however, is to emphasize equipment utilization rather than output per se.

(3) The plant is reasonably large. Very small establishments are not able to apply industrial engineering techniques of standardization and work measurement. Also, the span of control of the entire operation is usually within the limits of the owner’s capabilities, and the workers can be motivated on an individual basis. For example, bakeries, which average 28 employees per establishment, use output incentives for only 5 per cent of the work force. In larger establishments it becomes necessary and feasible to install output incentives.

(4) Other motivational devices are not available. For example, output incentives are used more frequently in industries characterized by high layoff rates and by female employment. When job tenure is short, it is not possible to appeal to employees with other rewards, such as pension plans, promotions, year-end bonuses, etc. Consequently, it is necessary to elicit performance with the direct stimulus of an output incentive.

The main advantage of output incentives is that they can elicit substantial motivation, since they are direct in nature and are applied to the individual or small group. Because of this “pull,” a company may need fewer foremen, and less discipline may be necessary.

However, output incentives produce many problems. These difficulties have been well analyzed by a number of English writers (Baldamus 1961; Behrend 1957; Marriott 1957; Shimmin 1959). In the massive study of collective bargaining in the United States (Slichter et al. 1960), many output schemes were characterized as demoralized: low effort, unduly high earnings, inequitable relationships between effort and earnings, and considerable allowance payments. While demoralization sometimes is the fault of management, quite often it stems from the inherent nature of the system itself. Output incentives teach people to beat the system, rather than to work more efficiently. People are constantly on guard against unwanted fluctuations in their take-home pay, and they press for special arrangements to handle nonstandard situations.

At the same time, production standards gradually loosen as creeping changes occur. As a result, the system slowly but steadily deteriorates. One survey showed that only 22 per cent of the output plans that had been put into operation during the preceding 15 years were still functioning as originally conceived (Payne 1951, p. 23).

Unions have played some part in this deterioration process. If a union is so inclined, it can challenge production standards and force management to loosen them; it can rally worker support for a program of restricting output; and it can demand that various allowance rules be incorporated into the labor agreement.

Evaluating the net result of output incentives is a difficult matter. Many claims have been made about the dramatic improvement that has taken place in output and unit labor costs after the installation of output incentives. However, one suspects that much of the improvement has come from the industrial engineering work that precedes the incentive application, rather than from the incentive itself (Belcher 1960, p. 98).

Cost-reduction plans. The cost-reduction approach received considerable attention in the United States after the negotiation of the Kaiser Plan in 1962. The plan distributes savings in labor and material costs between the employees and the company, roughly on a one-third-two-thirds basis. The plan also reckons with technological change by sharing the benefits from the introduction of capital in the same ratio as efficiency improvements and by guaranteeing that no one will lose employment because of mechanization (or because of the operation of the plan).

The Scanlon Plan is the best-known example of the cost-reduction approach. Since its inception in 1947, the Scanlon Plan has been installed in several dozen firms. The Rucker Share of Production Plan is another well-known system. It has been installed in about fifty or sixty firms.

The importance of these cost-reduction plans is not in the number of actual installations but in the attention which they have commanded. In common they seek to motivate employees to submit ideas for increasing efficiency and reducing costs. The plans (particularly the Kaiser and Scanlon plans) place considerable emphasis on involving union representatives and employees in solving the key problems of the business. To this end, management makes available considerable information.

Where have these cost-reduction plans been used? In some instances they have emerged as solutions to financial crises (Lesieur 1958, p. 102). There is also some suggestion that these plans have been used more frequently in closely held companies. A family-held business may be less hesitant to enter into an arrangement wherein employees can earn a bonus while, at the same time, the firm can be losing money.

The cost-reduction approach contains a number of important advantages. It focuses attention on costs rather than just on output. In many industries output is limited by technology; hence this approach appropriately directs attention to the areas of the business where achievement is possible. The Kaiser Plan also focuses attention on reducing non-labor costs—this emphasis encourages employees to improve efficiency without working themselves out of employment. The plans emphasize coordination and teamwork, not just on the factory floor but between all elements of the organization. Significantly, the Scanlon Plan includes indirect as well as direct employees, and the Kaiser Plan includes office employees.

But the cost-reduction approach also contains some weaknesses. It ties a company to a historical norm which may not reflect the competitive exigencies of the future. It is also possible to pay rewards on a continuing basis for improvements that should be viewed as “one-shot.”

On balance, it appears that cost reduction plans have worked reasonably well. During the first year of the Kaiser Plan, employees generated bonuses equal to 45 cents per hour. About one thousand suggestions were submitted, most of which were put into practice. Most of the Scanlon installations have remained in force, although there have been several abandonments (Gilson & Lefcowitz 1957). In the Scanlon installations, bonus earnings have averaged between 10 and 15 per cent of base pay and an important qualitative gain has come from the improvement in employee relations.

Profit-sharing plans. Profit sharing has been growing rapidly in the United States. The Council of Profit Sharing Industries has estimated that the number of installations grew from about nine thousand in 1950 to about 34,000 in 1960. Deferred plans have grown more rapidly than cash plans: deferred plans accounted for 24,000 installations in 1960, while cash plans outnumbered deferred plans in 1950 ("Profit Sharing Push” 1961). Profit sharing has been used with success in Europe. The coverage, however, does not appear to be as extensive as in the United States.

Usually profit sharing covers all employees in a company with the exception of a few top executives, who may be covered by a separate plan. The use of bonus plans for top executives appears to be quite extensive: at least 50 per cent of the companies use them (Belcher 1960, p. 108). Pat-ton observes that the industries which frequently use executive bonus plans (appliances, automotive, retail, and textile) tend to be divisionalized and tend to be industries where profits are influenced more by the quality of executive performance than by the ups and downs of the business cycle (Pat-ton 1961, pp. 136-137).

The recent growth in profit sharing has been due to the good business conditions which have prevailed during the postwar period. Contrastingly, profit sharing passed from the scene during the great depression of the 1930s, this falloff following a period of heavy use during the prosperous 1920s. The relative growth of the deferred plans stems from the favorable income tax status accorded monies set aside under profit sharing.

Profit sharing appears to be more prevalent in medium-sized companies. In such establishments the worker is in a better position to observe the relationship between extra effort and extra earnings (National Industrial Conference Board 1957, p. 61).

Profit sharing has some important advantages. Companies claim that it fosters economic education because people who are directly affected by the profits of a business come to learn something about the forces of the free enterprise system. Another important advantage is that profit sharing only shares rewards when they can be afforded.

But profit sharing contains a major weakness. Since profits are influenced by a wide range of forces, many of which are beyond the control of the people in the organization, people can work more industriously and receive no rewards for this.

Most companies report favorable results with profit sharing (Belcher [1955] 1962, p. 452; National Industrial Conference Board 1957). Rewards average between 10 and 15 per cent of base compensation. Quite importantly, companies also point to improvement in employee attitudes.

Measured daywork. Under measured daywork, the worker receives time wages yet management establishes—and in varying degrees discloses and enforces—production standards. While there is little statistical proof, many people feel that this form of wage payment has been growing (Marriott 1957, pp. 194, 195). The increased use of measured daywork probably reflects a greater use of industrial engineering techniques in plants that have traditionally paid time wages, rather than a major changeover from output incentives.

Measured daywork is most frequently used in large companies, where worker performance can be monitored through control techniques and sophisticated administration. This probably explains the greater use of measured daywork (and the more limited use of output incentives) in the United States than in other countries. To the extent that management in the United States can elicit employee effort through skilled supervision and advanced control techniques, direct incentives may not be as necessary as in other countries, where the industrial revolution has not proceeded as far.

Measured daywork is also being used for mechanized operations, where employees are required to work at the pace of the conveyor line or to work within the cycle of automatic machinery. Such would be the explanation for the extensive use of measured daywork in the automobile industry.

The most important advantage of measured day-work is that it avoids the difficulties inherent in output incentives. As mentioned before, the fault with many incentive systems is that they deteriorate. Since most incentive systems are difficult to abandon, the firm that operates on measured daywork has not locked itself into a difficult situation.

On the positive side, measured daywork allows a firm to introduce change with minimum resistance. Since the worker continues to receive his accustomed pay, he does not express the same resistance to new methods and production standards as does an incentive worker. Indeed, compa-anies operating on measured daywork feel that what they gain by being able to install new methods and equipment quickly and effectively more than offsets what they may lose in slower work pace.

Companies using measured daywork have also encountered some disadvantages. In order to elicit acceptable performance, it is necessary in some situations to use coercive techniques, such as discipline, often with a cost to employee relations. The automobile companies in the United States have frequently encountered strikes and slowdowns over what has been termed the “effort bargain” (Behrend 1957).

There is not much evidence about the operating results of measured daywork. One comparative study sheds some light on the question, however. Worker effort in some of the measured daywork plants was as high or higher than in the incentive plants. In the incentive plants where output was higher, the higher earnings more than offset the effect on costs of higher productivity (McKersie 1959).

Trends and conclusions. There appears to be an over-all growth in the use of different systems of incentive payment. In part this reflects the need felt by many managers to stimulate extra accomplishment. As certain developments have removed discretionary compensation from their hands—the emphasis on seniority for promotion, the emasculation of merit systems, and the growth of fringe benefits—managers have been forced to turn to wage payment plans in order to increase employee motivation.

The increasing accommodation between unions and the incentive principle has also contributed to this growth. Unions have curtailed their attempts to eliminate output incentives. Rather, they are more concerned with incentive abuses. In the case of cost reduction and profit sharing, unions have taken the initiative in proposing several installations.

While output incentives appear to be on a plateau, the more indirect and larger group plans (cost reduction and profit sharing) appear to be growing in importance. No doubt this trend will continue as mechanization increases and as white-collar employment continues to grow.

It is quite probable that unions will become more intimately involved in the operation of these plans. Their participation should influence the design of wage payment systems. In particular, job security will be given more attention—by defining achievement in terms of a more effective utilization of nonlabor factors of production and, in some cases, by actually guaranteeing that no one will lose employment as a result of the incentive plan.

Robert B. McKersie

[See also Industrial relations,article onreward systems and incentives.]

BIBLIOGRAPHY

Baldamus, W. 1961 Efficiency and Effort: An Analysis of Industrial Administration.London: Tavistock.

Behrend, Hilde 1957 The Effort-bargain. Industrial and Labor Relations Review10:503-515.

Belcher, David W. (1955) 1962 Wage and Salary Administration.2d ed. Englewood Cliffs, N.J.: Prentice-Hall.

Belcher, David W. 1960 Employee and Executive Compensation. Pages 73-131 in Industrial Relations Research Association, Employment Relations Research: A Summary and Appraisal.New York: Harper.

