Labor Unions
Labor Unions
I. Theories of the Labor MovementMark Perlman
II. Labor Movements and Collective Bargaining in EuropeAdolf Sturmthal
III. The Structure of Unions in the United StatesPhilip Taft
IV. Influence on WagesH. Gregg Lewis
I. THEORIES OF THE LABOR MOVEMENT
There are three principal groups currently developing theories of unionism. The communist group, primarily located in the Soviet Union, is concerned with how unions can increase productivity in communist societies. To them unions are workers’ morale rale agencies, and it seems irrelevant whether the state or the plant management (in loco employer) dominates the organization.
Most Western neoclassical economists who are concerned with unions focus on the effects that worker-controlled nonopolies will have on factor prices, levels of employment, labor mobility, technological change, and profits. Hence they view unions as logical constructs, units intent on maximizing gains for all or only some of their members —much as a business firm (also a construct) is supposed to operate. They are not usually interested in exploring the forces that bring unions into being and keep them operating effectively.
The third group includes most of the well-known Western labor union theorists. They attempt to ex-plain workers’ historical behavior regarding wages, hours, working conditions, and job rights by answering all or most of four questions: (1) Why did unions come into existence? (2) How do unions choose their goals and tactics? (3) What explains varying individual worker attitudes toward unions? (4) What is the impact of national cultural and legal experience on worker organizations? In other words, these writers study the historical records of institutions.
Each of these three groups is discussed separately below—the third group first because chronologically it first developed the bulk of labor union theory.
The Western historical tradition
Originally the central questions about unions were why and how socially inferior bargaining groups managed to redistribute wealth and gain recognition. Answers were first furnished by theologians and philosophers, historians and economists. Later, as the newer disciplines of psychology and sociology developed, they, too, influenced the analysis of unionism. All had in common an interest in the dynamic nature of workers’ group identification and organization. I have discussed the relationship of the disciplines to the theories more fully elsewhere (M. Perlman 1958).
The revolutionaries
In the western European tradition, interest in union theory stemmed originally from the study of Marx’s law of historical materialism. From it was fashioned a theory of revolutionary unionism which, in its many forms, is based on a supposedly scientific understanding of the historical process. Technological advances change the relationships between the bourgeois or property-holding classes and the working or propertyless classes. Inevitably the bourgeoisie, motivated by profits, exploits the workers, and workers frequently retaliate by forming unions. Thus, the proximate cause of class conflict is bourgeois exploitation of the workers, which in turn is caused by technological change.
What a specific union does in a given situation depends upon its leaders’ appreciation of the historical process. Unions that further the revolutionary ends of the working class are considered to be acting wisely. Unions that accede to the demands of bourgeois employers or that temporize with their revolutionary purpose, perhaps because of mis-leadership, are thought to be useless and undeserving of historical sanction. Only the socialist theorist, who can comprehend the full historical consequences of a given act, can be trusted to judge what is good and bad. These theorists consider any union leader who tries to make peace with a bourgeois employer to be a “faker” or class traitor.
According to this philosophy, unionism is a useful device in the workers’ class war against the bourgeoisie. But when a socialist society is achieved, the role of unionism is indistinct. Some Marxist writers believe that the workers’ organizations should not materially change their personalities. Others, discussed below, believe that a whole new set of objectives is essential.
Two discordant elements of the revolutionary theory stand out: its teleological nature and its dependence upon “proper” understanding of the historical process. The former gives certainty and confidence in the future. The latter creates dissension among the theorists since any human action or program, regardless of the alleged motives involved, encourages speculations about the “real” consequences. Advocates of the revolutionary theory have therefore each tended to give his own interpretation of the historical meaning of any action. These writers can be divided into three general categories: the right wing, the left wing, and the anarcho-syndicalists (M. Perlman 1958, chapter 3).
The right-wing theorists embrace political gradualism, like Bernsteinian revisionism. They agree that unions should represent the class interest of all workers (including particularly the less skilled) and should keep a steady economic pressure on employers for improved benefits and for the peaceful nationalization of industry. They advocate industrial unionism on the economic front (unions organized by industries or by geographic areas) and a labor party embracing gradualism on the political front. Where unions have become craft-particularistic, right-wing theorists urge constant education or “boring from within” to reorient them to their true historical purpose. The best-known American theorists in this tradition were such American Federation of Labor (AFL) socialist-pragmatists as Max Hayes, Nahum I. Stone, Victor Berger, and Norman J. Ware. All agreed that in the American context, craft-particularistic unions were ideologically misled, but nevertheless historically legitimate, representatives of the working class.
The left-wing theorists saw unions sometimes as the “protecting shield” and sometimes as the “avenging sword” of the working class. To them the particularistic craft unions, which they believed all AFL constituent organizations to be, were traitors to the working class and to thedeus ex machina of history. Despairing of changing the AFL unions, this group formed “dual unions.” Among the American leaders in this tradition were Daniel De Leon and William Z. Foster; both advocated political action and urged the development of aggressive, socialist-dominated unions as the main revolutionary instrument in bourgeois society.
The third or anarcho-syndicalist subcategory is often confused with the left-wing revolutionaries because both groups urged violence in bourgeois societies. Although both opposed craft-particularistic unions, they differed in their interpretation of the teleological nature of Marxism and of historical causation. The anarcho-syndicalists, lacking faith in the inevitability of the historical process, urged sabotage, duplicity in dealing with employers and public officials, assassination, and physical violence to bring about the disorder which they believed essential for the development of true freedom. The best-known American theorists in this group were William Trautmann, Vincent St. John, and William D. Hay wood.
The impact of such revolutionary theorizing has not been great in the United States. Many of the egalitarian and social welfare programs advocated by the revolutionaries, particularly the right wing, have become part of American public policy. However, these changes were not brought about by the realignment of class structure which the revolutionaries considered essential to the fruition of their plans.
The historical empiricists
Although the Marxians raised questions providing the most intriguing discussion, Sidney and Beatrice Webb used analytical methods that were the most widely admired. They went beyond the old Marxian question of why history should be dominated by worker organizations and asked how British workers had historically formed their organizations. The Webbs’s theory involved a new social science method, combining the patient fact collection of the German historical school with cautious, bit-at-a-time analysis. Their labor union theory was based on British labor history but goes beyond the recording of its facts to present generalizations. Since the Webbs were among the earliest British Fabians, it is not surprising to discover that their intellectual method incorporated political norms.
The Webbs noted that British unionism had, over time, developed several doctrines, each in response both to the workers’ desire to form their own continuous associations for improving wages, hours, and working conditions and to the key institutional features of a particular time (1897). Initially, when unions were small, weak, and at the mercy of employers and unsympathetic judges, workers embraced the self-interested doctrine of vested interests. They used the “method of mutual insurance,” implying mutual help within the workers’ group according to the self-insurance principle. The group was kept small in order to enhance job opportunity, and just large enough to spread the risk.
Later, as unions became stronger, successful attempts were made at bargaining with employers. In these instances the doctrine of vested interest was replaced by a market-price oriented “doctrine of [equilibration of] supply and demand.” The method of mutual insurance gave way to a newer “method of collective bargaining.” This was a bi-lateral arrangement which postulated worker solidarity but which could produce the flexibility sought by workers intent on improving their collective lot. The great shortcoming of this type of unionism was its parochialism—it gave benefits to the strong, that is, to those needing help the least.
Ultimately, the Webbs believed, unions would turn to a third, broader theory, namely a “doctrine of the living wage,” applying to all workers, not just to the strong among them (S. Webb 1918). This doctrine would be presented through the “method of legal enactment,” which was national in scope and parliamentary in means.
The Webbs favored the third doctrine as well as the third method. Indeed, so strong was their endorsement of these that there is a danger of assuming that they deprecated the others. On the contrary, they implicitly noted the usefulness of both the other methods and at least the doctrine of supply and demand. Even in a socialist state there would be a need for workers in advanced and profitable industries to bargain bilaterally with employers so that new and higher standards could be set where economically possible.
The burden of the Webbs’s normativism fell on their moral condemnation of what they considered to be an all-too-frequent union device, restriction of numbers—restricted entry into the trade and the union (S. Webb 1919). They condemned it as parochial, selfish, and inefficient. Rather, they endorsed another device, “the common rule,” a minimum working standard.
It is hard to overemphasize the impact of the Webbs’s work. It was soundly based on fact, where the earlier Marxian thinking was conjectural history at best. It advocated reasonable evolution where some Marxists suggested violent revolution. And most important, it did not present a teleological face, which practical men distrust.
The Webbs have had severe critics. Some have pointed out the influence of their inarticulated norms, namely a moral commitment to social efficiency (S. Perlman 1928); others have noted their anticapitalistic (anti-free-market) orientation (Cummings 1899); and some have even seen a coloring of history in their selection of episodes. Nevertheless, their monumental studies provided the groundwork for most of the later labor union theory. [SeeWebb, Sidney and Beatrice.]
Although there were several earlier American theorists of the labor movement, the impact of the Webbs’s work on John R. Commons and his subsequent reformulation of it mark the beginning of the most important American school. Using the Webbs’s research techniques, Commons and his students at the University of Wisconsin fashioned a general Spiethoffian theory of the development of the labor movement. It is built in large part on extending the concept of private property to include job rights and on analysis of the effects of geographical expansion of the product market on factor-market (particularly wage) relationships.
Commons believed that unionism was a protective device for social and economic inferiors, who had to be protected from the impersonal rigorousness of the market. He took care to point out that this impersonality was caused by the growth of the market (leading to transactions between strangers) and by technological change. Commons did not believe in natural harmony; unlike the Webbs he had little hope for the development of social consensus on economic matters. Rather, he put his faith in the functioning of countervailing large groups, which he thought would be able to fashion acceptable modi vivendi with good faith and a modicum of luck. [See the biography ofCommons.]
Contemporary theorists
The best-known (and the most controversial) theory of labor unions was presented by Selig Perlman, who was Commons’ student and collaborator and the first holder of the John R. Commons chair in economic research at the University of Wisconsin (where both Commons and he did most of their academic work). Perlman, a Russian-Jewishemigre, had in his youth been a Social DemocraticBundist (a Jewish right-wing Marxian socialist). Commons’ influence on his thinking led him to shed his earlier Marxian conclusions. In the Commons–Perlman approach, unions are considered to have an importance in the western European democratic social structure quite apart from their ability to raise wages and improve working conditions. Unionism is the technique by which workers band together to secure social and industrial rights. Having secured these rights, workers use unions to protect themselves from the effects of market competition, which through history has been intensified by the growth of markets (bringing once separate communities into direct economic competition) and changes in production techniques (resulting in competition among the workers in any one area). Thus unionism has both an offensive and a defensive characteristic, but invariably it was intended to promote the social interests of its membership rather than the needs of an abstract “working class.” The Commons-Perlman approach to unionism devolved from a study of historical records showing how specific groups of workingmen sought to protect their jobs. These records were found in court trials for conspiracy and in the development of job customs or working rules in particular trades and industries.
According to Commons and Perlman, the older a union gets the more it tends to draw on its own historical experience. It tends to rely upon the “common will” as interpreted by union leaders. Consequently, unions are reluctant to follow the advice, however well intentioned, of outside sympathizers and experts. Unions generally attempt hegemony over only a limited job territory, having discovered by experience what limits are most effective. In Anglo-American countries, unions pursue policies intended to create property rights in job possession similar to those already guaranteed to ownership of real estate and chattels.
Although there are some differences between Commons’ and Perlman’s ideas, it is erroneous to think that the two men were ever in conflict. However, it must be realized that Commons, unlike Perlman, was not particularly interested in developing a general theory of unionism.
Perlman’s theory drew heavily on the historical generalizations of Max Weber and Werner Som-bart. From Sombart he took an appreciation of the role of psychology in explaining both the Zeitgeist and the individual’s reactions to changes in economic opportunities.
Perlman took care to define the confines of worker group loyalty because he did not believe that in the American context there was any “class” loyalty as such. Rather, the ambit of such loyalty was set by the manual workers’ innate fear of factor-market competition for the job. Perlman’s phrase “dynamic job-consciousness” sought to ex-plain the changes made by technology and market expansion in a once tightly defined craft concept of job ownership. The “manual worker” (an ideal type in the Weberian tradition) was the key to Perlman’s theory of unionism. These workers, he felt, were willing to trade a purported freedom of contract for a small share of the collective voice controlling job ownership because they had little faith in their ability to survive in the junglelike factor market in which they sold their services. Perlman emphasized that the diffusion of wealth, and unfavorable worker experience with industrial self-government, made American workers and labor movements reluctant to embrace abstract social experiments like socialism (S. Perlman 1928). Later in his life, after the Nazi holocaust, Perlman added that even mature labor movements appeared unable to resist highly nationalistic leaders; indeed, they did not always try. At no time was he sanguine about colonial labor movements’ abilities to resist the appeals of nationalistic or socialistic reformers. [See the biography ofPerlman.]
The traditional Marxians, the Webbs, Commons, and Perlman had much in common in their labor movement theories. Each tried to explain the development of the labor movements of western Europe, North America, and Australasia. Where Marx had first asked “why organization?”, the Webbs added “how does organization work?” Commons and, particularly, Perlman reformulated the basic questions to “under what conditions will a labor movement develop, and what specific rational and historical factors shape its development?”
There have been several vigorous criticisms of the Wisconsin theory. Some have disliked its selection of the parochial, American ideal-type union, with its limited social objectives, arguing that Perlman “liked” or “approved” of that type of union’s particularism (see Hardman 1950; Laski 1948, pp. 221–222; 1950, pp. 37–40; Sturmthal 1953; Gulick 1948, vol. 1, pp. 294–297, 300 ff.; Gulick & Bers 1953). Others have felt that the theory was applicable only to nonexpanding industries (Meyer & Conrad 1957) or to already industrialized and capitalistic societies (Kerr & Siegel 1955).
Criticism from the last point of view has resulted in several interesting discussions, most of which are not presented as formal theories. Two of these were written by John T. Dunlop. The first (1948) summarized both the questions earlier theorists had discussed and what he thought were the shortcomings of their work. He argued that individual theorists had neglected some of the dynamic economic roles in the development of the labor movement played by changing technology, product-market and factor-market forces, public opinion, and the government. Subsequently Dunlop (1958) further clarified his dissatisfaction with labor movement theory. The key to this analysis is an interest in the process by which production and distribution decisions are made by firms within society and by society itself (cf. Moore 1960). He relegated the labor movement to a role somewhat different from the one that had become traditional. Unlike earlier writers who viewed the labor movement partly as a political and social force, Dunlop directed his attention exclusively to the processes of decision making in the allocation of labor. Consequently, unionism in his framework varies in importance among countries and requires a subtheory of labor movements. Where unions play an important industrial role (for example, in bipartite collective bargaining), their significance is great; where they play only a political role, their significance is less.
Communist tradition
At the turn of the century, in response to public opinion, the tsarist government attempted to set up “legal unions,” which anticipated both Italian fascist unions and American company unions. As substitutes for genuine unionism these organizations were not successful, and in 1905 the Soviet of the Workers Deputies of St. Petersburg was founded. In October of 1905, at the time of the granting of a constitution, the first Russian Trade Union Conference was held, made up of 26 Moscow unions and 10 from other localities. With the failure of the Russian revolution of 1905, the reactionary government reasserted its authority, and in 1907 the police claimed to have disbanded 107 unions and exiled or imprisoned their leaders.
In March 1917 the Petrograd soviet reappeared in the central role and, together with revolutionary soldiers, declared a strike. The following month 82 such local Soviets were represented at the first Soviet Conference meeting in Petrograd. Two months later a Russian Trade Union Conference was held whose 220 delegates represented a total membership of nearly 1.5 million workers. Within the organization there was a strong Menshevik, or Social Democratic, opposition. Bolshevik leadership was far from assured of support from the renascent older trade unions. From this distrust of the older group emerged both the Leninist policy toward unions and, later, the Leninist theory of unions. Lenin made an important speech in January 1919. His views were later circulated in 1921 and appeared as “The Task of the Trade Unions.” Lenin was concerned with the deterioration of factory equipment during World War I, which aggravated the country’s previous backwardness. He sought to improve production facilities and wanted the unions to abandon all policies which directly or indirectly hindered production. In both his 1919 speech and his reformulated views, he endorsed Taylorism, or scientific management.
Taylorism as Lenin knew it gave the workers virtually no voice in the handling of industrial problems. Nonetheless, Lenin sought to provide a place for union activity in the general framework of a scientifically managed industrial system. Trade unions were to discipline the workers, on the one hand, and be a quasi-public watchdog over managerial efficiency, on the other. The union was to minimize production loss and eliminate discrimination against particular workers, rather than improve wages generally or even improve working conditions for the workers in a given plant or area.
After Lenin’s death the government union system, which Lenin originally urged, was taken up by Leon Trotsky. Trotsky wanted the state to control and eventually absorb all unions. He was opposed by Michael Tomsky and the functional leaders of the trade unions. Up to the time of his dismissal as head of the trade union’s executive committee in 1929, Tomsky held strongly that the unions should occupy an independent place in the development of industrial policy. A third or middle view, popular within the party itself, held that the union should be apparently independent but actually controlled by party cells within it.
With the gradual sidetracking of Trotsky following Lenin’s death, the first view ceased to be an important issue. The second view, Tomsky’s, obtained until the New Economic Policy (NEP) yielded to the first five-year plan. Thereafter the third, or party, plan became official.
Under the NEP there was a tripartite control arrangement with three supposedly equal parties: the plant manager, representatives of the local unions, and representatives of the party. The manager was usually chosen by the party, and the leaders of the local unions were chosen by popular acclamation—which in practice also meant by the party. This troika arrangement continued for some time. Yet it is clear that by 1930 the unions had ceased to count; plant managers were in fact in unchallenged charge of all plant activity, and the trade unions had lost their voice in policy determi-nation. By World War II the unions themselves seemed to have become primarily interested in increasing productivity through the Stakhanovite and the Udarniki movements. They were also concerned with providing rest resorts for deserving workers and the processing of a few types of grievances against unpopular plant managers. The best-known theorists of Soviet unionism have been Lenin and A. Lozovskii, who became the party’s semiofficial spokesman. [See the biographies ofLeninand Trotsky.]
The neoclassical approach
Although most neoclassical economists have concentrated on the economic effects of wages and of particular union policies, such as strikes and secondary boycotts, several have also written on union power to affect the rate of technological change within the firm and the right of licensure (entry to the trade) [seeLicensing, Occupational; see also Hicks 1932; Friedman 1962]. Until fairly recently the typical neoclassical analysis of unions concentrated upon the conditions under which a union would raise the wages of its membership at the expense of employers or other workers (Simons 1944; Machlup 1952). Now, however, attention has turned to the various alternative goals which unions can maximize. These include wage rates, wage payments, and employment levels as possible goals of the membership as a whole or of a dominant (even if minor) fraction of the membership. Moreover, these models admit a different selection of goals by a given union at different stages of the business cycle and during different phases of a long-term technological development. Different assumptions are also made about the price elasticity of demand for the product, and there has been recognition of the importance of treating each firm within an industry differently for tactical reasons. Finally, the impact of tax provisions, among other factors, leads to different types of wage payments —as in the case of pensions and nontaxable services such as health benefits[seeWages, article onFringe Benefits;see also Dunlop 1944; Cartter 1959; Pen 1950; and Rees 1962].
The recent emphasis on statistical measurement has increased our knowledge of the effects of unionization on wage rates [seeLabor Unions, article oninfluence on wages]. Some economists have held that the gains made by one group of workers have been offset by a rise in prices or by a decline in the earnings or employment of others (Friedman 1962). Other writers, not necessarily rejecting that finding, have emphasized the “possible” result; one “guesses” that the “average effects of all American unions on the wages of their members in recent years will lie somewhere between 10 and 15%” (Rees 1962, p. 79).
One of the best summaries of the neoclassical position concludes simply that the direct effect of unions on the economy is detrimental but that the total social effect of unions (going beyond considerations of prices, wages, and the allocation of resources) turns the conclusion on its head (Rees 1962, pp. 194–195). This judgment, however, merely reflects the dichotomy within the field of Western labor union theory. One group of theorists approaches the topic from the institutional and historical position, while the other approaches it from the standpoint of economic theory.
Mark Perlman
BIBLIOGRAPHY
Alexander, Robert J. 1962 Labor Relations in Argentina, Brazil, and Chile. New York: McGraw-Hill.
Bauder, Russell S. 1943 Three Interpretations of the American Trade Union Movement. Social Forces 22:215–224.
Cartter, Allan M. 1959 Theory of Wages and Employment. Homewood, I11.: Irwin.
Commons, John R. (1899–1913) 1913 Labor and Administration. New York: Macmillan. ? Contains essays and speeches previously published in various publications.
Commons, John R. 1919 Industrial Goodwill. New York: McGraw-Hill.
Croce, Benedetto (1900) 1922 Historical Materialism and the Economics of Karl Marx. London: Allen & Unwin; New York: Macmillan. ? First published as Materialismo storico ed economica marxistica.
Cummings, E. A. 1899 A Collectivist Philosophy of Trade Unionism. Quarterly Journal of Economics 13: 151–186.
[Dridzo, Solomon A.] (1934) 1942 Marx and the Trade Unions, by A. Lozovskii [pseud.]. New York: International Publishers. ? First published as Karl Marks i profsoiuzy.
Dunlop, John T. (1944) 1950 Wage Determination Under Trade Unions. New York: Kelley.
Dunlop, John T. 1948 The Development of Labor Organization: A Theoretical Framework. Pages 163–193 in Richard A. Lester and Joseph Shister (editors), Insights Into Labor Issues. New York: Macmillan.
Dunlop, John T. (1949) 1953 Collective Bargaining: Principles and Cases. Rev. ed. Homewood, I11.: Irwin.
Dunlop, John T. 1958 Industrial Relations Systems. New York: Holt.
Friedman, Milton 1962 Capitalism and Freedom. Univ. of Chicago Press.
Gulick, Charles A. 1948 Austria From Hapsburg to Hitler. 2 vols. Berkeley: Univ. of Calif. Press. ? Volume 1: Labor’s Workshop of Democracy. Volume 2: Fascism’s Subversion of Democracy.
Gulick, Charles A.; and Bers, Melvin K. 1953 Insight and Illusion in Perlman’s Theory of the Labor Movement. Industrial and Labor Relations Review 6:510–531.
Hardman, J. B. S. (1950) 1951 From “Job-consciousness” to Power Accumulation. Pages 146–157 in Industrial Relations Research Association, Proceedings of Third Annual Meeting. Chicago: The Association.
Hicks, John R. (1932) 1964 The Theory of Wages. New York: St. Martins.
Kerr, Clark; and Siegel, Abraham 1955 The Structuring of the Labor Force in Industrial Society: New Dimensions and New Questions. Industrial and Labor Relations Review 8:151–168.
Kerr, Clark et al. 1960 Industrialism and Industrial Man: The Problems of Labor and Management in Economic Growth. Cambridge, Mass.: Harvard Univ. Press. ? A second edition waspublished in paperback in 1964 by the Oxford University Press.
Laski, Harold J. 1948 The American Democracy: A Commentary and an Interpretation. New York: Viking.
Laski, Harold J. 1950 Trade Unions in the New Society. London: Allen & Unwin.
Lenin, Vladimir I. (1902) 1961 What Is to Be Done? Volume 5, pages 347–529 in Vladimir I. Lenin, Collected Works. 4th ed. Moscow: Foreign Languages Publishing House. ? First published as Chto delat’?.
Lenin, Vladimir I. (1919) 1921 Scientific Management and Dictatorship of the Proletariat. Pages 177–199in John R. Commons (editor), Trade Unionism and Labor Problems: Second Series. Boston and New York: Ginn. ? First published in Russian.
Lenin, Vladimir I. (1921) 1945 The Task of the Trade Unions. Volume 23, pages 503–518 in Vladimir I. Lenin, Collected Works. New York: International Publishers. ? First published in Russian.
Machlup, Fritz 1952 The Political Economy of Monopoly: Business,Labor and Government Policies. Baltimore: Johns Hopkins Press.
Meyer, John R.; and Conrad, Alfred H. 1957 Economic Theory, Statistical Inference and Economic History. Journal of Economic History 17:524–544.
Moore, Wilbert E. 1960 Notes for a General Theory of LaborOrganization. Industrial and Labor Relations Review 13:387–397.
Pen, Jan (1950) 1959 The Wage Rate Under Collective Bargaining. Cambridge, Mass.: Harvard Univ. Press. ? First published as De loonvorming in de moderne volkshuishouding.