Bolle de Bal, Marcel 1964 En guise de réponse: Au dela de la crise de la rémunération au rendement. Sociologie du travail6, no. 2:177-187.

Campbell, Robert W. (1960) 1966 Soviet Economic Power: Its Organization, Growth, and Challenge.2d ed. Boston: Houghton Mifflin.

Dejean, Christian 1961 La crise du salaire au rendement: Un exemple beige. Sociologie du travail3, no. 2:124-139.

Gilson, Thomas Q.; and Lefcowitz, Myron J. 1957 A Plant-wide Productivity Bonus in a Small Factory: Study of an Unsuccessful Case. Industrial and Labor Relations Review10:284-297.

Guilbert, Madeleine 1960 Rémunération au temps et rémunérations au rendement dans la métallurgie parisienne. Sociologie du travail2, no. 2:107-121.

Hoffmann, Charles 1964 Work Incentives in Communist China. Industrial Relations3:81-98.

Hutchinson, John G. 1963 Managing a Fair Day’s Work: An Analysis of Work Standards in Operation.Univ. of Michigan, Bureau of Industrial Relations, Report No. 15. Ann Arbor: The Bureau.

Jehring, John J.; and Helburn, I. B. 1961 A Comprehensive Bibliography on Total Group Productivity Motivation in Business.Madison: Univ. of Wisconsin, School of Commerce, Center for Productivity Motivation.

Kennedy, Van Dusen 1945 Union Policy and Incentive Wage Methods.New York: Columbia Univ. Press.

Lajoinie, Guy A. 1961 La suppression des salaires au rendement: Un essai de solution. Sociologie du travail3, no. 2:140-160.

Lesieur, Frederick G. (editor) 1958 The Scanlon Plan: A Frontier in Labor-Management Cooperation.Cambridge, Mass.: M.I.T. Press.

Lewis, L. Earl 1960 Extent of Incentive Pay in Manufacturing. U.S. Bureau of Labor Statistics, Monthly Labor Review83:460-463.

Lupton, Thomas 1963 On the Shop Floor: Two Studies of Workshop Organization and Output.Oxford: Per-gamon.

Lutz, Burkart; and Willener, Alfred 1960 Niveau de mécanisation et mode de rémunération.Luxembourg: Haute Autorité, Communauté Européenne du Charbon et l’Acier.

McKersie, Robert B. 1959 Incentives and Daywork: A Comparative Analysis of Wage Payment Systems. Ph.D. dissertation, Harvard Univ., Graduate School of Business Administration.

McKersie, Robert B. et al. 1964 Some Indicators of Incentive Plan Prevalence. U.S. Bureau of Labor Statistics, Monthly Labor Review87:271-276.

Marriott, R. 1957 Incentive Payment Systems: A Review of Research and Opinion.London: Staples.

Metzger, Bertram L. 1964 Pro/it Sharing in Perspective: In American Medium-sized and Small Business.Evanston, 111.: Profit Sharing Research Foundation.

National Industrial Conference Board 1957 Sharing Profits With Employees. Studies in Personnel Policy, No. 162. New York: The Board, Division of Personnel Administration.

Patton, Arch 1961 Men, Money and Motivation: Executive Compensation as an Instrument of Leadership.New York: McGraw-Hill.

Payne, Bruce 1951 Incentives That Work. Pages 23-28 in Society for Advancement of Management, Fall Conference, 1951, Proceedings.New York: The Society.

Profit Sharing Push. 1961 Wall Street Journal September 13, p. 41, col. 1.

Robertson, Donald J. 1960 Factory Wage Structures and National Agreements.Cambridge Univ. Press.

Sellie, Clifford 1961 Incentives Versus Measured Day Work. Industrial Management3:15-19.

Shimmin, Sylvia 1959 Payment by Results: A Psychological Investigation.London: Staples.

Shultz, George P.; and Mckersie, Robert B. 1967 Stimulating Productivity: Choices, Problems, and Shares. British Journal of Industrial Relations5, no. 1:1-18.

Slichter, Sumner H.; Healy, James J.; and Livernash, E. R. 1960 The Impact of Collective Bargaining on Management.Washington: Brookings Institution.

Whyte, William F. 1955 Money and Motivation: An Analysis of Incentives in Industry.New York: Harper.

Wolf, William B. 1957 Wage Incentives as a Managerial Tool.New York: Columbia Univ. Press.

IV. WAGE AND HOUR LEGISLATION

In its broadest sense, the subject of wage and hour legislation includes all attempts by governments to regulate the remuneration or working hours of labor. In the narrower and also popular sense used here, the subject is confined to laws to compel or induce private employers to pay certain minimum wages and/or to limit the hours they require their employees to work. Excluded, consequently, are both determinations by governments of the wages and hours of their own employees and attempts to control maximum wages in private employment. Since the English legislation of 1802, the principal purpose of maximum-hour and minimum-wage legislation has been to improve the terms of employment of the least-fortunate wage earners. Complementary purposes have been evident at various times, however, such as to help settle union-employer disputes, to help equalize the labor costs of competing employers, to help spread available work among more persons, and to help increase aggregate purchasing power.

Evolution toward laissez-faire. Social standards of remuneration and of schedules of work have been historically the rule rather than the exception. In relatively simple and stable societies, like those of medieval Europe, custom, rather than individual decisions or formal laws, largely controlled these matters. In Europe from the thirteenth to the eighteenth centuries, customs were successively replaced by the more explicit regulations of the craft guilds, the towns, and the emerging national states, so as to control wages and hours in the face of the sweeping changes of those times in trade, population, and production techniques. None of these earlier regulations was intended to equalize or to improve the position of labor groups.

The French minimum-wage law of 1270 attempted to preserve the status quoagainst the competition of new production methods. The English Statute of Labourers of 1357 attempted to prevent wages from rising after the Black Death had cut the population in half. Henry vn even attempted to increase the hours of farm workers (from about 10 to 13 or 14 hours per day) by an English royal statute of 1495. Elizabeth i, in 1563, tried to cope with rising and chaotic prices by the Statute of Apprentices which, among other things, standardized hours (at 13½ or 14½ per day for farm work) and provided that wages of all kinds were to be fixed annually in each county by the justice of the peace. For some centuries employers used this judicial fixing of rates as an instrument of the mercantilist policy of keeping wages as low as possible. Toward the end of the eighteenth century, the early unions tried to get legal wages fixed in an effort to protect workers against the competition of those dispossessed by the enclosures, and of the cheap output of the domestic and early factory systems, but for this purpose the unions appealed to the justices in vain. Gradually, as the Statute of Apprentices fell into disuse, wage and hour decisions were left entirely to individual employers. In this area laissez-faire had become the usual practice in Britain, even before Adam Smith popularized the doctrine that economic affairs in general should be left to control by market forces alone, rather than by legislation. This laissez-faire doctrine dominated economic policy in Britain until the 1880s; it spread throughout the world as each country became industrialized; and until the 1930s it was enshrined in U.S. court decisions which repeatedly nullified legislative attempts to control wages and hours. [See Laissez-faire.]

Reactions against laissez-faire. Around the beginning of the nineteenth century, the most helpless of the uprooted British workers were toiling for as long as 18 hours a day for wages so low as to severely undermine health and shorten life. Such conditions gradually led to humanitarian demands for some legal minimum terms of employment. The resulting new legislation, in contrast to that of the preindustrial era, represented attempts to improve the worst of the existing conditions of employment rather than to preserve established standards.

Hours of work were more easily observed and controlled than were wages. Accordingly, the British statutes of 1802, 1819, 1833 and 1847 began by progressively limiting the working hours, first of children and then of women, until by 1847 the ten-hour day had become the general standard in British factories. Because factory operation required uniform working schedules, the working hours of men were in practice reduced along with those of children and women. Because the lowest of the actual wages paid had been close to a starvation level, the previous total earnings generally had to be maintained, even though the hours were reduced. Thus an indirect effect of the hours legislation was to raise wage rates.

The further reduction of standard working hours from 10 per day (for at least a six-day week) to 48 per week was first achieved by means of direct labor union pressure on employers rather than by hours legislation. These shorter hours, as provided for in collective bargaining agreements, became effective at different times in different countries, districts, and industries, with Britain generally in the lead from 1850 to the 1880s and Australasia leading from the 1800s to the 1930s. During or shortly following World War I, the majority of the unionized workers in the more industrialized countries had attained a normal eight-hour day.

In Australasia, continental Europe, and Latin America, legislation was generally used to extend to other wage earners the limitations on hours which unions had already obtained for their own members. During the 1920s and 1930s, the most common type of hours legislation provided for a basic 48-hour week, with overtime rates ranging from 110 to 150 per cent of regular rates of pay. The spread of such standards was encouraged by the International Labour Organisation (ILO), whose hours conventions of 1930 and 1935 provided for basic hours of 8 per day or 48 per week and for overtime rates of 125 per cent of regular pay. A proposed ILO convention which would have set a 40-hour standard was defeated in 1939 because of the pressure for rearmament at that time, and since World War n there has been no broad international movement toward a further shortening of hours. By 1962, however, the ILO 48-hour conventions of 1930 and 1935 had been ratified (subject to some reservations and exceptions) by 38 countries, including the newer nations of Burma, India, Iraq, Israel, Kuwait, Pakistan, Syria, and the United Arab Republic.

Modern minimum-wage legislation lagged about a century behind maximum-hours legislation, appearing in New Zealand in 1894, in Great Britain in 1909, and in the United States (Massachusetts) in 1912. Employers showed the stiffest resistance to the curtailment of their most cherished freedom to offer such wages as they might choose. At the same time labor unions, at first and especially in the United States, often feared that governmental wage fixing would undermine their opportunities to organize and to engage in free collective bargaining. Governments themselves also faced novel technical problems in framing practical and enforceable wage legislation, in view of the wide diversity and fluctuations of actual wages.

In New Zealand in 1894 and in the Australian Commonwealth and its states from 1896 to 1910, the government fixing of both minimum wages and maximum hours was generally combined with compulsory governmental arbitration of labor disputes. Thus the purpose of improving the terms of employment was joined with the purpose of preserving industrial peace. Not only minimum wages but also the whole structure of wages has come largely under government control in Australasia, as the arbitration courts have made awards in dispute cases while wage boards have set legal minimum wages for trades not sufficiently well organized to threaten to strike. Nationwide wage relationships in Australia have also been coordinated by the government, as the Commonwealth Court of Conciliation and Arbitration (established in 1904) has made awards in interstate disputes and, since 1907, has exercised considerable influence on the arbitration courts and wage boards of the individual Australian states.