Perlman, Mark 1958 Labor Union Theories in America: Background and Development. Evanston, 111.: Row, Peterson.
Perlman, Mark 1960 Labor Movement Theories: Past, Present, and Future. Industrial and Labor Relations Review 13:338–348.
Perlman, Selig (1922) 1950 A History of Trade Unionism in the UnitedStates. New York: Kelley.
Perlman, Selig (1928) 1949 The Theory of the Labor Movement. New York: Kelley.
Reder, Melvin W. 1960 Job Scarcity and the Nature of Union Power. Industrial and Labor Relations Review 13:349–362.
Rees, Albert 1962 The Economics of Trade Unions. Univ. of Chicago Press.
Simons, Henry (1944) 1948 Some Reflections on Syndicalism. Pages 121–159 in Henry Simons, Economic-Policy for a Free Society.Univ. of Chicago Press. ? First published in Volume 52, No. 1 of the Journal of Political Economy.
Sturmthal, Adolf 1953 Unity and Diversity in European Labor: AnIntroduction to Contemporary Labor Movements. Glencoe, I11.: FreePress.
Taft, Philip 1952 Theories of the Labor Movement. Pages 1–38 in Industrial Relations Research Association Interpreting the Labor Movement. Madison, Wis.: The Association.
Webb, Sidney 1918 The Works Manager To-day: An Address Prepared for a Series of Private Gatherings of Works Managers. London: Longmans.
Webb, Sidney 1919 National Finance and a Levy on Capital: What the Labour Party Intends. London: FabianSociety.
Webb, Sidney 1920 The Root of Labour Unrest: An Address to Employers and Managers. London: Fabian Society.
Webb, Sidney; and Webb, Beatrice (1897) 1920 Industrial Democracy. New ed. London and New York: Longmans.
II. LABOR MOVEMENTS AND COLLECTIVE BARGAINING IN EUROPE
Trade unionism is a child of industrialization; therefore it is not surprising that the country that pioneered in industrial development, Great Britain, was also among the first to develop modern trade unions. No attempt will be made here to retrace the history of unionism in Great Britain or elsewhere. However, some general considerations regarding the development of unionism are appropriate.
When trade unions first arose in the Western world they were regarded everywhere as criminal organizations; even if their existence was tolerated most of their activities were unlawful. The difficult stage of early industrialization was, therefore, passed by most Western nations without the intervention of effective unions. The British Combination Acts of 1800, the Loi Le Chapelier of 1791 in France, and the fairly general prohibition of unions until late in the nineteenth century in the countries that later formed Germany are examples of this proscription.
Union-political party relations
The growth of unions, since this first stage, has followed widely different courses in different parts of the world. On the Continent, unionism typically developed as part of a wider labor movement, with a political party of the workers usually taking the lead. Frequently the party organized the unions and provided them with leadership. In view of the low educational level of the workers, the extremely long hours of work, and the low incomes on the Continent at the time when unions came into being, it is doubtful whether labor organizations could have been formed at the time without the active participation of political leaders who were, for the most part, intellectuals. Moreover, the effective defense of the workers’ interests at the time encompassed far more than the specialized activities of unions. Throughout the Continent, workers were denied equality of status as citizens, in education, and in the life of the community in general. To meet the demand of the workers for political, social, and cultural equality, as well as to make the unions effective instruments for the attainment of the economic objectives of their members, a wider labor movement was required, of which the unions formed a part. The political organization was most commonly the elite of the movement, and the unions accepted party guidance. It seemed obvious that only a profound transformation of the social and political system would permit an improvement of the worker’s standard of living and the attainment of his full status as a citizen and a human being. This transformation had to be brought about primarily by the party. Moreover, the long-run situation on the labor market made collective bargaining an effective device for skilled workers only. Underemployment in agriculture was constantly being created, or at least was a threat, because of technological advances; and the actual or potential weight of this supply of unskilled labor greatly impaired the bargaining power of unskilled groups. This fact, too, tended to make political action appear relatively more promising than collective bargaining.
Political control of the unions led in many countries to politically divided labor movements, since different political parties established competitive unions. Until World War II the predominant current was socialist; Christian (mostly Roman Catholic), nationalist, democratic, and at times communist organizations represented the main minority groups. In Scandinavia, however, the socialist blue-collar unions remained almost without competition. French syndicalism, primarily representing skilled workers, rejected political control of the unions; indeed, the syndicalists regarded any political activity as futile. But after the decline of syndicalism, the Communist party assumed the leadership of the strongest trade union federation, the Confédération Générate du Travail (CGT).
Starting under political guidance, the unions gradually acquired more status within the labor movement as industrialization progressed. Education was provided for leaders arising in increasing numbers from the working class. Union membership rose and with it union funds and prestige. While retaining their basic political philosophy, the unions claimed and obtained equality of status with the party. The Mannheim agreement of 1906 in Germany is one example of this shift in relative weight. In Great Britain, the evolution proceeded from a different starting point. Unions of skilled workers, rejecting social reform ideas and concentrating on bread and butter issues, arose after the failure of the Chartist reform movement. These “business unions” dominated the movement until attempts at organizing unskilled workers were successful in the last two decades of the century. This introduced socialist—although non-Marxian—ideas into the labor movement. A major court decision, the Taff Vale case, which endangered the effective operations of the unions, greatly increased the unions’ support of the organization of the Labour party. This party was thus essentially a creation of the unions, and for a long time it remained primarily their political instrument—the political branch of the movement. These differences in the origin of unions and party have lost most of their former significance. Different systems of union-party relations remain. In Great Britain and some parts of Scandinavia, unions are collectively affiliated with the Labour party or the Social Democratic party of their country. Union members pay a “political levy” along with their union dues, unless, as they may do in Britain, they “contract out,” i.e., state in writing their refusal to do so. In other countries union leaders are at the same time party leaders; in a formal or informal way, the union federation is represented on the party executive and vice versa. At the other extreme is the tradition of the French “Charter of Amiens”—nowadays often honored in the breach—which, following the syndicalist ideas, requires complete separation of unions and parties.
International labor organizations
The problem of union-party relations has also affected the life of the international organizations of labor. The early international socialist congresses were at the same time international trade union gatherings. Separate international union organizations arose first under the name of international trade secretariats. These were associations of unions of the same craft or industry. The typographical unions were the first to establish their international association in 1889, followed by the miners and others, so that by the turn of the century almost thirty of them were operating. The national trade union centers—federations of unions of different crafts or industries in the same country—held their first independent international gathering in Copenhagen in 1901. They established a small international secretariat, whose administration was entrusted to the German trade unions, then the leading labor movement of the world. The activities, however, were modest and purely technical as the unions continued to accept in policy matters the leadership of the socialist parties and of their international. This aroused the protest first of the syndicalists in the French CGT and later of the American Federation of Labor, which joined the secretariat after an exploratory visit by Samuel Gompers in 1909. Under their combined pressure, the international secretariat was renamed the International Federation of Trade Unions (IFTU) in 1913, and its activities were gradually expanded. This development was interrupted by World War I. When the IFTU was reconstituted in Amsterdam in 1919, the AFL withdrew, primarily because of the “revolutionary principle” to which the organization, in Gompers’ view, was now committed. Instead, the AFL turned increasingly toward collaboration with Latin American unions. Shortly before World War II, however, the AFL rejoined the IFTU, whose activity, soon afterwards, was paralyzed by the advance of Hitlerism. The reorganization of the international labor movement into the World Federation of Trade Unions (WFTU) after the war brought the Russians and the CIO into the movement, while the AFL opposed the WFTU. Sharp conflicts between communists and anticommunists led to the breakup of the WFTU. The immediate issue was the attitude to be taken toward the Mar-shall Plan. The noncommunist unions, including both the AFL and the CIO, formed the International Confederation of Free Trade Unions (ICFTU) in 1949, with headquarters in Brussels. The WFTU has continued to operate, primarily as an international body of the communist-dominated organizations.
Unions of socialist persuasion were, and continue to be, the predominant current in the inter-national labor movement, but they are by no means without competition. Until the 1880s, business unionism set the tone in Great Britain while anarchosyndicalists won out over Marxists in France and Spain and formed an influential group in Italy. In Germany, France, Belgium, and other countries with strong Roman Catholic populations, Christian trade unions arose, often under the inspiration of the papal encyclicalRerum novarum. As a result, divided and competitive trade unions arose in several countries of the Continent. Communist unions came and went after 1919, following the complicated zigzag of the party line, which at times ordered the communists to conquer socialist unions by “boring from within” and at other times to defeat them by setting up communist unions in competition with them. In some countries, such as Germany and Gaullist France, nationalist movements set up their own unions, in general with only modest success except at times among the white-collar workers.
Plant organizations and workers’ councils
British unions, from an early stage, operated in the plant or close to the plant. Being craft unions, they were greatly concerned with “working rules,” which often were determined by mutual agreement at the local or plant level. Still more important in bringing the union into the plant was the necessity for regulating piecework rates, which rarely can be done in a satisfactory way outside the plant. The handling of grievances arising out of the piece rate system and, later, out of the collective agreement as a whole, required a union organization in the plant. Many, although by no means all, British unions perform such functions in the plant, even though their formal organization and their constitutions do not always reflect this orientation.
Entry into the plant was, of course, the result of a long battle against hostile laws and employers. The repeal of the Combination Acts in 1824–1825 had less impact upon union activities than the enactment of labor laws in 1871 (amended in 1876) and 1875. By 1910, as a government report of that year indicated, collective bargaining still concerned only a minority of “industrial workers,” but a substantial part of the bargaining that did occur proceeded on the plant level. When World War I greatly enlarged the area of collective bargaining and caused a shift in emphasis to national agreements, union activity in the plant was already established.
Quite different was the evolution in those countries of the Continent in which competitive unions arose. Compared with England, industrial development in most of these countries was either delayed or slower; and unionism and collective bargaining were less advanced until the revolutionary movement following World War I gave the labor movements in many countries the power to impose union recognition and collective bargaining. Thus for a long period, unions existed without being able to enter the plant itself. The employers’ resistance to union activity within the plant was strengthened by political-religious divisions of the unions. Union operations in the establishment, it was said, would carry ideological disputes into the plant and interfere with its operation.
These differences between the British and the Continental evolution of the movement are to some extent reflected in the organizational structure of the unions. There are, of course, great variations among the unions of different crafts and different industries, as well as among those of different countries. But a frequent pattern on the Continent is represented by the union whose lowest administrative level is geographic, combining union members in several plants or firms. Even where, as in France, a lower level organization has been added informally to the localsyndicat, this group, the so-called section, will usually still cover several plants. The communist CGT has been working to make thesyndicat coincide with the membership in one plant, but so far it has met with only limited success. In Germany, some unions, particularly the metalworkers union (I.G. Metall), are engaged in setting up union stewards in the plants to work with the geographically based local (Ortsverwaltung). British unions, on the other hand, sometimes have the same organizational pattern as American unions and are officially represented at the workplace, although many unions also have geographic branches at their base.
In the absence of well-developed plant-union groups in most countries on the Continent, plant organizations of a different kind have arisen to fill the void. There is a need for bargaining within the plant, in order to interpret the collective agreement, to make it fit the particular conditions of the plant, to supplement it where necessary, and, especially, to handle individual grievances. The organizations that meet these needs go under different names in different countries. Their German designation, translated as “workers’ council,” will serve here as a general term for all of them. They show considerable variations as to their exact functions, election systems, relationships to unions, etc. Some are designed to enable the workers to participate in managerial functions. Characteristically, most of them perform tasks in the plant that in the United States are basic union activities. The council members are elected by all workers, whether union members or not. The councils are consequently not an organic part of the union itself and derive their existence and authority from legislation rather than collective agreements. The industrial relations function that is of most immediate relevance to the individual workers is thus in the hands of a body that does not form part of the union. At times, in some countries, union–council relations have been competitive rather than complementary. In Britain, shop steward movements, particularly after World War I, threatened for a while to turn into rival labor movements.
Union structure
There are few other organizational features that European labor organizations have in common. They vary greatly in their degree of centralization. In most Scandinavian countries, the union federation has a high degree of control over its affiliated In Great Britain, the Trades Union Congress (TUC) has little power under the statutes but considerable moral stature; collective bargaining, however, is fully in the hands of the affiliated unions. The German Union Federation (DGB), organized by Hans Böckler in 1949, has lost a good deal of its influence in favor of its affiliates. In France, the national union has only limited functions. The centers of gravity of the movement are, on the one hand, the (local)syndicats and, on the other hand, the union federations—CGT, the Christian CFTC, the socialist Force Ouvriere (CGT-FO), and the white-collar Confederation Générate des Cadres (CGC). Significantly, it is the syndicats, not the national unions, that are represented at the federation congresses.