In Great Britain, as in New Zealand and Australia, the evils of excessively low wages had been the subject of protest by reformist groups from the 1800s onward, and the British Trades Boards Act of 1909 was directly inspired by the Australasian examples. Britain rejected the compulsory arbitration feature of the Australasian statutes but did provide for the appointment of tripartite boards to make minimum wage awards for trades in which the Ministry of Labour found wages to be “exceptionally low.” As in Australasia, the minimum-wage awards in Britain applied to adult men as well as to women and children, and they frequently went beyond the fixing of a single minimum by specifying several graded occupational piece rates or hourly rates in any one industry. During World War i, the scope of the trades boards was extended with the aim of setting legal minimum wages in all industries where wages were not already established by employer-union agreements. Thus by 1921, 63 boards had been appointed for 39 industries which employed a total of some 3 million workers. Since 1921 the scope of the trade board system has reverted to the more limited concept of attempting to raise only exceptionally low wages, and by 1935 the awards of 44 boards then in effect applied only to industries employing about 1.1 million workers (excluding Ireland).

Some form of government control of the wages which can be paid by private enterprises is now attempted in almost every country that has a stable government, but there are wide variations between countries in the objectives, coverages, mechanisms, and effectiveness of their wage-control laws. The International Labour Office convention of 1928 (amended in 1946) attempted no more than to get agreement on the machinery for fixing of minimum wages. By 1960, 27 nations had ratified this convention, and at least 4 other nations which had not ratified had, in fact, enacted minimum-wage laws (Japan, Jordan, the Philippines, and the United States). Apart from direct governmental wage fixing, the steadily growing importance of government employment throughout the world had increased the indirect influence of the wages that governments pay to their own employees on the wages that actually must be paid by private employers.

U.S. experience. Each of the several diverse purposes of wage and hour controls has been illustrated by the history of such legislation in the United States. Although maximum-hours and minimum-wage laws were initiated in the United States almost as early as in Britain, prior to the 1930s they had been enacted by only a minority of the individual states and only for a small minority of the wage earners in those states (usually only women in selected industries). Even when enacted, such laws could be enforced only if the courts found them to involve no more than a “reasonable” restraint on the constitutionally guaranteed freedom of individual contract. The adoption of uniform national standards was impeded not only by the wide range of actual wages and hours within the country but also by the constitutional division of authority between the federal and the many state governments and also between the executive, legislative, and judicial branches of each of these governments. In sharp contrast to the early experience, however, wage and hour legislation in the United States since the 1930s has been exceptionally comprehensive, flexible, and effectual.

The U.S. courts have always been disposed to permit any governmental body to require compliance with any labor standards that government might stipulate as a condition of its own contracts with private employers. Exercising this right, the federal government set a basic ten-hour day for its own employees as early as 1840. This was reduced to eight hours and the requirement extended to its contracted work under statutes of 1868, 1892, and 1912. Since 1931 (Davis-Bacon Act) the public works contractors of the federal government have been required to pay neither less nor more than the officially determined “prevailing local wages rates” to their employees at the sites of such construction, and since 1936 (Walsh-Healey Public Contracts Act) the “prevailing rate” requirement has been extended to the manufacture, assembly, and handling of materials purchased by the federal government (under contracts of more than $10,000 apiece). Maximum hours on federally financed public works were set as low as 30 per week between 1932 and 1936, in an attempt to spread work during the depression, and since 1936 the maximum basic hours under both the public works and procurement contracts have been limited to 8 per day and 40 per week. In addition, individual state and local governments frequently set maximum hours and occasionally set minimum wages as conditions of their own contracts with private employers.

Except for government contractors and for work that was either exceptionally hazardous or that especially involved the safety of the public, the U.S. courts long maintained that any legal regulation of hours or wages represented an unconstitutional interference with freedom of contract. Maximum hours laws for women were admitted to be both the most necessary and most nearly justified as a reasonable restraint of individual freedom. But while a ten-hour law was passed as early as 1847 (New Hampshire), such legislation was nullified by the courts in 1895 (Ritchie vs. People, Illinois) and not accepted by the U.S. Supreme Court until 1908. By 1915, however, the U.S. Supreme Court had upheld even an eight-hour law for women (Miller vs. Wilson); and by 1961 all but 2 of the 52 states and territories had set some legal limitations on women’s working hours, the maximum being 48 hours or less in 22 states.

Minimum-wage legislation for women, initiated by Massachusetts in 1912, appeared to have obtained acceptance by the U.S. Supreme Court in 1917 (Stetter vs. O’Hara, 243 U.S. 629) but in 1923 was struck down by that Court (Adkins vs. Children’s Hospital,261 U.S. 525) in a decision that was not reversed until 1937 (West Coast Hotel Companyvs. Parrish,300 U.S. 329). As of 1961, minimum-wage laws had been enacted for women in 35 of the 52 states and territories, and 15 of these laws covered men also. However, the federal legislation of 1938 effectually limited the jurisdiction of any state to firms whose trade was essentially confined within the area of that state.

The great depression of the 1930s brought about the first attempt by the federal government at comprehensive wage and hour control under the National Industrial Recovery Act (NIRA) of 1933. This legislation aimed not only to improve the condition of the lowest-paid workers but particularly to stimulate business recovery, both by stopping the competitive undercutting of wage rates and by raising wage rates to increase aggregate purchasing power. The wage and hour standards under the NIRA were contained in some 557 “codes of fair competition” as proposed by the employers for each industry. The experiment was short-lived, being discontinued in 1935, after the U.S. Supreme Court decided against enforcement of the act in the first test case brought before it (Schecter Poultry Company vs. United States, 295 U.S. 495).

Comprehensive federal wage and hour controls were restored in modified form by the Fair Labor Standards Act (FLSA) of 1938 (the “wage-hour law”), following the aforementioned reversal in 1937 of the Supreme Court’s general position on the constitutionality of minimum-wage legislation. While the separate state laws have continued to set maximum hours and minimum wages for many of the local industries (which tend to pay the lowest wages), the federal law has covered the dominant group of workers involved in interstate commerce and, as such, is the most influential single wage or hour statute in the world.

Conspicuously omitted from the 1938 act were the trade-practice provisions and the price-fixing opportunities of the former NIRA, as well as the varying wage and hour standards previously formulated by employer groups for each industry. Instead, the FLSA initially fixed a single, nationwide minimum wage of 25 cents per hour and a uniform 50 per cent premium rate for all hours worked by an individual in excess of 40 per week. This “floor under wages and ceiling over hours” was expected to do something to prevent extreme forms of wage-cutting competition which might intensify a future business depression. However, the primary purpose of the wage-hour law was not that of attempting to stimulate business recovery by raising all wage rates, but rather the more familiar purpose of bringing the terms of employment of a minority of the least fortunate workers up to a minimum level “necessary for health, efficiency and general well-being.” This primary aim was not to be achieved immediately but “as rapidly as practicable …without substantially curtailing employment or earning power.” These phrases from the statute’s statement of policy reflected awareness of some price elasticity of the various demands for labor. Accordingly, standards were raised from a low initial level, and the coverage of the act was extended only gradually and cautiously.

Tripartite industry committees were used to recommend minimum rates above 25 cents an hour in the territories and above the second-year mandatory rate of 30 cents elsewhere. The swift rise in prices and wages after the outbreak of World War ii made it easy to achieve a universal 40-cent minimum by 1944 (except in Puerto Rico and the Virgin Islands) without fear of substantially curtailing employment.

After World War II, the minimum rates for the continental United States were raised by amendments to the 1938 act rather than by the industry-by-industry committee procedure. A 1949 amendment raised the minimum to 75 cents an hour, and a 1955 amendment further advanced the minimum to $1.00 an hour. In 1961 the coverage of the act was importantly extended for the first time to certain trade and service industries. To give the newly covered industries more time to adjust to the federal minimums, two schedules of automatic future increases were adopted. For all previously covered employees, the minimum was raised in 1961 to $1.15 and automatically advanced to $1.25 in 1963. The minimum for those covered for the first time in 1961 remained at $1.00 until 1964 and then rose to $1.15 and further to $1.25 in 1965.

Control of working hours by requiring overtime pay at 1J times each employee’s regular rate (not merely the minimum rate) has been retained without change, and for employees covered before 1961 the 40-hour basic week has also remained unchanged. Since the workers first covered in 1961 consisted largely of those industries where actual hours—especially in peak seasons—had frequently been considerably longer than 40 per week, a gradual adjustment to the 40-hour standard was permitted. No premium overtime pay was required for the newly covered workers during the years 1961 and 1962, and the 50 per cent premium began only after 44 hours in 1963, after 42 hours in 1964, and after 40 hours per week from 1965 onward.

Effectiveness and effects. Wage and hour legislation sometimes has consisted of a mere proclamation of a standard, the mere authorization of some machinery to fix wages or hours, or the issuance of some regulations which were not enforced. Furthermore, during periods of general price and wage inflation, a minimum wage may fall behind the lowest wages actually paid, so that even a nominal increase in the legal minimum rate may require little change in the money wages that would have been paid in any case. Even when the wage or hour legislation does require some change in minimum standards, the great majority of employers and employees may not be affected, at least not immediately and directly. In all such cases, both the proponents and opponents have been prone to exaggerate the actual consequences of wage and hour statutes.

Judgment as to the actual results of effective wage and hour controls is necessarily dependent on which of their several purposes is chosen for the evaluation. The most usual, but by no means the only, purpose of such controls since the early nineteenth century has been to improve the terms of employment of some least-fortunate fraction of the workers. This purpose has been highly objectionable, understandably and almost universally, to those employers who previously had worked their employees for longer or different hours or had employed them at lower wages than those required under the proposed legislation. The loyalty of employers to each other’s interests usually has been sufficiently strong to induce employers generally either to support the objectors or to remain silent. However, once moderate improvements in labor standards have been enforced long enough to allow employers to adjust to them, their repeal has seldom been demanded, and employers have even praised them as helping to provide a basis for “decent” competition.

Self-interested objectors to improvement of labor standards by law have usually sought to buttress their protests by appealing to the broader social-welfare doctrine of laissez-faire. The modern form of reasoning used to support that doctrine (marginal productivity theory) often has been interpreted to mean that any enforced improvement of the terms of employment must adversely affect the volume of employment and, after adjustments are made, must obstruct the most economical allocation of existing resources to satisfy consumers’ wants. Whatever its validity as an expression of general, economy-wide tendencies, this reasoning would support objection to moderate improvements of labor standards by law only if it were valid to presume that every employer was paying wages determined in a perfectly competitive labor market. Under modern labor market conditions, that presumption is quite unwarranted for many important reasons. A large increase in the wage will certainly result in some reduction of employment, but the short run wage-employment relationship is not necessarily a close or sensitive one, and employers frequently do pay less for a given quantity of labor than they could be made to pay without adverse effects on employment. Restriction of employment opportunities is, at the most, only one of the several choices actually made by the employer when hours are reduced or the wages of particular workers are raised. Even in the usual situation where improved labor standards do force up both unit costs and output prices, it may be the social value judgment that a slight reduction in the welfare of consumers as a whole is of less importance than the mitigation of the poverty of the particular workers who otherwise have little chance to improve their conditions.