In contrast to the United States, the issue of craft versus industrial unions rarely has led to serious internal cleavages. In most countries, both types exist within the same federation. In Germany, where the unions were rebuilt afresh after the Hitler era, a relatively neat organizational layout was established: 16 industrial unions divide the job territory among themselves. One of them, the Metalworkers Union, with close to two million members, represents almost one-third of the entire DGB membership. At the other extreme, Great Britain shows a complex pattern of overlapping organizations ranging in size from a few hundred to more than a million members. Amalgamations have tended to simplify the system somewhat, but in the main it continues to be unplanned. Two great general unions, the Transport and General Workers Union and the National Union of General and Municipal Workers, representing together more than a quarter of the total TUC membership, include among their members large numbers of unskilled workers in various industries. The existence of industrial federations of unions, comparable to some of the departments in the AFL-CIO, is one device by which the untidy pattern has been rendered more workable. Within the trade union movement in Italy, the existence of associations of sharecroppers (mezzadri) and farmers on different lease arrangements adds a confusing feature to the otherwise predominantly industrial organization scheme.
Even where the industrial union is the basic pattern, as in West Germany, it has been impossible to apply the principle without qualification to the white-collar employees. In many European countries, e.g., Sweden, Austria, and France, a considerable number of white-collar workers are unionized, including government employees. In Germany, the organized white-collar workers are distributed over the industrial unions, a special white-collar federation (DAG), and a separate organization of civil servants. While the DAG has relatively friendly relations with the DGB which is predominantly a manual workers’ organization, the highly effective civil servants’ union maintains its independence in line with the social distance that in Germany traditionally separates the civil servant from the public, and especially from the manual worker. The Swedish federation of white-collar unions (TCO) has friendly relations with the general trade union federation (LO) but is politically neutral, while LO maintains intimate contacts with the Social Democratic party. In France, however, the civil servants form the strongest single bloc within the CGT-FO. A separate organization, the CGC, exists for higher white-collar employees and professional workers outside the government.
Practically everywhere government employees are free to organize and to engage in some forms of collective bargaining. Most of them—in some countries even the police—are legally free to engage in strikes. In Britain, the laws enacted after World War II providing for the nationalization of the coal mines, the railroads, and the electricity, gas, and other industries require the governing boards of these industries to engage in collective bargaining with the appropriate unions. Strike restrictions imposed during World War II have been removed.
The unions’ role in management
Nationalization of industry has given rise to prolonged debates about the role of the unions in management. In Great Britain, in spite of the guild socialist tradition of advocating the handing over of industrial management to the unions, the movement came ultimately to the conclusion that union representatives could not serve as such on the management boards without running the risk of conflicting loyalties. The post-World War II nationalization laws thus provide that the management be composed of experts, including industrial relations experts. Trade unionists have been regularly appointed as such experts, but upon appointment they have had to resign their union offices. The French nationalization measures followed a scheme proposed by the Austrian socialist Otto Bauer, who in turn was inspired by the writings of the British guild socialist G. D. H. Cole. According to Bauer, nationalized industries were to be administered by tripartite boards consisting of equal numbers of representatives of the government, of the employees, and of the consumers. This was to ensure labor’s participation in the management of the enterprise, while avoiding the danger that labor would take advantage of monopolistic situations to exploit the public. Most observers agree that neither the British nor the French scheme has given the employees the sense that they are effectively participating in managerial decision making. German codetermination, limited to coal and steel, gives the employees themselves and the union the right to be represented on the supervisory boards of the corporations, together with an equal number of representatives of the stockholders and a jointly elected chairman. In addition, the labor director, one of the three managers of the company, must be appointed with the approval of the union. While this scheme was being discussed, extreme fears were expressed that it would destroy the effectiveness of company management and/or of the union. Events so far have done little to justify these fears. Nor has codetermination had any visible impact on productivity or the degree of the employees’ involvement in the enterprise. Other devices whose purpose is to ensure some measure of participation by workers in management are the workers’ councils, particularly in Germany and France, and the participation of shop stewards and union officials in consultative bodies in Great Britain. While the record of the French and German councils in the administration of plant welfare agencies is praiseworthy, their participation in consultation or decisions on technical and general business matters has so far only rarely been effective.
In the Yugoslav system of workers’ self-government, the councils are to act as managers, although the plant director himself is primarily responsible to the public authorities of the area and only partly responsible to the workers’ council. Under the general political and economic conditions of the country, the council system, which has aroused a good deal of interest, has only limited possibilities of development. When the general administration of a country is of an authoritarian nature, full self-government in the plant, which might create many problems, particularly in countries undergoing rapid industrialization processes, would appear to be impossible to attain.
Collective bargaining
Great Britain has been described as the motherland of collective bargaining; it has been traced back there to the end of the eighteenth century (Webb & Webb 1897). Shortly after the turn of the twentieth century it had become widely accepted in British industry. It has remained the predominant form of determining wages and working conditions, even though the introduction of a general legal obligation to engage in collective bar-gaining after the pattern of the Wagner Act in the United States has been rejected by the unions. An obligation to engage in collective bargaining exists only under the post-World War II statutes providing for the nationalization of certain industries.
On the Continent developments have been slower, in line perhaps with the less rapid progress of industrialization and prolonged legal restrictions upon the existence or the functioning of unions. Significant collective bargaining occurred in France only after World War I (syndicalist refusals to engage in collective bargaining may have contributed to the delay), and even then it was limited to a few years. In 1936, a new trend began during the socialist-led Popular Front regime. This was soon interrupted, however, by World War II. The present system is based upon legislation enacted in 1950. In Germany, the legal restrictions on unionism were reduced in the North German Confederation in 1867, but the antisocialist legislation under Bismarck did not allow collective bargaining to start on a larger scale until the 1890s. Unionism and collective bargaining flourished in the Weimar Republic and resumed after the fall of Hitler.
Yet, apart from England, collective bargaining has never achieved the importance in Europe that it has in the United States. In general, collective bargaining is relatively most important in the United States and Great Britain, of least significance in the newly industrializing nations, with the Continent somewhere in between. In most Continental countries, mixtures of collective bargaining and other methods of wage setting exist that make it difficult, if not impossible, to determine exactly where collective bargaining begins and other methods end. In many of the underdeveloped countries, governmental wage setting is customary, although not always enforceable. Two factors seem particularly relevant in explaining these differences in the importance of collective bargaining: (1) the tradition of regulating social relationships either by contract or by law and administrative action and (2) the long-run situation on the labor market. The first depends upon the continued impact of mercantilistic thinking or, alternatively, the breakthrough of economic liberalism. The second depends upon the long-run existence (or absence) of excess supplies of labor, mainly in the form of underemployment (primarily in agriculture, but also in small-scale crafts and trade.
The British wage councils are one example of such mixtures of bargaining and other forms of wage setting. These were set up, by special legislation, for industries in which inadequate collective bargaining machinery existed and also for some specifically designated industries (Agricultural Wage Act, 1948, 1949; the Road Haulage Act, 1938; and the Catering Wage Act, 1943). Although controlled by a minister in both composition and procedure, the councils operate in a way not very different from collective bargaining. In France, some of the nationalized industries are exempt from the obligations of collective bargaining; wages and working conditions in these industries are determined by a law, astatut. But the working out of thestatut involves a process not dissimilar to collective bargaining. Moreover, French law provides for a government-appointed committee to recommend a minimum wage for all occupations. This process itself is not very different from collective bargaining, as the committee comprises labor and management representatives as well as others. In addition, changes in the minimum wage often exert a direct impact upon the rates set in collective bargaining, since the lowest rates set in many agreements are often close to the legal minimum.
A further complication in estimating the importance of collective bargaining in different countries arises from differences in the function of collective agreements. In many countries, in periods of full employment, effective wage rates exceed contract rates. This phenomenon—labeled “wages drift” in Sweden—must, however, be distinguished from the German system. In the German system, it is expected at all times that the contract rates are merely a minimum upon which will be based effective rates, which are arrived at sometimes in agreements between the workers’ council and the management but more often in individual negotiations. The distance between effective and contract rates varies over time, but in periods of full employment it is significant enough to make union demands for changes in contract rates often appear irrelevant to a substantial portion of the union membership —the portion whose effective rates are above the level the union is aiming at. The labor organizations have tried to meet this problem, which threatens to devalue collective bargaining for the rank and file, by “bringing bargaining closer to the plant,” i.e., concluding union-sponsored agreements for smaller areas and moving contract rates closer to effective rates. Occasionally they have included a so-called effective clause in their contract demands, which stipulates that any improvements granted in contract rates would also be added to effective rates above the contract level. But the absence of effective union organizations at the plant level in most European countries has proven a serious handicap for such developments.
”Extension” of agreements
In a number of countries, among them New Zealand, Australia, Germany, France, and the Netherlands, contracts can under certain circumstances be extended, usually by the government. Extension usually has two attributes: (1) the agreement is given the force of a law and (2) it is made applicable to plants and workers not represented at the bargaining table. The main purpose of extension is to protect the firms operating under the standards established by the agreement against the competition of firms not subject to them. Extension thus reduces the incentive to organize these firms; at the same time it makes essential a shaping of the agreement in such a way that it can be extended according to the prevailing rules. In Germany, for instance, and with a slight modification in the Netherlands, extension is possible only if the employers’ association represents firms employing a majority of the employees in the industry or portion of the industry covered by the contract. No requirements for the coverage of the union are included in the law. Union concern for the extension of the agreement may restrain union wages policy, since excessive demands may induce marginal employers to leave the association, thus depriving it of the membership necessary for the extension of the agreement. In France, extension is limited to contracts concluded by the “most representative unions.” In 1936, when the principle was first introduced into the legislation by the Popular Front government, this term may have been intended to refer to the union representing a majority of the employees in the bargaining unit. This interpretation would have given an effective bargaining monopoly to the then socialist-controlled CGT. Under political pressure this interpretation was abandoned, and the term “most representative union” now refers to all major unions.
The emphasis on the coverage of the employers’ association rather than that of the union tends to reduce labor’s interest in union security clauses of the type found in the United States. However, concern for union strength, financially and in terms of membership, operates in the opposite direction, particularly since the traditional incentive for union membership—class consciousness and a sense of class solidarity—has been waning in the postwar climate of prosperity and diminishing social and political discrimination against workers. European legislation frowns upon such institutions as the closed shop or the union shop. To be lawful, French agree-ments must contain clauses about the freedom of workers to join unions or not to join them. In the Netherlands, the law specifically exempts, from extension, provisions of a collective agreement “which aim at …coercing employers or workers into joining an employers’ or workers’ association … bringing about discrimination in treatment between organized and unorganized parties or persons…” The close relationships between unions and certain political parties tend to make compulsory unionism appear a violation of freedom of conscience and thus to give moral support to the prohibition of union security clauses. However, the agency shop, which requires payment of a fee by the worker to the union to compensate it for representing a nonmember, appears to be legal in certain countries, for example, in Switzerland. Moreover, when unions aspire to union shop clauses, it is commonly understood that the worker can join any bona fide union rather than a specific union. This may go back to the sense of class solidarity, which, at least until recently, meant more to the rank and file than did interunion competition.
Extension tends to reinforce the fairly general emphasis in Europe and other industrial nations outside the United States on large area-wide agreements. One of the implications of this is that the basic agreement must be sufficiently general to be applicable to many different firms in a fairly large area, which is frequently the entire country, as in France and Great Britain, or large parts of it, as in Germany. As a result, the basic agreement is rather brief; essential clauses are worked out at lower levels. Negotiations in the plant, however, and at least portions of the grievance handling are often in the hands of a nonunion organization, namely the workers’ council.
Centralization of bargaining
From the point of view of the degree of centralization of bargaining, the industrialized countries can be grouped into two categories. The Netherlands and Norway are examples of countries with highly centralized wage bargaining. Britain stands for the group characterized by competitive sectional bargaining. This classification relates to the role that the trade union federation plays in relation to the collective bargaining policies of its affiliates. Another possible classification refers to the degree of public control of the wage-setting process. Control may go hand in hand with centralization, since the latter facilitates the former, but logically the two must be kept apart; control is possible, although perhaps difficult to enforce, without centralized bargaining systems. Finally, the term “centralization” might refer to the territorial coverage of the agreement: whether it is national, regional, or local in scope or whether it is a firm or plant agreement. In this sense, most European agreements tend to be far more centralized than most agreements in the United States.