Considerable empirical research has been conducted during the last quarter of a century, especially in the United States, to provide a basis for estimating the immediate direct and indirect effects of specific, alternative minimum-wage scales and for assessing the relative importance of the various actual employer adjustments to increased minimum wages—including the absorption of costs or the offsetting of costs by curtailing employment, increasing output prices, curtailing services to buyers, or adopting a variety of more efficient management methods. These studies do not, of course, support the view that wages can be raised or hours reduced to any desired extent without some adverse consequences even to the least fortunate workers themselves. They do suggest strongly that moderate and gradual enforced improvements in the poorest of the terms of employment have actually benefited the workers affected and the communities in which they lived, without major adverse consequences of any kind. At the least, wage and hour legislation appears to have been moderately successful in insuring that some of the fruits of rising average output per worker will be used to improve the welfare of the lowest-wage workers. Some disemployment of the workers who were the least highly regarded by their employers has indeed occurred, but the percentage reduction of employment has been generally much less than the percentage increase in wages (indicating a low short-run price elasticity of such demands for labor). Skill differentials in wages have generally been compressed immediately following an enforced rise in the minimum wages, but these differentials have tended to be partially restored over more extended periods of time.

The relatively low incidence of adverse effects of wage and hour legislation on either employers or employees has encouraged the fixing of standards by statute rather than by the more painstaking and painful examination of each situation by representative boards or committees. The statutory method of setting standards runs the risk, however, of either letting the improvement of standards fall behind what could be accomplished or of courting unintended adverse effects in the particular situations which fail to fit the general formula.

The research studies of the effects of wage and hour legislation, although promising, still leave very much to be desired. Essential to further progress are more studies that concentrate attention on those enterprises and communities actually affected to a considerable extent by a new legal standard, rather than studies of over-all results, including the negative ones and the mass of the cases which were not affected at all. Furthermore, much more accurate definition and measurement is needed of the economic climate in which each of the changes in a wage or hour standard is imposed. Finally, it should be recognized that empirical studies can furnish no more than a basis for informed inferences as to the pure effects of wage and hour legislation, because each situation will contain some uncontrolled and unmeasured variables and because the future is never completely predictable.

The U.S. federal legislation of 1966, although containing large increases in minimum rates and extensions in federal coverage, does not represent any change in the limited objectives and gradual-istic methods of modern minimum wage laws. The new federal rates, first effective in February 1967, ranged from $1.00 for newly covered workers to $1.40 for the 30 million previously covered workers, with both groups of rates scheduled to advance automatically up to $1.60 in future years. More of an innovation, perhaps, was the massive shift of some eight million workers from diverse standards to uniform federal standards and the coverage of some farm employees for the first time. Evaluation of these changes must await studies of specific future experience.

N. Arnold Tolles

[See also Labor force,article onhours of work.]

BIBLIOGRAPHY

Blum, Fred H. 1956 The Social and Economic Implications of the Fair Labor Standards Act: An Interpretation in Terms of Social Cost. Industrial Research Association, Proceedings9:167-183. → See pages 184-194 for discussions by J. H. Van Sickle, Lazare Teper, and N. Arnold Tolles.

Burns, E. M. 1933 Minimum Wage. Volume 10, pages 491-495 in Encyclopaedia of the Social Sciences.New York: Macmillan.

Cullen, Donald E. 1961 Minimum Wage Laws.Cornell University, New York State School of Industrial and Labor Relations, Bulletin No. 43. Ithaca, N.Y.: The School.

Douty, Harry M. 1960 Some Effects of the $1.00 Minimum Wage in the United States. EconomicaNew Series 27: 137-147.

Hicks, J. R. (1932) 1964 The Theory of Wages.New York: St. Martins. → See especially pages 179-216, “Wage-regulation and Unemployment,” and pages 218-226, “Hours and Conditions.”

Karlin, Jack I. 1967 Economic Effects of the 1966 Changes in the FLSA. Monthly Labor Review90, no. 6:21-25.

Kocin, Susan 1967 Basic Provisions of the 1966 FLSA Amendments. Monthly Labor Review90, no. 3:1-4.

Lubin, Isador; and Pearce, Charles A. 1958 New York’s Minimum Wage Law: The First Twenty Years. Industrial and Labor Relations Review11:203-219.

Lyon, Leverett S. et al. 1935 The National RecoveryAdministration: An Analysis and Appraisal.Washington: Brookings Institution.

Martin, Edward C. 1967 Extent of Coverage Under FLSA as Amended in 1966. Monthly Labor Review90, no. 4:21-24.

Millis, Harry A.; and Montgomery, Royal 1938 Labor’s Progress and Some Basic Labor Problems.New York: McGraw-Hill. → See especially pages 278-375, “Government Regulation of Wages,” and pages 463-536, “Hours of Work.”

New York State, Department of Labor, Division of Research and Statistics 1964 Economic Effects of Minimum Wages: The New York Retail Trade Order of 1957.Albany, N.Y.: The Department.

Princeton University, Industrial Relations Section 1955 Economic Issues in Federal Minimum Wage Legislation: Selected References. Princeton Univ. Press.

Princeton University, Industrial Relations Section 1963 Hours of Work: Selected References.Princeton Univ. Press.

Sells, Dorothy M. 1939 British Wages Boards: A Study in Industrial Democracy.Washington: Brookings Institution.

Tolles, N. Arnold 1959 American Minimum Wage Laws: Their Purposes and Results. Industrial Research Association, Proceedings12:116-133. → See pages 134-143 for discussions by Peter Henle and John M. Peterson.

Tolles, N. Arnold 1964 Origins of Modern Wage Theories.Englewood Cliffs, N.J.: Prentice-Hall. → See especially pages 108-125, “National Minimum Wages: NRA and the Wage-Hour Law.”

Weiss, Harry 1956 Economic Effects of a Nationwide Minimum Wage. Industrial Research Association, Proceedings9:154-166. → See pages 184-194 for discussions by J. H. Van Sickle, Lazare Teper, and N. Arnold Tolles.

Woytinsky, Wladimir 1932 Hours of Labor. Volume 7, pages 478-493 in Encyclopaedia of the Social Sciences.New York: Macmillan.

V. FRINGE BENEFITS

“Fringe benefits” is a term embracing a variety of employee benefits paid by employers and supplementing the worker’s basic wage or salary.

In an industrial society the basic method of compensating an employee for his labor has been the payment of a wage or salary for each unit of time spent on the job (or each unit of work completed). Traditionally, the obligation of the employer to his employee was considered complete with the payment of this basic compensation.

With increasing industrialization and rising incomes, the interests of employees in compensation has widened to include more than this basic payment, and concurrently the employer’s ability to provide supplementary items of compensation has improved. Conceivably the development of many supplementary compensation practices could have been financed through higher wages, thus permitting employees to purchase directly such fringe benefits as health insurance and retirement annuities. Two factors, however, have contributed to direct financing by employers. (1) Because employer payments can be made with “before-tax” dollars, direct financing can provide greater benefits to employees for a given sum of money than would be obtained if the same amount were distributed as wages, on which each employee is forced to pay taxes. (2) Lower costs for benefits financed through insurance can be obtained by covering a large number of employees.

There is no universally accepted group of practices embraced by the term “fringe benefits.” Generally, it can be said that a fringe benefit has to meet two tests: it must provide a specific benefit to an employee, and it must represent a cost to the employer. Clearly included within this concept are payments to employees for various types of paid leave (vacation, holiday, military duty, personal) and payments for various welfare benefits (retirement, health, life insurance, unemployment).

Another group of fringe benefits are those financed by employer payments under government programs. In the United States typical examples are the employer’s taxes paid to the federal government under the Old Age, Survivors, and Disability Insurance system, his state taxes for unemployment insurance, and his required payments for workmen’s compensation. In these instances, the law sets the standards under which payments are made and benefits distributed.

A final group of payments often classified as fringe benefits are made by employers for work performed at specific times or under specific conditions. An example is premium pay for overtime work or for work on Sunday or holidays, shift differentials, and additional payments for work performed under hazardous conditions.

Other examples of payments which, under some classifications, are considered fringe benefits are contributions to profit-sharing, thrift, or savings plans, moving expenses, educational or training expenses, and employer-financed recreational activities for employees.

Fringe benefits evolved slowly in the United States and other industrializing countries. The limitations imposed by shortages of capital, low productivity, and low levels of income invariably led to an emphasis on the basic wage payment. The early unions concentrated on higher wages and shorter hours.

While some fringe benefits originated almost at the start of industrial organization, these were largely confined to managerial or salaried employees. In some cases employers extended these benefits to manual workers, either on their own volition or as the result of actual or threatened union organizing efforts. The rapid development of fringe benefits in recent years, commencing with the 1930s, has brought greater parity in this regard between “blue-collar” and “white-collar” workers.

The growth of certain fringe benefits in the United States has been stimulated by governmental actions. Social security programs and the statutory overtime premium were adopted during the 1930s. The growth of private pension plans has been assisted by special tax treatment for money set aside by employers in a special retirement fund and by rulings of the National Labor Relations Board and the courts declaring pension plans within the scope of legal collective bargaining.

The development of fringe benefits started later in the United States than in many other industrialized nations, where public social security programs were initiated long before the New Deal. For many European countries, including France, West Germany, and Italy, fringe benefits constitute a higher proportion of total compensation than in the United States. In Great Britain, however, the proportion is roughly comparable to that of the United States, although it should be noted that certain social security programs financed through general tax revenue are more comprehensive than those in the United States. A major difference between the United States experience and that of most European countries is that in the United States the greater proportion of fringe benefits has been initiated through collective bargaining, whereas in European countries the prevailing practice has been set by legislation. For example, most European countries prescribe vacation and holiday practices through legislation, but not the United States. In recent years, however, European collective bargaining agreements have been including more provisions dealing with fringe benefits.

The various types of fringe benefits reflect the different attitudes or interests of workers and employers.

Vacations and holidays. The development of paid leave practices, consisting largely of paid vacations and paid holidays, reflects what might be called “the quest for leisure.” For most workers in the United States, paid leisure of this type represents a post-World War n phenomenon. A 1940 study disclosed that only about one-fourth of union members were eligible for annual vacations, and for most of these the maximum vacation period was one week. Although major holidays were frequently observed throughout industry, the practice of providing holiday pay for hourly rated employees was quite rare.

By 1967, almost all workers covered by bargaining agreements were eligible for paid vacations, and judging by the major agreements, over 70 per cent of these agreements provide at least a four-week vacation for longer-service employees. Five-week and longer vacations are now found in about 15 per cent of major bargaining agreements. Similarly, the most recent survey of holiday provisions in major bargaining agreements indicates that about 90 per cent of the workers covered were entitled to paid holidays. Nearly all of these were entitled to at least six holidays, and more than four-fifths of them received seven or more paid holidays.

Disability and unemployment. Another major stream in the development of fringe benefits originated with the worker’s need for supplementary income to meet hardships imposed when he is prevented from working by disability or lack of employment opportunities. In many countries, including the United States, government programs are generally available to assist on these occasions. However, United States government programs for wage replacement of losses resulting from non-occupationaldisabilities are limited to four states and railroad employees.

To fill the gap left by the lack of U.S. government programs for off-the-job accidents and illnesses and to supplement public programs for unemployment insurance, private arrangements have been developed through collective bargaining and through unilateral actions by employers. By 1967, three out of five workers in private nonfarm industry were protected by some type of temporary-disability benefit plan: a formal sick-leave plan, an insured accident and sickness plan, or a government-operated plan.

Private programs to supplement state unemployment insurance were first introduced in 1955, when this type of program was incorporated into the bargaining agreements in the automobile industry. Within nine months after this contract was adopted, these plans had spread to cover more than 1.2 million workers. At the present time, an estimated 2.5 million workers are covered by plans in the auto, steel, rubber, apparel, cement, flat glass and other industries.

Health. To meet the costs of personal health care of employees and their dependents, health insurance and prepayment plans have been widely adopted by American industry. Hospital and surgical benefits now cover three out of four wage and salary workers; three out of five have regular medical benefits coverage; and in a rapidly growing development, over one out of four has major medical expense coverage to help defray unusually large health care expenses. While the employer in nearly all cases pays some part of the cost, frequently the entire amount, the cost of dependent coverage is often borne by the employee.

In recent years, hospital and surgical benefits have been improved to cover additional costs and services. Supplementary protection has been extended by many plans to retired workers (who are covered after age 65 by the federal government’s Medicare program) and, to a lesser extent, to employees on temporary layoff. Some progress has been made in the last few years in extending benefits to the costs of dental and optical care and long-term disability.

Life insurance. Even more prevalent than hospital benefits are death benefits, which are usually provided through group life insurance. Currently, over two out of three wage and salary workers have such protection, and coverage is often extended to laid-off and retired workers and, to a lesser extent, to the dependents of active workers. Although the amount of coverage for active employees varies widely, the median is around one year’s earnings. Under most collectively bargained plans, the entire cost of at least some employee coverage is paid by the employer; under most nonnegotiated plans, the employee usually pays a substantial part of the cost. Additional coverage is often made available to employees at their own expense.

Pension plans. A similar development is the growth of private pension plans. The importance of these arrangements is underscored by the increasing life expectancy, under which the average employee in the United States can look forward to 14 years of retirement after reaching the normal retirement age of 65.

Government programs for retirement income are well developed in most countries. The basic United States program (Social Security) was one of the most significant of the New Deal measures enacted during the administration of President Franklin D. Roosevelt. Passed in 1936, it has been broadened several times by Congress and now includes, in addition to higher levels of benefits for retiring workers, benefits for survivors and for covered workers who become permanently and totally disabled. The emphasis in the public program is necessarily placed on providing basic retirement benefits to all members of the labor force. It provides the worker with low earnings a relatively larger benefit than the worker with higher earnings and sets a maximum limit to the taxable earnings on which benefits are based. Consequently, the need has arisen for supplemental private pension plans to yield for many workers a more satisfactory level of retirement income.

In 1967, private retirement plans, including deferred profit-sharing plans, covered over 25 million workers, or about half the employees in private nonfarm establishments. Collectively bargained pension plans covered about 11 million workers. The employer pays the entire cost of over two-thirds of the plans; in the jointly financed plans to which employees also make contributions, the employer pays the major proportion of the cost.

Included in the total are nearly one thousand multiemployer pension plans, covering about 4 million workers. Multiemployer pension plans differ from single-employer plans in two important respects: (1) their pension credits are portable, in that employees may accumulate credits by working for any employer who belongs to the plan; and (2) the employer members agree with the union to contribute to the plan at a certain rate—usually a specified number of cents per man-hour or per cent of payroll—rather than to provide specified pension benefits. The benefits are subsequently determined by a joint union-management board of trustees.

Benefit levels vary widely among pension plans and often within plans, depending on the specific benefit formula. Under the benefit formulas for current service in effect in 1963, the median pension payable at age 65 to workers with average annual earnings of $4,800 was about $54 a month for 20 years of credited service and $78 for 30 years. Contributory plans normally provide higher benefits than noncontributory plans.

Premium pay. Another major set of fringe benefits grows out of the practice of providing premium pay for work performed after the normal day’s or week’s work is completed or at inconvenient times or under especially hazardous conditions. While such premiums are common in industrialized countries, the premium rates in the United States are generally higher than in most other countries.

The rate of time and a half for work performed after 40 hours in any one work week was a basic provision of the 1938 Fair Labor Standards Act. In many industries collective bargaining agreements make provision for premium payments after eight hours in any one day and for work performed on the weekend, regardless of the total number of hours worked during the week. Higher premiums generally prevail for work performed on a Sunday or a holiday.

Higher payments for work performed on a second or third shift are also common. Typical shift differentials in the United States range between 8 and 11 cents an hour for the second shift and between 11 and 13 cents an hour for the third shift. Premium payment for especially hazardous work is found in the longshoring, maritime, and construction industries.

Extension of benefits. Fringe benefit practices are among the most rapidly changing aspects of industrial relations. In 1966, for example, almost four-fifths of all major collective bargaining settlements involved new or liberalized fringe benefits. The most rapidly changing benefits were health and welfare plans, modified in nearly three-fifths of the settlements. Changes in retirement plans and paid vacations appeared in almost half the newly negotiated contracts.

Changes have been of two types: certain existing fringe benefits have been liberalized, for example, the length of service required to obtain longer vacations has been reduced; in addition, certain types of fringe benefits have spread to parts of the economy where they were formerly considered impractical. The construction industry, despite its multitude of small contractors and high turnover rate, has been able to develop special funds not only for health and retirement benefits but also for vacation payments. In many cases, multiem-ployer retirement plans have been developed in industries where individual employer plans would be impractical.

Fringe benefits have also proved to be a useful tool in meeting specialized problems of adjustments to technological change involving possible reductions in the work force. In recent years supplementary unemployment benefits (SUB) have been revised not only to provide higher weekly payments and more weeks of benefits but also to cover short work weeks and to pay other benefits from SUB funds. In the case of the automobile agreements, the funds are also used to finance separation allowances; hospital, medical, and surgical insurance for laid-off workers eligible for SUB benefits and their dependents; and if there are sufficient funds, a cash payment is made at the end of the year.

Another way in which fringe benefits have helped to meet industry problems of adjustment is by the adoption of more liberalized early retirement in situations where older workers have been affected by layoffs and plant shutdowns. The use of severance payments for laid-off workers is also increasing and was a crucial element in the final arbitration award settling the 1959-1963 dispute in the United States between the railroads and the unions of operating employees. The award authorized the discharge of firemen but awarded them severance payments depending upon their accumulated service.

Problems. The widespread adoption of fringe benefits has raised several issues involving the effective functioning of an economy. A few of the most important are the following.

Increasing cost of fringe benefits. In 1962, the average United States manufacturing establishment paid an amount equal to 22 per cent of payroll in fringe benefit costs. This represented an increase from 20 per cent in 1959. Comparable expenditures in mining were 26 per cent in I960; and in finance, insurance, and real estate 23 per cent in 1961. Among individual industries, wide differences in fringe benefit costs generally reflect similar differences in wage rates.

Increases in fringe benefit costs affect an individual firm’s total labor costs and become one factor in affecting the firm’s decisions regarding prices. For an economy as a whole, increases in total compensation (wages plus fringe benefits) which exceed increases in productivity lead to pressures for rising prices. Thus, the size of increases in fringe benefit expenditures is one important factor in determining changes in the level of prices.

Another aspect of fringe benefit costs that has drawn attention is the impact of certain costs (particularly paid leave, welfare plans, and legally required payments) on the employer’s willingness to hire additional employees. Since these benefits add to the cost of placing new workers on the payroll, the employer who has a choice may find that the additional cost of scheduling overtime work has diminished compared to the cost of hiring additional workers. Action to modify this disparity has become an issue of public policy in the United States, with labor unions contending that the overtime penalty rate under the Fair Labor Standards Act should be raised, while employers oppose this as an unduly burdensome and costly approach which would not yield additional jobs.

Coordination of private and public programs.The development of private welfare programs has led to possible conflict and duplication with public programs in the fields of retirement, disability, and unemployment. Certain federal and state legislation in the United States is directed at these problems; but even at best, such legislation could not be expected to meet the many diverse needs throughout the country. The use of collective bargaining or employer personnel policies to supplement legislative standards has provided more adequate benefits and encouraged the development of programs tailored to meet the needs of specific industries, firms, or groups of workers.

On the other hand, the question has been raised whether this procedure may not allow a relatively fortunate few to achieve high standards of protection while neglecting the more basic legislative standards on which the many must depend.

Effect on worker’s relation to his job.Another issue that has been raised is whether some fringe benefits have had the effect of tying the worker too closely to a single employer, thus preventing him from being more effectively used by another employer, possibly at a higher skill.

A number of fringe benefits increase in value as the worker accumulates service on the job. Examples are the length of an employee’s paid vacation or the size of his pension. Because of his equity in these benefits, an employee may be reluctant to shift jobs even for a higher basic wage rate. An employee shifting jobs at age 50, for example, may find upon retirement that he does not have enough service to qualify for a pension from either his old or his new employer.

In any analysis of this issue, the difficulty has been to separate the specific effects of fringe benefits from the more general effect of seniority as a whole. It is true that job shifts are less common among older employees, but much of this simply reflects the individual’s continued satisfaction with his job together with the security he has accumulated over a period of years. Fringe benefits may play but a small role in his reluctance to change.

One approach to this question has been to develop modifications of fringe benefits to encourage greater mobility of the work force. One example is the introduction in private pension plans of provisions which vest in the individual employee who meets certain age and service requirements his right, at retirement age, to all or part of his accrued pension benefits regardless of his employment status at that time.