In the Netherlands, central control over bargaining has been in force since the end of the war. The key role in the process is played in fact, although not under the law, by the Labor Foundation. Officially this body merely provides advice on questions of general importance, but this advice has been so consistently accepted that it must be interpreted as almost decisive. In composition the foundation follows the general pattern of bipartite bodies: it consists of equal numbers of representatives of management and the unions. The union representatives come from the three main union federations, the socialistic NVV, the Roman Catholic KAB, and the Protestant CNV. This ranking corresponds to their numerical strength. The NVV has close to 500,000 members, the KAB 400,000, and the CNV more than 200,000. All three endeavor to organize workers in all industries. On the employers’ side a similar division exists, except that the religious organizations are relatively far weaker. The combined coverage of the employers’ organizations, expressed as a percentage of those eligible for membership, is double that of the unions.
Collective bargaining usually begins with discussions in the foundation, which lead to an expression of opinion as to mandatory or permissive changes in the existing agreements. These are then translated into more concrete terms by negotiations on lower levels. All agreements negotiated at the level of the firm or industry must be approved by the central authorities before they become legally valid. This decision is in fact in the hands of the Social and Economic Council, although the council merely advises the government, which has the legal authority to approve or reject agreements through a board of government mediators. The council, consisting of union, employer, and public representatives, seems to follow the recommendations of the foundation rather closely, and the minister in turn almost always accepts the council’s advice. The two elements of centralized bargaining and government control of wage change are thus closely intertwined. The wage structure is regulated by way of centrally determined principles of job evaluation and regional differentials. Since 1962 this system has undergone considerable change; it has been decentralized and controls have been reduced. The main impetus for these changes has come from extreme labor shortages.
Centralization without, or at least with a far lesser degree of, government control than the earlier Dutch system is exemplified by Norway, Sweden, and Denmark. In all three countries union membership comprises about two-thirds or more of all employees. This, together with the small size of the countries, facilitates centralized wage bargaining. The Norwegian system started after World War Ii with compulsory arbitration of contract changes. This was later limited to cases where wage demands had not received the approval of the central labor federation. In 1952, the principle of compulsory arbitration was altogether abandoned. As a result, wage policy is set in negotiations between the trade union federation and the employers’ association. However, the close relationship between the unions and the Labor party, which has been the government for most of the time, has provided for informal mutual coordination. In emergencies, such as the sharp price increases of 1958, the government has not hesitated to invoke compulsory arbitration procedures. The Swedish system, although similar in some respects to that in Norway, has been relying less on compulsory arbitration, which is used, if necessary, to settle disputes about the interpretation of agreements, and more on the internal discipline of employers’ and union federations, whose position is buttressed by their control over substantial funds. Cooperation between the government—led by the Social Democrats—and the socialist-led unions provides a high degree of assurance that union wage policy and government policy run along fairly parallel lines.
A higher degree of decentralization in bargaining has gradually evolved in Germany. While union organization after the fall of the Third Reich began on the local and regional level, the center of gravity of the movement shifted rapidly toward the main trade union federation, the DGB, when it was set up in 1949. Part of the DGB’s authority derived from the fact that whereas in the Weimar Republic the labor movement had been divided into competitive organizations along religious and political lines, the new unions were unified and at first exerted considerable public influence. Essentially this unity has been preserved in spite of the rise of a small Roman Catholic federation, whose main strength is concentrated in the Saar district, and the establishment of separate organizations, one for civil servants and another for part of the white-collar labor force. With the DGB acting as its official spokesman, the movement concentrated its efforts in the first phase of its life toward the attainment of changes in the structure of industrial management and the re-establishment of workers’ representation at the plant level. The first was achieved for coal and steel by the system of codetermination, the second by the Works Constitution Law. The main provisions of this law relate to the establishment of workers’ councils within the plants of private enterprise; later legislation extended this institution to public services. Although during the struggle for passage of these laws the federation took the lead, the later shift of emphasis to collective bargaining brought with it a transfer of power to the 16 industrial unions affiliated with the DGB, since the unions have full power in wage movements. The Metalworkers Union, in particular, has played a key part in bargaining developments. In spite of large membership figures in absolute terms, the DGB unions represent only about 35 per cent of the wage earners. The percentage is constantly declining, and union membership is frequently shortlived so that a considerable recruitment effort is necessary merely to maintain membership at a steady level.
Within the individual unions in Germany there is a high degree of centralization of authority. The full-time district officials, who were at first elected in some unions, are now usually appointed by the national executive committee of the union. This has particular significance since collective bargaining is carried on typically in the districts or for even larger areas and involves the top union officers. The plant organization of the unions, on the other hand, although eagerly promoted, is rather weak, as the workers’ councils, which are elected by all workers, perform the union functions at the plant level.
German employers’ associations—about which relatively little detail is known—are strong and have great discipline. The associations are combined in one central confederation, the BDA. Combativeness, after a long period of social peace devoted to the reconstruction of the country, is growing on both sides. The weapon of the lockout has been employed on a large scale, although it met with only modest success during a metalworkers’ strike in early 1963.
Although agreements are most commonly concluded for large areas, there are few really nation wide agreements. The basic contract is of limited scope, as far as subject matter is concerned, partly because the well-developed social security system and supplementary legislation regulate many fringe issues and partly because only principles are set-tied in this master contract and the details are regulated in agreements with less geographic coverage. Not infrequently special contracts with differing geographic application settle different subject matters. In particular, plant agreements worked out with the assistance of the workers’ council frequently supplement and improve, from the workers’ point of view, the terms of the master contract. Many companies offer voluntary fringe benefits beyond those set by law or by contract. The administration of these voluntary benefits and the supervision of the observation of contracts and laws form one of the main activities of the workers’ council. As the rates set in master contracts or other large-area agreements are usually quite modest in order to facilitate extension of the contracts, great emphasis is placed upon the improvements that plant or personal agreements can bring.
While there may be a consistent wage policy on a national scale on the part of the well-organized and disciplined employers, unions set their policy primarily within the limits of the individual industrial union. Attempts at a national wage policy —for example, a proposal that the government appoint a body of experts to review wage demands —have been without result. The government, however, has intervened rather frequently in wage negotiations by pleas for wage restraints.
British collective bargaining is far more decentralized. Union organization and bargaining systems have grown up over the decades without plan. Written and unwritten agreements and law and custom regulate industrial relations in an almost chaotic fashion. The centra] organizations on both sides—the Trades Union Congress (TUC) and the British Employers’ Confederation (BEC)—have only limited power.
The TUC combines pure craft, amalgamated craft, industrial, and general unions ranging from a few hundred members to more than a million. Some unions are not affiliated with the TUC, some because of legal inhibitions and a few out of choice. With close to ten million members, the movement represents less than half the wage earners. A certain amount of order is established for bargaining purposes by the creation of federations of unions in related industries, as for instance those in shipbuilding or the National Federation of Building Trade Operatives. A slow process of amalgamation of smaller unions has been under way for many years, encouraged by TUC resolutions. The haphazard character of the organizations, the relatively weak position of the TUC, and the tradition of union autonomy in bargaining have pre-vented the movement so far from playing a part in formulating a national wage policy. Yet the idea is gaining ground that such a policy may be required and that the unions should accept their responsibilities within the framework necessary to formulate and implement it.
The experience of the Labour government that came into office in 1945 does not facilitate the task of the partisans of a national wage policy. “Voluntary wage restraint” was inaugurated in 1948 without union participation, and union support was lukewarm from the beginning. Yet in order to support the Labour government, the union drive for higher wages slowed down noticeably for some two years. Contract rates continued to rise, but distinctly less rapidly than prior to the “wage pause.” Under the impact of full employment, however, effective rates increased more rapidly than contract rates. The British unions thus suffered the same experience that the Swedish unions suffered under similar circumstances during the same period. The resulting tension within the movement led the unions in both countries to abandon the policy of voluntary restraint. A new experiment in a government-sponsored “incomes policy” has been under way in Britain since 1961. A balance of payments crisis in 1966 caused the government to establish a temporary wage freeze. A Prices and Incomes Board has been set up whose functions are not quite clear. The role that unions are to play under full employment, particularly in the systems of democratic planning that some countries in Europe are endeavoring to develop, is still uncertain.
The most immediate channel through which the government can exert influence upon wages is its control of the nationalized enterprises. Collective bargaining in many of these industries is subject to government instructions to the managing boards, and the impact of the wage bargains in nationalized industries is considerable since these industries are of great importance in the total economy of Great Britain, France, and several other European countries. Various forms of arbitration are potential levers for the implementation of a wage policy, as arbitration about the terms of a new contract, contrary to U.S. custom, is not infrequent in British industry. This may occur through an existing agreement or, as is more frequently the case, through the Industrial Court, which was established by the Industrial Courts Act of 1919. In fact, however, arbitrators have been more concerned with settling conflicts rapidly than with implementing any particular policy. No long-run solution has yet been found for the problem of fitting free collective bargaining into full employment economies.
Unionism and prosperity
Most observers agree that the prolonged boom in western and central Europe since the early 1950s has been accompanied by substantial changes in the spirit and operations of the unions. Full employment, indeed a labor shortage, rapidly rising real wages, and the continuing decline of social discrimination against workers have led to a transformation of European society. One aspect of this has been a distinct decrease of class consciousness among the workers and considerable modifications of the meaning of the traditional socialist allegiance of the majority of the workers. This has been one factor weakening union-party relations. Another has been the transformation of the structure of the labor force, particularly the growing share of white-collar and professional employees. As a result, union membership has been stagnating and the social democratic and labor parties have tended to change from class parties into people’s parties by de-emphasizing their working-class character.
Within the unions these developments have sharpened the need for stronger representation in the plant and for a shift of the center of gravity of collective bargaining away from union headquarters. At the same time, however, newly developing efforts toward some form of democratic planning extending into the area of personal incomes accentuate the necessity of centralized union leadership that is capable of implementing income policies arrived at by the planning process. The outcome of these contradictory tendencies is uncertain. Some relationship may exist between the trend toward planning and the decreasing use of the strike weapon in western Europe. However, the latter phenomenon also has been observed in countries rejecting the use of planning.
The decline of class consciousness threatens to remove the strongest pillar upon which European unionism has rested in the past. Effectiveness in the performance of the practical tasks of unionism is increasingly the appeal that the labor organizations are holding out to the younger generation. This, however, involves a transformation of spirit, structure, and emphasis, which long-established organizations are often reluctant to undergo.
Adolf Sturmthal
BIBLIOGRAPHY
Bry, Gerhard 1960 Wages in Germany, 1871–1945. Princeton Univ. Press.
Clegg, Hugh A. 1960 A New Approach to Industrial Democracy. Oxford: Blackwell.
Clegg, Hugh A.; Fox, Alan; and Thompson, A. F. 1964 A History of British Trade Unions Since 1889. Vol. 1. Oxford: Clarendon.
Cole, G. D. H. (1925–1927) 1948 A Short History of the British Working-class Movement: 1789–1947. Rev. ed. New York: Macmillan; London: Allen & Unwin.
Edelman, J. Murray; and Fleming, Robben W. 1965 The Politics of Wage-Price Decisions: A Four-country Analysis. Urbana: Univ. of Illinois Press.
Ehrmann, Henry W. 1957 Organized Business in France. Princeton Univ. Press.
Flanders, Allan 1964 The Fawley Productivity Agreements: A Case Study of Management and Collective Bargaining. London: Faber & Faber.
Flanders, Allan; and Clegg, Hugh A. (editors) 1954 The System of Industrial Relations in Great Britain: Its History, Law and Institutions. Oxford: Blackwell.
Galenson, Walter 1949 Labor in Norway. Cambridge, Mass.: Harvard Univ. Press.
Galenson, Walter (editor) 1952aComparative Labor Movements. Englewood Cliffs, N.J.: Prentice-Hall.
Galenson, Walter 1952i>The Danish System of Labor Relations: A Study in Industrial Peace. Cambridge, Mass.: Harvard Univ. Press.
Horowitz, Daniel L. 1963 The Italian Labor Movement. Cambridge, Mass.: Harvard Univ. Press.
Johnston, T. L. 1962 Collective Bargaining in Sweden: A Study of the Labour Market and Its Institutions. Cambridge, Mass.: Harvard Univ. Press.