Peter Henle

BIBLIOGRAPHY

Beier, Emerson H. 1967 Terminations of Pension Plans: 11 Years’ Experience. Monthly Labor Review90, no. 6: 26-30.

Bernstein, Merton C. 1964 The Future of Private Pensions.New York: Free Press.

Block, Joseph W. 1956 Problems in Measurement of Expenditures on Selected Items of Supplementary Employee Remuneration: Manufacturing Establishments, 1953.U.S. Bureau of Labor Statistics, Bulletin No. 1186. Washington: Government Printing Office.

Engen, Gunnar 1967 A New Direction and Growth in Profit Sharing. Monthly Labor Review90, no. 7:1-8.

Federal Employment—Retirement Insurance and Medical Care Programs. 1964 U.S. Civil Service Commission, Library Personnel Bibliography Series,No. 14.

Greene, Dorothy [Kittner] 1960 Health and Insurance Plans Under Collective Bargaining: Hospital Benefits, Early 1959.U.S. Bureau of Labor Statistics, Bulletin No. 1274. Washington: Government Printing Office. → Summarizes provisions of 300 plans.

Greene, Dorothy [Kittner]; and Davis, Harry E. 1959 Health and Insurance Plans Under Collective Bargaining: Accident and Sickness Benefits, Fall 1958.U.S. Bureau of Labor Statistics, Bulletin No. 1250. Washington: Government Printing Office. → Summarizes provisions of 300 plans.

Health Insurance Institute, New YorkA List of Current Health Insurance Books. → Published since 1959, under varying titles.

Holland, Daniel M. 1966 Private Pension Funds: Projected Growth.National Bureau of Economic Research, Occasional Paper No. 97. New York: The Bureau; Columbia Univ. Press.

Kittner, Dorothy R. 1961 Health and Insurance Plans Under Collective Bargaining: Life Insurance and Accidental Death and Dismemberment Benefits, Early Summer 1960.U.S. Bureau of Labor Statistics, Bulletin No. 1296. Washington Government Printing Office. → Summarizes provisions of 300 plans.

Kittner, Dorothy R. 1962 Digest of Profit-sharing, Savings and Stock Purchase Plans, Winter 1961-1962.U.S. Bureau of Labor Statistics, Bulletin No. 1325. Washington: Government Printing Office. → Summarizes 20 selected plans.

Kittner, Dorothy R.; and Davis, Harry E. 1960 Health and Insurance Plans Under Collective Bargaining: Surgical and Medical Benefits, Late Summer 1959.U.S. Bureau of Labor Statistics, Bulletin No. 1280. Washington: Government Printing Office. → Summarizes provisions of 300 plans.

Kolodrubetz, Walter W. 1964 Labor Mobility and Private Pension Plans: Vesting, Early Retirement and Portability Provisions.U.S. Bureau of Labor Statistics, Bulletin No. 1407. Washington: Government Printing Office.

Kolodrubetz, Walter W. 1967 Growth in Employee-benefit Plans, 1950-1965. Social Security Bulletin30, no. 4:10-27.

Landay, Donald M. 1961 Health and Insurance Plans Under Collective Bargaining: Major Medical Expense Benefits, Fall 1960.U.S. Bureau of Labor Statistics, Bulletin No. 1293. Washington: Government Printing Office. → Summarizes provisions of 300 plans.

Mcgill, Dan M. (1955)1964 Fundamentals of Private Pensions.2d ed. Homewood, III.: Irwin.

Metzger, Bertram L. 1966 Profit Sharing in Perspective.Evanston, III.: Profit Sharing Research Foundation.

Munts, Raymond 1967 Bargaining for Health.Madison: Univ. of Wisconsin Press.

Pfeffer, Aloysius R. 1963 Employer Expenditures for Selected Supplementary Remuneration Practices for Production Workers in Mining Industries, 1960.U.S. Bureau of Labor Statistics, Bulletin No. 1332. Washington: Government Printing Office.

Rosenbloom, Henry S. 1960 Paid Sick Leave Provisions in Major Union Contracts, 1959.U.S. Bureau of Labor Statistics, Bulletin No. 1282. Washington: Government Printing Office.

Slichter, Sumner H.; Healy, James J.; and Livernash, E. Robert 1960 The Impact of Collective Bargaining on Management.Washington: Brookings Institution.

Somers, Herman M.; and Somers, Anne R. 1961 Doctors, Patients, and Health Insurance: The Organization and Financing of Medical Care.Washington: Brookings Institution.

Strasser, Arnold 1964 Employer Expenditures for Selected Supplementary Compensation Practices for Production and Related Workers, Meatpacking and Processing Industries, 1962.U.S. Bureau of Labor Statistics, Bulletin No. 1413. Washington: Government Printing Office.

U.S. Bureau of Labor Statistics 1959 Premium Pay for Night, Weekend, and Overtime Work in Major Union Contracts. Bulletin No. 1251. Washington: Government Printing Office.

U.S. Bureau of Labor Statistics 1962a Employer Expenditures for Selected Supplementary Remuneration Practices for Production Workers in Manufacturing Industries, 1959.Bulletin No. 1308. Washington: Government Printing Office.

U.S. Bureau of Labor Statistics 1962b Paid Leave Provisions in Major Contracts, 1961.Bulletin No. 1342. Washington: Government Printing Office.

U.S. Bureau of Labor Statistics 1964 Digest of Fifty Selected Health and Insurance Plans for Salaried Employees, Spring 1963.Bulletin No. 1377. Washington: Government Printing Office.

U.S. Bureau of Labor Statistics 1965a Digest of 100 Selected Pension Plans Under Collective Bargaining, Late 1964.Bulletin No. 1435. Washington: Government Printing Office.

U.S. Bureau of Labor Statistics 1965b Employer Expenditures for Selected Supplementary Compensation Practices for Production and Related Workers: Composition of Payroll Hours, Manufacturing Industries, 1962.Bulletin No. 1428. Washington: Government Printing Office.

U.S. Bureau of Labor Statistics 1965c Major Collective Bargaining Agreements: Severance Pay and Layoff Benefit Plans.Bulletin No. 1425-2. Washington: Government Printing Office.

U.S. Bureau of Labor Statistics 1965d Major Collective Bargaining Agreements: Supplemental Unemployment Benefit Plans and Wage-Employment Guarantees.Bulletin No. 1425-3. Washington: Government Printing Office.

U.S. Bureau of Labor Statistics 1965e Research Sources on Employees’ Supplementary Benefits. Compiled by William Gerber. Mimeographed manuscript. → Covers all areas of the world.

U.S. Bureau of Labor Statistics 1965f Supplementary Compensation for Nonproduction Workers, 1963.Bulletin No. 1470. Washington: Government Printing Office.

U.S. Bureau of Labor Statistics 1966a Digest of 50 Selected Pension Plans for Salaried Employees, Summer 1965.Bulletin No. 1477. Washington: Government Printing Office.

U.S. Bureau of Labor Statistics 1966b Digest of 100 Selected Health and Insurance Plans Under Collective Bargaining, Early 1966.Bulletin No. 1502. Washington: Government Printing Office.

U.S. Bureau of Labor Statistics 1966c Financing Supplemental Unemployment Benefit Plans.Bulletin No. 1483. Washington: Government Printing Office.

U.S. Bureau of Labor Statistics 1966d Private Pension Plan Benefits. Bulletin No. 1485. Washington: Government Printing Office.

U.S. Bureau of Labor StatisticsWages and Related Benefits: Major Labor Markets. → Published since 1951/1952. An annual survey reporting prevalence of fringe benefits for individual labor markets (Part i) and for economic regions (Part n).

U.S. Office of Labor-Management and Welfare-Pension Reports 1967 Welfare and Pension Plan Statistics: Characteristics of 163,500 Plans Filed as of July 1, 1965.Washington: Government Printing Office.

U.S. Securities and Exchange CommissionPrivate Non-insured Pension Funds. → Published annually since 1966.

Weintraub, Gerald D. 1964 Employer Expenditures for Selected Supplementary Remuneration Practices: Finance, Insurance, and Real Estate Industries, 1961.U.S. Bureau of Labor Statistics, Bulletin No. 1419. Washington: Government Printing Office.

WAGES

WAGES. The history of wages in early modern Europe is a study of contrasts. To begin with, most people toiled on family farms or in family enterprises. Hence wages were a dominant part of income for only a small fraction of the population. Nevertheless, hiring workers for wages and working for someone else part of the time were extremely common. The tension between these two facts has informed the two key debates about wages in the early modern period. The first debate, accepting the ubiquity of paid labor, uses wages to infer standards of living and thus examine Malthusian cycles. The second debate involves both the extent of wage labor and the institutions that made it respond or not respond to the laws of supply and demand.

In 1798, the British social theorist Thomas Malthus (1766–1834) argued that in agrarian economies (agriculture absorbed two-thirds of all workers in nearly all European regions prior to 1800) incomes depended on the ratio of land to population. More land allowed higher output per person; more people drove down the output per person. Because land rents increase when land is scarce, wages are even more sensitive to scarcity than output per person. This narrative has been adopted, with slight variation, by many different scholars who believe that these iron laws held firm for millennia. For some, these shackles were eventually broken by the increased use of coal, for others by access to the agricultural output of the New World, or even by technical change broadly defined and dated to sometime in the mid-eighteenth century. From the time of the Black Death (mid-fourteenth century) to the 1750s, wage series did follow a broad Malthusian pattern. In England, for instance, wages started from a low in the mid-1300s, rose for nearly a century and a half in response to the epidemic's massive mortality rate, then fell for an equally long time, bottoming out in the seventeenth century. The rise of wages in the eighteenth century was not pronounced, but wage stability in the face of massive population growth was nonetheless an important achievement. Bits and pieces of this story can be seen in all European countries, though each in its fashion raises questions about the standard Malthusian model.

In recent years, Malthus has been under strong challenge. First, as Van Zanden states, wages are not income. At the individual level, nonwage compensation—from common rights, or home manufacture, for instance—was an important element of most families' income in the early modern period. At the national level, earnings from land, capital, skills, and entrepreneurship were of considerable value, even though their distribution was quite different from that of wages. As the recent historical record suggests, economies can experience massive growth without witnessing much real wage increase for the unskilled. In the past as in the present, one should investigate wages with some concern for inequality.

Second, and more problematic, is the evidence that comes from examining regional patterns in wages. Regional variation in wages at any point in time is of the same order of magnitude as the two-century variation in wages of a local Malthusian cycle. If we compare high wage regions to low wage regions Malthus's theory fails again. In fact, high population areas did not have low wages. On the contrary, economically leading areas were most often very densely populated relative to the European hinterland. Northern Italy, the Low Countries, and England all were or became densely populated in their period of economic leadership; and they were all also high wage economies. Economic historians now argue that Malthus's emphasis on endowments and demography explained in part the evolution of economies and wages. Political institutions and economic institutions have at least as much importance.

The second debate arrays two sides. On one side scholars argue that families in the early modern era preferred self-sufficiency to the uncertainty or unfairness of market interaction. Therefore they avoided labor markets. These scholars also argue that, unlike in modern society, workers and their employers were enmeshed in a web of social relations that only capitalism would break. In this view, labor exchange was relational rather than market-driven. In such a situation, one would prefer to employ an acquaintance at a higher wage rather than hire an outsider for less. In contrast, the argument continues, modern factory workers have no social relations either with management or with the distant shareholders of the corporation they work for; hence wages are free to reflect the iron law of supply and demand.

That view has come under repeated challenge. In part, this is because the arguments that seek to differentiate early modern from modern labor markets have been made on unsound quantitative evidence and are based on a very naive view of how labor markets operate. When scholars take into account that labor markets are always imperfect, differences between those of the preindustrial and contemporary eras cease to be differences in kind.

The market-avoidance argument fails for empirical reasons: only a small fraction of farms and enterprises were the right size to have an exact balance between their labor demand and their family labor supply. Imbalances arose for different reasons, including seasonal peaks in labor demand at harvest, the demographic cycle in crafts, and the difficulty of adjusting farm size to family size. Therefore many, probably most, families either bought or sold days of labor, earning or paying wages. These wages did reflect supply and demand, rising in summer as demand for labor increased, and falling when population growth was rapid and during bad harvests, when the amount of work was reduced, and so on.

There were some important exceptions. For instance, in eastern Europe the strengthening of serfdom stymied labor markets. But there were other areas, like the Low Countries, where wage labor was quite prevalent by the end of the Middle Ages. Overall, the extent of wage labor seems to have paralleled the extent of markets in general: where trade and commerce were more active, one could observe more active labor markets.

See also Capitalism ; Commerce and Markets ; Laborers ; Servants .

BIBLIOGRAPHY

Allen R. C. "The Great Divergence in European Wages and Prices from the Middle Ages to the First World War." Explorations in Economic History 38, no. 4 (October 2001): 411–448.

Brown, Henry Phelps, and Sheila V. Hopkins. A Perspective of Wage and Prices. London, 1981.

Campbell, B. "Agricultural Progress in Medieval England: Some Evidence from Eastern Norfolk." Economic History Review 36, no. 1 (February 1983): 26–46.

de Vries, J., and A. van der Woode. The First Modern Economy: Success, Failure and Perseverance of the Dutch Economy 1500–1815. Cambridge, U.K., 1997.

Epstein, S. "Craft Guilds, Apprenticeship, and Technological Change in Preindustrial Europe." Journal of Economic History 58, no. 3 (September 1998): 684–713.

Grantham, G. "Contra Ricardo: On the Macroeconomics of Pre-Industrial Economies." European Review of Economic History 3, no. 2 (August 1999): 199–232.

Hoffman, P. T. Growth in a Traditional Society. Princeton, 1998.

Hoffman, P. T., D. Jacks, and P. Lindert. "Real Inequality in Europe Since 1500." Journal of Economic History 62, no. 2 (June 2002): 322–355.

Jones, E. L. The European Miracle: Environments, Economies, and Geopolitics in the History of Europe and Asia. 2nd ed. Cambridge, U.K., 1987.

Malthus, T. R. An Essay on the Principle of Population. 1798.

Mokyr, J. The Lever of Riches: Technological Creativity and Economic Progress. New York, 1990.

North, D. C. Structure and Change in Economic History. New York, 1981.

Ozmucur, S., and S. Pamuck. "Real Wages and Standards of Living in the Ottoman Empire 1489–1914." Journal of Economic History 62, no. 2 (June 2002): 293–321.

Reddy, W. M. The Rise of Market Culture: The Textile Trade and French Society, 1750–1900. Cambridge, U.K., 1984.

Van Zanden, J. L. "Rich and Poor before the Industrial Revolution: A Comparison between Java and the Netherlands at the Beginning of the Nineteenth Century." Explorations in Economic History 40, no. 1 (January 2001): 1–23.

Wrigley, E. A. Continuity, Chance and Change: The Character of the Industrial Revolution in England. Cambridge, U.K., 1988.

Wrigley, E. A., and R. S. Schofield. The Population History of England 1541–1871: A Reconstruction. Cambridge, Mass., 1981.

Jean-Laurent Rosenthal

Wages

THE LABOR MARKET UNDER PERFECT COMPETITION

A DEPARTURE FROM A PERFECTLY COMPETITIVE LABOR MARKET

BIBLIOGRAPHY

Wages are remuneration for labor services, in the form of either cash or some other mechanism of compensation. Wages have been the subject of extensive empirical and theoretical inquiry, the focus of which has included causes and consequences of wage variation, implications of evolving labor market conditions, and the effects of wage increases on productivity and growth. Generally, these studies approach such questions through the prism of two alternative labor market circumstances: perfectly competitive markets, in which consumers and producers have no market power to affect prices, and imperfectly competitive markets.

THE LABOR MARKET UNDER PERFECT COMPETITION

Traditionally, economic analysis is predicated on the assumption that labor markets are perfectly competitive. For this to be reasonable, three criteria must hold: There must be no barriers to entry; agents must have complete information; firm and labor mobility must be costless. When these conditions obtain, wage rates of a given quality are determined by the intersection of the labor supply and demand curves. The labor demand curve reflects the marginal productivity of labor confronting profit-maximizing firms. The labor supply curve reveals the willingness of utility-maximizing individuals to supply labor at each conceivable wage. The intersection of these two curves determines the market clearing wage rate, at which all persons willing and able to work will find employment. In this framework, the wage rate is a market price that effectively allocates labor across alternative (and competing) uses.

One vein of the empirical work mentioned above is essentially rooted in this perfectly competitive framework. For instance, there is a tradition of fairly straightforward examination of inter-temporal wage patterns implicitly interpreted through the lens of perfect competition. Katz and Autor (1999) examine longitudinal wage data in the United States, revealing several interesting patterns. The ratio of wages of those in the ninetieth percentile of workers to those in the tenth percentile has grown substantially in recent decades. Large wage disparities have also been detected by education and experience levels. Nonetheless, wage gaps between white and black males and males and females overall have narrowed.

There is, however, no consensus regarding the causes of these structural changes. An often-cited theory explains wage differentials as a result of investments in human capital (e.g., through education, health, or training). Becker (1964) formalized this idea by demonstrating that greater human capital expenditures could account for empirically observed patterns of wage variation via their impact on the marginal revenue product of labor. This theory received empirical support from Mincer (1974), whose estimates of the returns to education were positive and increasingly significant with controls for years of experience in the labor market.

Numerous studies question this explanation of observed wage differences. For instance, Pritchett (2001) shows that education has contributed very little to economic growth. He speculates that, among other possibilities, this might reflect the irrelevance of knowledge transmitted through formal schooling (in terms of the skills actually required in the workplace). From a slightly different perspective, some (e.g., Spence [1974]) have argued that education contributes little directly to labor market productivity, but instead serves as a kind of signal for ability that allows employers to sort high and low ability workers.

Another avenue for wage variation generally approached through the framework of perfect competition is immigration. LaLonde and Topel (1991) consider the impact of newly arrived immigrants on the wages of workers born in the United States as well as those who had immigrated to the United States earlier. They find that immigration has a negative, but small, impact on the wages of earlier immigrants and native-born workers who represent close substitutes (i.e., low skilled workers) for these more recent arrivals. This relationship attenuates as the tenure of earlier immigrants in the United States increases. Others (e.g., Borjas [2003]) have disputed these findings, claiming that they stem from the failure to control for the migration of native-born Americans to other cities and states.

A DEPARTURE FROM A PERFECTLY COMPETITIVE LABOR MARKET

The other major theoretical and empirical tradition focuses on possible market failure resulting in involuntary unemployment (a situation where workers willing to work at the prevailing market wage rate cannot find employment). Efforts to explain such failures have centered on two major theoretical possibilities: market failure rooted in the purposeful decisions of optimizing agents and failure as a consequence of government intervention.

The former possibility (failure stemming from the actions of optimizing agents) has received considerable research attention. To begin with, there is the obvious potential for a “menu costs” type situation: Firms may resist adjusting wages to a new equilibrium level if the transaction costs associated with doing so are sufficiently high. Another possibility that has attracted more interest is commonly referred to as “efficiency wages” (e.g., Akerlof and Yellen [1986]). This is the idea that wages might be set above market equilibrium for the purpose of inducing higher worker productivity and efficiency.

Several specific motivations for efficiency wages have been suggested. First, they might reduce shirking and increase the financial penalty associated with termination. Shapiro and Stiglitz (1984) demonstrate that in the face of limited monitoring resources firms are willing to pay efficiency wages for these reasons. Efficiency wages might also lessen turnover costs by rendering workers less likely to quit their jobs (more experienced workers tend to be more productive than new employees, whose training can also be costly). Particularly in the setting of lower income countries, higher wages can improve the health and overall well-being of workers (through greater food and health consumption), potentially raising their productivity. Finally, efficiency wages can help firms avoid adverse selection by encouraging more skilled (and thus more productive) workers to apply for jobs.

Another potential source of market failure comes from the timing of wage contracts. In essence, contracts offer a specified wage payment over some fixed interval of time during which there might be shifts in labor supply or demand (and hence in the fundamental equilibrium wage that would prevail in an unfettered market). However, such contractual agreements can prevent adjustment in wages toward these new equilibrium conditions (a situation referred to as “wage stickiness”).

State intervention is another major reason that equilibrium conditions in the labor market may fail to obtain. For instance, a legally mandated minimum wage (a kind of price floor) may result in labor market disequilibrium if that minimum wage level is binding in the sense that it exceeds the equilibrium level that would emerge in its absence. Most empirical and theoretical work on the subject suggests that minimum wages that are binding in this sense yield, ceteris paribus, lower employment levels. However, a few studies (e.g., Card and Krueger [1994]) have received a great deal of attention for reaching the opposite conclusion. Card and Krueger (1994) examine the effects of an April 1992 minimum wage increase (from $4.25 to $5.05 an hour) by exploiting the fact that the increase was implemented earlier in some states than others. Employment levels before and after the change showed no signs of the decline anticipated in the face of an increase in the minimum wage: Employment levels actually increased. However, Card and Krueger (1994) have received a great deal of methodological criticism.