Kerr, Clark et al. (1960) 1964 Industrialism and Industrial Man: The Problems of Labor and Management in Economic Growth. 2d ed. New York: Oxford Univ. Press.
Knoellinger, Carl E. 1960 Labor in Finland. Cambridge, Mass.: Harvard Univ. Press.
Lorwin, Lewis L. 1953 The International Labor Movement: History, Policies, Outlook. New York: Harper.
Lorwin, Val R. 1954 The French Labor Movement. Cambridge, Mass.: Harvard Univ. Press.
Mccarthy, William E. J. 1964 The Closed Shop in Britain. Berkeley and Los Angeles: Univ. of California Press.
Meyers, Frederic H. 1961 European Coal Mining Unions: Structure and Function. Monograph Series, No. 7. Los Angeles: Univ. of California, Institute of Industrial Relations.
Perlman, Selig (1928) 1949 The Theory of the Labor Movement. New York: Kelley.
Reid, Graham L.; and Robertson, Donald J. (editors) 1965 Fringe Benefits, Labour Costs and Social Security. London: Allen & Unwin.
Roberts, Benjamin C. 1956 Trade Union Government and Administration in Great Britain. Cambridge, Mass.: Harvard Univ. Press.
Ross, Arthur M.; and Hartman, Paul T. 1960 Changing Patterns of Industrial Conflict. New York: Wiley.
Sturmthal, Adolf (1943) 1951 The Tragedy of European Labor: 1918–1939. New York: Columbia Univ. Press.
Sturmthal, Adolf (editor) 1957 Contemporary Collective Bargaining in Seven Countries. Cornell International Industrial and Labor Relations Reports, No. 4. Ithaca, N.Y.: Cornell Univ., Institute of International Industrial and Labor Relations.
Sturmthal, Adolf 1958 The Labor Movement Abroad. Pages 174–205 in Neil W. Chamberlain et al.(editors
A Decade of Industrial Relations Research, 1946–1956. New York: Harper.
Sturmthal, Adolf 1964 Workers Councils: A Study of Workplace Organization on Both Sides of the Iron Curtain. Cambridge, Mass.: Harvard Univ. Press.
Sturmthal, Adolf (editor) 1966 White Collar Trade Unions: Contemporary Developments in Industrialized Societies. Urbana: Univ. of Illinois Press.
Turner, Herbert A. 1962 Trade Union Growth; Structure and Policy: A Comparative Study of the Cotton Unions in England. Univ. of Toronto Press.
Webb, Sidney; and Webb, Beatrice (1897) 1920 Industrial Democracy. New ed. London and New York: Longmans.
III. THE STRUCTURE OF UNIONS IN THE UNITED STATES
The structure of labor unions is the result of a variety of influences Industrial processes, the size and form of the employing unit, the skills, character, and ethnic and national origin of the work force, as well as the time at which the union was first founded, all exercise some influence upon the structure and functions of labor organizations. While one might reduce all unions to a few primary types, many variants have been developed in an effort to adjust the activities of the organizations to the needs of specific groups.
The local. The basic unit of structure and government of labor organizations is the local union. The local takes a number of forms, though all have some characteristics in common. It may be made up of workers of a single craft—bricklayers, plumbers; of employees of an industry—automobile or steel workers; or of a number of occupational groups—but not all those employed in an industry. The International Typographical Union allows only one local to function in a community. Thus Big Six, the New York local, includes in its ranks the more than eleven thousand union printers in New York City. The basic working conditions and problems in the industry are sufficiently uniform so that labor-management issues can be handled by one local. Of course, the size of the business unit and conditions of employment in the newspaper, book, and job branches are different, but these variations may not be significant for collective bargaining or contract administration.
The majority of unions, even those which were originally craft organizations, usually allow more than one local to function in a given community. This practice, however, does not automatically give any group within the organization the right to set up new local units. The basis for multiple local unions is not always clear, especially in organizationsestablished in the early years of the century or before. In some early unions, locals were established on the basis of language. Thus, Swedish or German carpenters, Italian or Puerto Rican dressmakers, and German- or Yiddish-speaking printers might establish local unions which were chartered to allow non-English-speaking groups to conduct union affairs efficiently.
Multiple local unions were also established because the union included more than one type of worker. For example, there are separate local unions of plumbers and steamfitters; linemen and journeyman electricians; garment cutters and operators; carpenters and workers manufacturing doors, trim, and window frames; coal- and milk-wagon drivers; and garage mechanics and machinists. From the beginning it may have been believed that special problems could be better handled if members were segregated on the basis of skill or occupation. Another basis for union organization is residence of members. Here again, the purpose of establishing more than one local in a community is to facilitate union business.
The rise of unionism in manufacturing created new grounds for multiple local organizations. For example, unions operating in the steel, automobile ,and electrical manufacturing industries also recruit workers from a variety of other related industries. Under such conditions local unions may be based upon the employer unit or even a division or department of a single establishment. No common rule for deciding the number of local unions exists, although internationals may seek to prevent the existence of two locals in a particular area. It is not always easy to eliminate locals which have shrunk in size or which might be merged with another without loss of efficiency. Loyalty to a local union or pride ofleadership may be obstacles to mergers. National officers sometimes seek to combine two or more locals to eliminate political opponents, but opposition to mergers arises more generally because of loyalty to the old unit.
The local union transacts its business at periodic meetings and chooses officers to manage its affairs. If the number of members is small, the business of the union will be conducted by unpaid staff members who receive nominal payments and wages for time lost on union business. Locals are governed by their constitutions and bylaws, and if they are affiliated with a national or international union, their rules and policies must conform to the general constitution and bylaws.
The authority and the importance of local unions vary. In some organizations, the local union has almost complete discretion in the negotiation of the terms of employment. There may, at the same time, be some general objectives or standards which the locals are obligated to incorporate in their agreements; for example, the 5–day week or the 35–hour week. The power of the local incollective bargaining is influenced by the type of industry and marketin which it operates, and especially by whether the employing firms cater to a local, regional, or national market. Because their product is nottransferable, unions in the building construction industry have usuallyexercised a great deal of autonomy in collective bargaining and in themanagement of their own affairs. On the other hand, unions that bargain with firms or groups of firms operating in national product markets arelikely to limit more severely the discretion of their local unions inthe bargaining area. The national unions in coal mining, men’s clothing, glass, automobile, steel, and pottery manufacture are among those that exercise a more direct influence over the bargaining of theirlocals than do unions in the service trades and in building construction.
Locals of every union play important roles in carrying out the purposes of the organization, in collecting dues and assessments, inhandling grievances, and in protecting the rights and prerogatives ofmembers. Unless a local is a purely administrative unit, as in the seagoingunions, the local can manage much of its business without interference from higher union authorities. On many issues, however, the local union mustconform to rules and standards developed by the national or international organization. The local is formally a creature of theinternational union, and in many instances locals have arisen from the efforts of organizers employed by the international union.
As a rule, the local holds periodic business meetings at whichexpenditures are approved, appeals for financial or other kinds ofassistance are made, and general issues are discussed under the heading“good and welfare.” Attention has often been called to the lowattendance at local union meetings, the lack of interest shown in union affairs by large numbers of members, and the ability of minorities todetermine policy. The question is whether low attendance is a recent development or a problem that faced unionists in earlier generations. Acertain answer cannot be given, but evidence of an indirect kind doesexist. The fines levied for non-attendance and the rewards given for beingpresent indicate that union meetings were not normally overcrowded in the past, for reasons that are not difficult to perceive. The emphasis uponattendance at local meetings springs from an implicit view that unionsare general organizations for discussing social and economic policies. Few members regard the union from that point of view. The pri-mary functionof the union, from the point of view of the member, is to deal with the terms of employment. Attendance at local meetings will therefore be greateras contract-renewal time approaches, or when serious difficulties arise inthe administration of the contract. Normally, however, the business of theunion is conducted by a few activists— stewards, local officers, andthose deeply concerned with union policy.
To argue that poor attendance at meetings indicates the absence ofdemocracy is to give that term a curious definition. Democracy is a system ofgovernment that provides citizens with an opportunity to influence policydirectly or through elected officers. There are countries that compel voting in elections, but the United States does not. The reasons for lowattendance are complex. The greater dispersion of members makes the gathering of large numbers more difficult than in the past. Members may findthat to attend a union meeting they will have to eat out, reach home late, and miss their customary radio or television entertainment. Although failure to attend meetings may reflect inertia and lack of interest, it might also show satisfaction with the performance of the union.
Under present conditions it is not difficult for large numbers of workers in most industries to withdraw from a union if they aredissatisfied with its policies. While decertification requires effort andeven courage, workers intent upon ridding them-selves of a particular labororganization can usually succeed, providing they can muster a majority. Lowattendance at union meetings is accompanied by willingness to accept thedecisions of the organization, and this is the test of loyalty to theunion.
On the whole, trade unions are among the most democratic institutions in the United States, even though some local unions and larger bodies havebeen guilty of oppression and undemocratic practices. The typical ratherthan the extreme instances must be held in view when making judgments.Investigations of election complaints under the Landrum-Griffin Act haverevealed few violations of the members’ rights. Of the more than 52,000 local unions which held elections between September 14, 1959, and March 31, 1963, investigation disclosed violations in 380, but in only 207 instances might the results have been affected. In a number ofthese instances it is doubtful whether a court of law would overturn theelection were it to public office.
Intermediate bodies
The existence of a number of local unionsin a community has led to the development of such intermediate bodies as the joint or district board or council, a delegate body from a number of local unions of the same organization. These organizations exercise a considerable, and in some unions a predominant, influence over policy. In others, major power is retained by the locals. In the International Ladies Garment Workers Union, the joint board has the primary role in negotiating and administering contracts. The district, as it is called in the International Association of Machinists, is an important delegate body engaging in organization, negotiations, and assistance to local unions, but the locals retain considerable autonomy. District councils are established in a number of the buildingtrades whenever more than a given number of local unions operate in a particular area. In some of these unions the district councils supervise the activities of the business agents and initially consider appeals from members. The teamsters’ union requires that locals in a particular area join the district council, but the power these bodies exercise is not uniform throughout the union. In some areas, the district councilis powerful in collective bargaining and in its influence upon the affiliatedlocals; in others, its influence is only nominal. A district council maycover a particular community or stretch over a large area, as do the district councils of some of the building trades. These councils maynegotiate and administer contracts with employers.
The need for rapid and decentralized decisions has compellednational unions to develop regional forms of organization. Regional headsmay be ap-pointed, or they may be vice-presidents or members of the executiveboard elected on a regional basis. In some unions the officer is chosen by the dele-gates or members from the district he represents, and in others he is elected by the entire delegate body at a convention or by the general member-ship in a referendum.
The United Mine Workers of America was organized on a district basisfrom its beginning. Some districts were established prior to the nationalunion. District organizations became small-scale replicas of the national union, with a president, vice-president, and executive boardchosen by the district membership. At one time, the districts in the UnitedMine Workers were much more powerful administratively and politically than they are now. Most national leaders served apprenticeships in district offices, although John Mitchell and John L. Lewis, the greatest and most creative leaders, did not occupysignificant offices on that level. The districts exercised sufficient autonomy so that they could refuse to accept the interstate agreementswith the operators if they believed their interests were not wellserved. During the 1920s the United Mine Workers of America faced a seriousinternal crisis that was generated in part by the decline in economic importance of the older, northern districts. The districts’ loss ofmembership and power led in a number of instances to their being placed inprovisional status, with their officers appointed by the international president. West Virginia and Kentucky, which were newly organized in theearly 1930s, were also placed under provisional rule, so that the districtsbecame mere administrative units under the control of the international office. This development was influenced not only by the shift of the locus of power within the union from the older organized districts ofIndiana, Illinois, Ohio, and Pennsylvania to the newer, southern regionsbut also by economic shifts which required a single national policy andsubordination of regional leaders to the needs of the nationalorganization.
Even when the regional organization is not formally established in the same way as in the mine workers’ union, it may perform important organizational and administrative functions, especially when the regional office is headed by an international officer. Usually theregional director or vice-president will direct a number of organizers, who may also handle issues that cannot be settled through directnegotiation at the plant level. The regional head is ultimately responsiblefor developments within his region, and in some unions he rules initially on appeals for or against members involving internal union affairs orquestions of collective bargaining affecting such issues as seniority and grievance processing. The regional head may be responsible fororganizing work, act as a mediator between rival groups in locals, andparticipate in the more difficult or more important collective bargaining negotiations within the region. Regional directors often acquire anindependent position within the union. They may have a considerablefollowing among the locals, and they are among the chief molders of the organization’s policy.