SEE ALSO Becker, Gary; Employment; Labor Demand; Labor Force Participation; Labor Market; Labor Supply; Signals

BIBLIOGRAPHY

Akerlof, George A., and Janet L. Yellen, eds. 1986. Efficiency Wage Models of the Labor Market. Cambridge, U.K. and New York: Cambridge University Press.

Becker, Gary. 1964. Human Capital: A Theoretical and Empirical Analysis, with Special Reference to Education. New York: National Bureau of Economic Research.

Borjas, George. 2003. The Labor Demand Curve is Downward Sloping: Reexamining the Impact of Immigration on the Labor Market. Quarterly Journal of Economics 118 (4): 1335–1374.

Card, David, and Alan Krueger. 1994. Minimum Wages and Unemployment: A Case Study of the Fast Food Industry in New Jersey and Pennsylvania. American Economic Review 84 (4): 772–793.

Katz, Lawrence F., and David H. Autor. 1999. Changes in the Wage Structure and Earnings Inequality. In Handbook of Labor Economics, vol. 3a, eds. Orley Ashenfelter and David Card, 1463–1555. Amsterdam: North-Holland. http://economics.harvard.edu/faculty/katz/papers.html.

LaLonde, Robert, and Robert Topel. 1991. Labor Market Adjustments to Increased Immigration. In Immigration, Trade, and the Labor Market, 167–199. Eds. John Abowd and Richard Freeman. Chicago: University of Chicago Press.

Mincer, Jacob. 1974. Schooling, Experience, and Earnings. New York: National Bureau of Economic Research.

Pritchett, Lant. 2001. Where Has All the Education Gone? The World Bank Economic Review 15 (3): 367–391.

Shapiro, Carl, and Joseph Stiglitz. 1984. Equilibrium Unemployment as a Worker Discipline Device. American Economic Review 74 (3): 433–444.

Spence, A. Michael. 1974. Market Signaling: Informational Transfer in Hiring and Related Screening Processes. Cambridge, MA: Harvard University Press.

Mai Noguchi Hubbard

Wages

What It Means

A wage is the payment that a worker or employee receives for his or her labor. A wage is commonly in the form of money, but it can also be in goods. A wage is usually paid for a specified quantity of labor, which most often is measured as a unit of time. For example, it is common for a worker to make a certain dollar amount for each hour, day, or week the worker provides labor to an employer. The wage differs from another form of monetary compensation for work, the salary, which typically is not based on a specified quantity of labor and is usually paid on a weekly, biweekly, or monthly basis.

The United States government determines a federal minimum wage, which is the lowest dollar amount per hour of labor that an employer is allowed to pay. In 2007 a new minimum wage law was enacted, which specified that the minimum wage rate of $5.15 per hour in force since 1997 would be raised in three stages, increasing to $5.85 per hour in July 2007 and peaking at $7.25 per hour in mid-2009. An employer may choose to pay a worker the minimum rate or an amount above it. The rate of the wage for a particular job is sometimes the focus of negotiations between a worker and an employer before the job agreement is completed.

The wage rates paid in the United States are determined mainly by market forces, that is, the supply of, and the demand for, goods and services. The economic systems in other countries determine wage rates according to different criteria. In Japan, for example, cultural tradition and social structure influence wage rates more than do market forces. Beginning in the last decades of the twentieth century, some developing countries that formerly had provided the developed world with goods and services produced with extremely cheap labor have established minimum wage rates that must be paid to workers.

When Did It Begin

Wages paid as compensation for labor have been in existence for many centuries. When economic systems were more agriculturally based, the wage for farm labor commonly came in the form of a portion of the farm’s harvest rather than as money. The oldest definitions of wage encompass more than monetary compensation for work. Even among contemporary economists, the satisfaction that workers may gain from their labor is considered a type of wage.

The federal minimum wage rate in the United States was established in 1938. It was 25 cents per hour at that time, and it provided a higher pay for a large part of the country’s workforce. The minimum rate, set by the U.S. Congress, has increased numerous times since then, though typically within a political climate influenced by workers who wanted higher pay and employers who wanted to contain business expenses. In 2007 the federal minimum wage rate was $5.85 per hour, but there were higher minimum wage rates in 20 states, such as Illinois, where it was $7.50 per hour. Unfortunately the prices of goods and services rose between 1938 and 2007 as well. When adjusted for this inflation, the purchasing power (the value of money measured by the items it can buy) of the minimum wage was highest in 1968.

More Detailed Information

Many American workers who are paid hourly wages are members of labor unions, which are legal, formally organized groups of workers who gather to represent their collective views on wages, hours, and working conditions. Wage rates in the United States are sometimes set by a negotiation, known as collective bargaining, between union representatives and company leaders. The Fair Labor Standards Act of 1938, the law that established the first national minimum wage in the United States, also guaranteed time-and-a-half pay (compensation amounting to 50 percent more than the standard wage rate) for overtime (working time above the agreed-upon amount per day, week, or month) in certain jobs.

A living wage in the United States is the wage a full-time worker would need to earn to support a healthy family. In most instances the requirements of a living wage are above those of a minimum wage. Living wages are usually defined by local ordinances (laws of a city or county). Living-wage ordinances cover specific groups of workers, typically those employed by businesses that supply goods or services to city or county governments. The city of Milwaukee, Wisconsin, for example, had a living-wage ordinance in 2007 mandating $6.25 per hour, and the city of Santa Cruz, California, had one mandating $12 per hour.

Employers are sometimes required by federal or state governments to follow a wage rate known as the prevailing wage. This is the hourly wage, including benefits and overtime pay, paid in the largest city in each county of a state to the largest group of workers, laborers, and mechanics. The U.S. Department of Labor and Industries establishes prevailing wages for each trade and occupation (for instance, electricians, ironworkers, or carpenters) that are part of the labor force used to complete public works (construction or engineering projects that a state government puts in place for local communities).

Recent Trends

The wage gap is an indicator used in the United States since the mid-twentieth century to illustrate the difference between women’s and men’s earnings at a given time. It also has been used to compare earnings of other ethnicities to those of white males, since white males form a group that typically is not subject to sex- or race-based discrimination. The wage gap is expressed as a percentage; for comparing women and men, it is calculated by dividing the median annual earnings for women by that for men. For example, women in 2006 earned 77 percent of what men earned, as they had each year for about a decade previous.

The Equal Pay Act, signed into law in 1963, made it illegal for American employers to pay unequal wages to men and women who hold the same job and do the same work. Nevertheless, a wage gap between male and female workers has persisted.

Wages

Sources

Rations. Wage payments in the Old Kingdom (circa 2675-2130 b.c.e.) are known from the Abu Sir Papyri written during that period. In the Middle Kingdom (circa 1980-1630 b.c.e.) there are temple documents, biographies, and archaeological data that include information about wages. New Kingdom (circa 1539-1075 b.c.e.) wages are known from Deir el Medina and from documents pertaining to shipping. All of these sources show that wage payments were made in rations of bread, beer, grain, meat, and cloth that were the necessities of life.

Bread and Beer. Most often rations were expressed in units of bread and beer, the two staples of an Egyptian diet. It seems likely that the lower salaries, which were close to subsistence level, were actually paid in bread and beer. Just as modern coins are guaranteed to contain standard amounts of metal, each loaf of bread was baked from a standard recipe, using equal amounts of ingredients, and had a standard nutritional value.

Baking Value. Standardization was assured through a system called pefsu, translated as “baking value.” The employer could use pefsu to ensure that a predictable number of loaves would result from a known amount of grain. The baking value was based on the number of loaves or beer jars produced from a set measure of grain. The higher the value, the smaller the loaves, or the weaker the beer, the

smaller the jars. Most wage lists seem to assume that a standard pefsu has been used in baking and brewing.

Tokens or Tallies. Standardization could also be assured through the use of tokens or tallies. During the Middle Kingdom at Fort Uronarti in Nubia, ceramic tallies have been discovered in the shape of a standard loaf of bread. Presumably this tally could be used to check that a worker’s wages in bread loaves were all the proper size. Beer jars were also presumably a roughly standard size.

Basic Wage. The standard basic wage was ten loaves of bread and one-third to two full jugs of beer per day. Egyptian beer was much less alcoholic than modern brews and higher in caloric content. This ration was for the lowest-paid staff members. Others were paid in multiples of this standard wage, varying from two to fifty times the standard wage for very highly paid people. Various methods could be used for apportioning wages. For example, documentation exists for a particular ship’s crew where the captain and other officials received twice the ration of the ordinary sailors. There is another case where the highest-paid official received 38 1/3 loaves while the lowest-paid worker received 1 1/3 loaves.

Temple Day. In another example from the Middle Kingdom, it appears that the staff of a temple received a commission on all the goods that came to the temple. One inscription describes this occurrence as a “temple day.”

As for a temple day, it is 1/360^th part of a year. Now, you shall divide everything that enters this temple-bread, beer, and meat-by way of the daily rate. That is, it is going to be 1/360^th of the bread, the beer, and of everything that enters this temple for (any) one of these temple days that I have given you.

In this temple the regular staff received 2/360ths of the total revenue of the temple while the chief priest received 4/360ths.

Keeping Accounts. In another case from the Middle Kingdom an expedition leader received five hundred loaves a day as his “ration.” Clearly, large sums like this were not paid out in actual loaves of bread or jars of beer. It seems highly unlikely that an expedition leader could actually take his ever-increasing number of loaves of bread— fifteen thousand loaves after a month—with him on an extended trip into the desert. It also seems impossible that he could eat this much, even with a large family and servants to support. Thus, it seems possible that five hundred loaves of bread is actually a unit for measuring out commodities, approximating the modern idea of a unit of money. It must also have been possible to save and to draw against an account of bread and beer.

Deir el Medina. The New Kingdom craftsmen at Deir el Medina received all the necessities of life from their employer. Their houses were owned by the state. Food and clothing rations were given to them as well as most of the other necessities including water, fuel for their ovens, and the tools they needed to perform their duties. Yet, the robust trade among the workers demonstrates that they required additional goods and services that the state did not provide.

Sources

Gae Callender, “The Middle Kingdom Renaissance (c.2055-1650 BC),” in The Oxford History of Ancient Egypt, edited by Ian Shaw (Oxford: Oxford University Press, 2000), pp. 148–183.

Barry J. Kemp, Ancient Egypt: Anatomy of a Civilization (London & New York: Routledge, 1991).

Alan B. Lloyd, “The Late Period, 664-323 BC,” in Ancient Egypt: A Social History, edited by Bruce Trigger and others (Cambridge: Cambridge University Press, 1983), pp. 279–348.