Unions on the railroads have established lodges at various terminal points, and the lodges on a particular carrier are members of a systemfederation. Local lodge chairmen perform the same functions as businessagents or staff members in other unions, and general chairmen might beregarded as regional representatives.
A number of unions have established state conferences of local unions. These may be formally organized as in the bricklayers’ union, or theymay be informal meetings of local representatives to discuss commonproblems. A variety of structural forms have developed as a result of thebreakdown of the old American Federation of Labor principle of exclusive jurisdiction. Under this rule, a union claimed the right to organize all workers within its jurisdiction, and other affiliates were obliged to avoid infringement upon its jurisdiction by refraining from organizing such workers. This principle could not be effectively applied in manufacturing industries, especially in those that used mass-production methods. Moreover, when the AFL ceased to be the only federation, the principle could not be enforced. As a result, unions have been organizing a greater variety of workers than previously, and new types of units have been devised to allow special groups to work out their problems.
Unions may set up conferences or divisions for particular industries or occupations within their jurisdiction. The Textile Workers’ Unionof America has a dyers’ division because the problems and pay scales of dyers are different from those of other textile workers. Longshoremen may allow warehousemen who belong to the same union to have a separate unit.The automobile workers’ union has established a skilled trades department to handle problems of training, bargaining, apprenticeship, and other matters affecting skilled workers. The union’s primary activities are in an industry made up of firms with many establishments. A variety of departments—General Motors, Ford, Chrysler, aircraft, foundry, die casting, and many more—have been set up to coordinate activities. The heads of departments are appointed by the president of the union with the consent of the executive board.
The teamsters’ union has developed two types of intermediate units. The geographical conference, the Western Conference of Teamsters, was initiated by David Beck. This was, at first, an informal grouping of teamster locals from 11 western states and British Columbia. The principal reason given for the formation of this body was the need to coordinate bargaining with employers who operated over a wide geographical area. Geographical conferences have now been established over the entire United States and Canada. In addition, conferences of localsin particular industries have been set up on a regional and national level. Meetings are held periodically at which bargaining and organizing methodsare discussed.
The Associated Actors and Artists of America is a federation of eight unions in the entertainment industry, although the component organizations appear as divisions of the federation. The Actors’Equity Association, American Federation of Television and RadioArtists, and the Screen Actors Guild contain about 70 per cent of the total membership. The Seafarers International Union of North America is an all-inclusive organization of unlicensed personnel on the East and Gulf coasts, but it has three nominal departments on the Pacific coast antedating the formation of the SIU—the Sailors’ Union of the Pacific, the Marine Cooks and Stewards Union, and the Pacific Coast Marine Firemen, Oilers and Water tenders Union.
District 50 of the United Mine Workers of America presents another exception. Originally established by the parent organization to recruit workers in nonmining occupations, it functioned for a long time as a districtof the miners’ union. It has, however, organized several hundredthousand workers in utilities, manufacturing, and construction and now functions as a sort of English-type general union.
National and international unions
For many purposes the international or national union is the most important unit of structure and government. It is autonomous, and it establishes its own rules for managing its own affairs and for collective bargaining. Normally, international unions claim that their geographical jurisdiction extends over the United States and Canada. A number of the unions formed prior to World War I began as national unions and changed their title when they expanded intoCanada. Not all international unions function in every state, province, and region of the two countries, nor are the unions spread uniformly over them. The spatial dispersion of membership is determined by the location of the employing units whose workers the union recruits, by the responses of workers it seeks to organize, and by the resistance it encounters from employers.
Officers of the international unions are elected by the delegates to the convention or by the membership in a referendum. It has been noted that the tenure of international officers is long, regardless of the structure of the union, the kinds of workers recruited, and even the political coloration of the union. Contests for higher union office are unusual. Heads of unions, like their counterparts in other kinds of organizations, become well known. Unless there is serious division within the official hierarchy or widespread dissatisfaction with performance or policy, incumbents of national office tend to be re-elected repeatedly. In contests for office, incumbents have several inherent advantages. They perform tasks that bring them into direct contact with the membership, their reports appear in the official press, and they may handle collective bargaining and grievance negotiations for large groups within the union. Because the functions of union officers are likely to be more administrative than political, an argument can be made for long tenure. Contests for office at this level have recently increased, and it appears that this change reflects the influence of the Landrum–GriffinAct.
Local unions and district councils hold their charters from the international unions, and the jurisdiction, rights, and duties of the locals are defined in the international union’s constitution. The international union has the right under defined conditions to investigate the policies and conduct of its subordinate units, and if evidence of malfeasance, incompetence, or dishonesty is found, it may impose a receivership or trusteeship upon the local.
Trusteeships Trusteeships are regulated by the Landrum–GriffinAct. They must be reported to the United States Department of Labor, and the reasons for their imposition must be given. They can continue for only 18 months and can be renewed only upon a show of necessity and with the permission of the secretary of labor. The need for occasional intervention in local affairs is generally recognized, but the possibility of abuse of this power is obvious. Evidence of dishonesty,failure to carry out organizational tasks or conform to union rules, and factional rivalries that make it difficult for the union to function or administer its contracts fairly and effectively may warrant the intervention of the international union. The rigor with which trusteeships are applied is not uniform. In some instances a trusteeship may simply meanthe appointment of a representative of the international to supervise or to oversee the local’s activities. In other cases it can mean the displacement of some or all of the local officers and thetransfer of their functions to an appointed representative.
Appellate tribunals. International unions also serve as appellate tribunals to which members can appeal penalties imposed bysubordinate units or make complaints against local officers or policies. Unions as a rule permit appeal to more than one tribunal; in some a member dissatisfied with the verdict can appeal successively to the district council, regional vice-president, the international president, andfinally the convention. Some unions limit the number of appeals on certain questions and allow appeals on other issues to go to the convention at the request of the appellant. As of 1965, the requirement of charges being presented in writing and a trial being allowed was written into law. However, in the past, unions generally required charges to be made in writing and laid down trial procedures and methods of appeal. In many unions the membership of the local ultimately determines, through a vote, the guilt or innocence of a defendant and assesses the penalty. Some writers have argued that this is unfair because members are interested parties. It might be noted that this method does not always work to the disadvantage of the defendant. There have been more than a few cases in which members rejected overwhelming evidence against a popular defendant and refused to impose a penalty. Moreover, many penalties arereduced or set aside on appeal. Most of the charges made against members are for violation of union rules that govern conduct on the job.Charges do not customarily involve internal political questions. However, the latter are likely to result in charges of denial of rights.
Other international functions. The most important direct activity of the international union is the organization of the workers in its jurisdiction. The size of the organizing staff, the amounts expendedfor such purposes, and the significance of such activities differ among organizations. Several of the international industrial unions maintain large numbers of organizers and other staff people who perform functions similar to those of local business agents in other unions. In these cases, the central organization manages more of the local’s business than, for example, in the printing and building trades unions, where organizing activity by the international is likely to be largely in the form of monetary or staff aid to locals.
The international union sometimes administers benefit programs. The older unions, largely although not exclusively in the skilled trades, were organized before governmental systems of social security were in operation. These unions were formed to carry on both protective and beneficial activities. The latter include mortuary, sickness, pension, and unemployment benefits, which may be paid under specified conditions.
In the seagoing unions the international union is the only significant unit. Locals are mainly administrative centers to collect dues and to handle grievances and other problems. The shore delegates forthe various ports are elected by a general membership referendum or by conventions.
Finances
The income of the locals is derived mainly from dues, initiation and readmission fees, and assessments, although the last are not a significant source in most unions. Dues tend to be higher in the unions of skilled workers who earn high wages. It is those organizations which provide the more extensive welfare benefits, and dues must be high enough to finance them. As a rule, international unions tend to discourage the levying of assessments by their locals. Because these contributions are imposed irregularly they are likely to arouse dissatisfaction and opposition amongmembers. Consequently, in many organizations the local must demonstrate serious need before such a levy will be approved by the national officers. The Landrum-Griffin Act provides that increases of local dues or levying of assessments can be made only by a majority vote through a secret ballot of members in good standing at a general or special meeting.
Like dues, initiation and readmission fees tend to be higher among the unions of skilled than of semiskilled and unskilled workers. Some unions base their dues and initiation fees upon skill and upon the benefit rights particular members enjoy. Dues are usually a specified amount per month or quarter, but a few organizations charge a percentage of income earned at the trade or calling. International unions may require that all locals levy uniform dues and initiation fees, others allow a range of charges, a third group places a floor under these payments, and a fourth places a ceiling upon them.
International unions are mainly financed by a tax of a given amount per member per month. The size of the per capita tax also depends upon the kind of workers the union recruits, their skills, and the benefits the union provides. Per capita taxes tend to be higher in the skilled trades unions in building, printing, and entertainment. The tax may be divided into separate portions for carrying on various activities or placed in a general fund. For example, many unions allocate a set amount to a journal or publication fund. Another part may be placed in the mortuary, pension, or health and welfare fund, and a third into a general fund for administrative expenditures. Organizing costs may be drawn from a general or a special fund. Many unions maintain separate funds for the payment of strike benefits and allocate some part of the per capita payment for this purpose. Assessments are another source of income, although there is a tendency to avoid such levies. The conditions under which international assessments can be imposed are frequently defined in union constitutions; in a number of organizations they require the approval of the membership. Some unions also require locals to remit part of their initiation and readmission fees. Interest from securities and rents from property are also sources of income for international unions.
Officers and salaries
International officers are chosen either by the delegates at the conventions or by the membership in a referendum. The head of the union is usually the president, who as the chief executive officer is in charge of organizing and acts in general as a supervisor of the affairs of the locals under his jurisdiction. He ordinarily has authority to appoint organizers and much of the staff at headquarters. He is likely to be the most powerful person in the union, but in some unions the executive board may have more influence on final decisions. Unions also elect a secretary and a treasurer or combine the two offices into a secretary-treasurer.
Below the president, secretary, and treasurer in rank are the vice-presidents and members of the executive board; in some unions the vice-presidents constitute the executive boards. These bodies hold considerable power in many unions. They can review most decisions made by the officers stationed at headquarters, and they must approve a wide variety of decisions. To some extent their power depends upon the caliber of the persons serving. Vice-presidents and executive board members are selected by the entire union or by districts and serve for specific terms.
Salaries of union officers can be divided into several levels. The highest are the salaries of the presidents, secretaries, and treasurers. Next in line are vice-presidents and executive board members; staff men may be included in the latter group or occupy a position of their own. The lowest group are the local officers, although in some unions the heads of large locals may be paid salaries equal to those of organizers or even international vice-presidents. In a few organizations the leaders of large locals also serve as executive board members or vice-presidents of the international union and may receive extra compensation for these services. As a rule the business agents and other full-time officers in local unions of skilled trades will receive salariessome what in excess of the wages received by the members. In addition to salaries, union officers receive expenses. Organizers and other international officers are allowed either a specified amount for daily expenses and travel expenditures or are paid for expenses incurred.
The AFI-CIO
The American Federation of Labor-Congress of Industrial Organizations is made up (as of 1965) of 137 national and international unions. About another fifty international unions remain unaffiliated. There are, in addition, several small federations of unaffiliated unions and a large number of independent single-plant unions. Some of the national unions outside of federation have withdrawn because of policy differences, as did the United Mine Workers of America; some were expelled by the AFL on charges of corruption, as were theteamsters’ and bakery and confectionsery workers’ international unions; some were ousted by the CIO on charges ofcommunist domination, as were the United Electrical, Radio and Machine Workers of America and a number of others.
The AFL-CIO is engaged in organizing, or in assisting its affiliates to organize, and carries on educational campaigns on behalf of the labor movement. It seeks to settle jurisdictional differences among its affiliates, and it aids those occupying the same or contiguous jurisdictions to merge, an activity in which it has had only modestsuccess. It is engaged in two types of political activity.First, it supports before Congress legislation it regards as benefiting the population, irrespective of whether it affects the organized or the unorganized. Its legislative committee examines every bill introduced in the Congress to determine its effect upon labor. This committee carries out the instructions of conventions or the directives ofthe executive council. Second, through the Committee on Political Education, the AFL-CIO makes known the voting records of congressmen, aids in registering voters, and endorses and aids those candidates it holds worthy of support. It publishes bulletins and leaflets and seeks to rally union members and friends in support of its endorsements.
The supreme governing body of the AFL-CIO is the biennial or special convention. The officers are a president and a secretary-treasurer, who are the executive officers, and 27 vice-presidents, each of whom must be a member of an affiliated organization and be chosen by the convention. The president, as the chief executive officer of the organization, supervises its affairs and presides at all conventions and at meetings of the executive council and the general board. He has the power to interpret the constitution, but he can be overruled by the council. The secretary-treasurer is in charge of monies and property. The executive council is the governing body between conventions, and it carries out the directives and decisions of conventions. In addition, an executive committee of the president, secretary-treasurer, and six vice-presidents consults with and advises the president on policy matters. A general board is made up of members of the executive council and a principal officer of each of the affiliated national or international unions and of each trade and industrial department. It meets at least once ayear and decides questions submitted to it by the executivecouncil.
A department of organization is prescribed by the constitution. In addition, the constitution calls for the appointment by the president of 14 standing committees to carry out the tasks of the federation and the directives of the convention. These committees deal with legislation, civil rights, political education, ethical practices, international affairs, education, social security, community services, economic policy, housing, research, public relations, safety and occupational health, andveterans’ affairs.
A large staff carries out the work of the federation. The organizing department maintains 23 regional offices, each headed by a regional director. The organizing staff of the federation can be of considerable importance in a given area or situation. Generally, however, it serves as an aid or auxiliary to the activities of affiliated unions. The social security division assembles information on various aspects of welfare legislation and prepares memoranda for the officers and releases for the press. The research department analyzes economic data, prepares briefs and testimony, and issues a monthly economic bulletin. The publications division issues a weekly newspaper, a monthly magazine, and other printed materials.
Central bodies. Central bodies are directly affiliated with the AFL-CIO on a city, state, or regional basis. These are composed exclusively of locals of national and international unions and organizing committees affiliated with the federation. In addition, a number of local unions and organizing committees are directly chartered by the federation. Whenever a sufficiently large group of locals in a particular trade or industry is affiliated with the federation, it may at the decision of the executive council be formed into a national council, but it remains directly affiliated with the federation. Finally, groups of local unions may be authorized to combine into autonomous national and international unions. Groups of directly affiliated local unions may request the executive council to authorize such combinations. The federation performs all the services and exercises the same authority toward directly affiliated locals as national and international unions do in the case of their locals.
The AFL-CIO is supported by a per capita tax levied upon all affiliates and upon directly chartered local unions. Revenue may also be derived from assessments authorized by a convention, and the executive council can call for the payment of limited assessments.
Trade and industrial departments. The need to devise a method of cooperation among unions operating in contiguous trades was early recognized. Local trades councils in the building and metal trades were formed in the 1880s. These organizations sought to develop cooperation among the separate trades employed in the same industry. The AFL recognized the building trades councils and allowed these bodies to affiliate with it. The AFL, however, feared that formal cooperation by a group of international unions might lead to the emergence of a competing power in the trade union world. After several unofficial efforts, the AFL recognized the need for formal cooperation and established departments in a number of industries in 1908. The building and construction trades department charters trades councils and seeks to devise general policies that will promote the unions and the construction industry. It has developed, after many trials, a fairly successful method of handling jurisdictional disputes among its members, who have been drawn closer together by the desire to present a common front against the industrial unions in the work assignment disputes. The functions of the metal trades department are similar to those of other trades departments, except that it has not faced many internal disputes. It charters trades councils in the metal-working industries that sometimes negotiate common contracts with employers. The union label and service trades department conducts campaigns for the use of union-made products and the patronage of service establishments which display the union insignia.
The railway employees department is made up of six international unions whose members maintain and service railroad equipment. These unions bargain jointly with the carriers. The maritime trades department seeks to enroll unions whose members are employed in seagoing services, or in serving vessels, or in handling cargo. Similarly, the food trades department is made up of unions in the food processing and distributing industries. The industrial union department is made up of a wide variety of unions, and it functions more like a general federation than a department. It has sought to develop common organizing drives and unifiednegotiations with employers—coalition bargaining —in firms where a number of unions hold bargaining rights. It also carries one ducational, research, and lobbying activities.
The Government Employees’ Council was formed in 1945 so that unions of government employees would be able to prepare common programs for legislative and administrative action. The Railway Labor Executives’ Association is made up of the chief executive officers of the railway unions, all but two of which are affiliated with the AFL-CIO. Most of the railway unions have virtually all of their members in the railroad industry, but eight have members outside of that industryand several have the majority of their members outside. The association, formed in 1926, functions as a policymaking body on legislation and other issues of interest to railroad workers. It is not a federation ofunions.
Philip Taft
BIBLIOGRAPHY
Barbash, Jack 1961 Labor’s Grass Roots. New York: Harper.
Chamberlain, Neil W. 1965 The Labor Sector: An Introduction to Labor in the American Economy. New York:McGraw-Hill.
Commons, John Rogers et al. 1918–1935 History of Labour in the United States. 4 vols. New York: Mac-mill an.
Galenson, Walter 1960 The CIO Challenge to the AFL: A History of the American Labor Movement, 1935–1941.Cambridge, Mass.: Harvard Univ. Press.
Gitlow, Abraham L. (1957)1963 Labor and Industrial Society. Rev. ed. Homewood, 111.: Irwin. ? First published asLabor Economics and Industrial Relations.
Hoxie, Robert (1917) 1936 Trade Unionism in the U.S. 2d ed. New York: Appleton.
Leiserson, William M. 1959 American Trade Union Democracy. New York: Columbia Univ. Press.
Miernyk, William H. 1965 The Economics of Labor and Collective Bargaining. Boston: Heath.
National Industrial Conference Board 1956 Sourcebook of Union Government, Structure and Procedures. New York: The Board.
Saposs, David J. 1926 Left Wing Unionism. New York: International Publishers.
Slighter, Sumner H.; Healy, James J.; and Livernash, E. ROBERT 1960 The Impact of Collective Bargaining on Management. Washington: Brookings Institution.
Taft, Philip 1954 The Structure and Government of Labor Unions. Cambridge, Mass.: Harvard Univ. Press.
Taft, Philip 1964 Organized Labor in American History. NewYork: Harper.
Ulman, Lloyd 1955 The Rise of the National Union: The Development and Significance of Its Structure, Governing Institutions, and Economic Policies. Cambridge, Mass.: Harvard Univ.Press.
IV. INFLUENCE ON WAGES
The main questions to which this article is addressed are empirical ones: (1) To what extent have labor unions affected the relative wages of different groups of labor in the United States?(2) What factors account for the observed differences in effects ofunions?(3) How variable were the relative wage effects from one date toanother during the period 1929–1958? (4)What regularities appear in the variations of theeffects over time?
Absolute and relative wages
Throughout this article, thetermwages covers the earnings of selfemployed persons as well as the compensation of wage and salary workers. The latter includes all remuneration for time worked at either straight-time or premium rates and for time not worked (vacation, holiday, sick leave, call-in pay, and other paid leave) and employer contributions to public and private unemployment, pension, health, and other employee welfare funds. A worker’s absolute moneywage per hour—for short, hismoney wage—is the sum in dollars of all of the pecuniary rewards to him per hour of his labor services. His real wage is the same sum expressed in dollars of fixedpurchasing power.
Thegeneral wage level, which may be expressed in either money or real units, is the average (mean) wage per hour of all workers in the U.S. labor force.
Therelative wage of a group of workers is the ratio of the average (mean) wage of all workers in the group to the average wage of all workers in the labor force, i.e., to the general wage level. Relative wages are indicators of the percentage differences in wages among groups of workers. For example, if the relative wage of one group is 1.5 and that of a second group is 0.75, the average wage of the first group is 50 per cent greater than the average wage of all workers and 100 per cent greater than the average wage of the second group, and the average wage of the second group is 25 per cent below the average wage of all workers.
It is impossible for the wage of every worker to exceed (or, alternatively, to be below) the average wage of all workers. Indeed, the average relative wage of all workers is always exactly equal to unity. Therefore, if the relative wages of some workers exceed unity, the relative wages of some other workers must be less than unity.
This article deals only with the effects of labor unions on relative wages. Unionism affects the relative wages of different groups only if it causes the absolute wages of the groups to change by percentage amounts that differ among the groups. Furthermore, if unionism raises the relative wages of some groups, it thereby lowers the relative wages of some other groups. For example, if nonunion workers outnumber union workers in the labor force in a three to one ratio, a 9 per cent increase in the average relative wage of union workers implies a 3 per cent decline in the average relative wage of nonunion workers.
On the other hand, that unionism has increased the average wage of union workers by, say, 12 per cent relative to the average wage of nonunion workers does not imply that unionism has raised theaverageabsolute money or real wage of union workers by 12 per cent rather than some different percentages.The effects of unionism on the general level of either money or real wages cannot be deduced from knowledge only of the relative wage impact of unionism.
Methods of estimation
The average relative wage of a group of labor in the presence of the unionism existing at a particular date can be observed directly. What the corresponding average would have been in the absence of unionism, however, can be estimated onlyuncertainly from observable data. Two different types of approaches have been used to estimate the relative wage effects of unionism:
Theindirect approach estimates the effect of unionism on the relative wage of a group of labor from estimates of the effects of unionism on other economic magnitudes pertaining to the group, such as its relative employment, its relative quit rate, and the excess supply (queuing) of labor to the group. Evidence that such nonwage effects are not zero often can be taken as evidence that unionism has affected the relative wage of the group. It is usually difficult, however, to estimate the size of the implied relative wage effect from evidence on the nonwage effects.
In thedirect, orwage comparison, approach the relative wage effects are estimated from comparisons of the wages or changes in wages of two or more groups of labor differing in their degree of unionization. The relative wage effect estimates presented later in this article are based almost entirely on studies that have used thisapproach.
Four types of wage comparisons may be distinguished in these studies:
Variant 1 compares the average wage of a given, usually more or less highly unionized, group at a particular date with the average wage at the same date of a bench mark group. The latter group commonly is a nonunion group considered highly comparable to the given group. In a few studies,however, the bench mark group approximates all labor in the economy. The difference in average wage between the two groups is adjusted to eliminate the effects of factors other than unionism, and the adjusted difference is attributed to unionism. This procedure yields an estimate of the effect of unionism on the average wage of the given group relative to the average wage of the bench mark group at the particular date.
Variant 2 treats simultaneously three or more groups differing in degree of unionization. This variant adjusts the average wages of the covered groups for the effects of factors other than unionism andcorrelates the adjusted average wages with the degree of unionization of the groups. Commonly the adjustment and the correlation take place simultaneously in a multivariate analysis of wages, degree of unionization, and other variables. Variant 2 estimates the effect of unionism on the average wage of union workers in the covered groups, relative to the average wage of the covered nonunion workers. The resulting estimate may be biased, however, even when the adjustment of the wage data for the effects of factors other than unionism is perfect. In general, the algebraic size of the bias can be calculated onlyroughly.
Variant 3 compares the change from a base date to a given date in the average wage of a given group with the corresponding change for a bench mark group and adjusts the difference between the two wage changes for effects of factors other than unionism. The effect estimated by this procedure is the change from the base date to the given date in the effect of unionism on the average wage of the given group relative to the average wage of the bench mark group.
Variant 4 correlates the adjusted changes from a base date to a given date in the average wages of several groups of labor with the degree ofunionization of the groups at the given date. This variant estimates the change between the two dates in the effect of unionism on the average wage of covered union labor relative to the average wage of covered nonunion labor. This estimate, like that from Variant 2, may be biased.
The central problem in all of the wage comparison procedures is adjusting the wage data for the effects of variables, other than unionism, that may be correlated with the effects of unionism. The data and procedures that have been used to make such adjustments are too detailed andtoo specialized to the individual studies to be summarized briefly. It should be said, however, that in general the adjustments have not been so refined as to leave little margin for error in the estimates of the relative wage effects of unions. For this reason the estimates given below should be regarded as quite tentative.
Recent research in the United States
Although unionism in the United States is well over a century old, it was not until the end of World War ii that union members comprised as much as 25 per cent of the U.S. labor force.A decade earlier union membership was about 8 per cent of the labor force, which was only slightly higher than the corresponding percentage in 1905. Since 1945 unionization in the U.S. labor force has remained around 25 per cent, rising by a small amount from 1945 to 1953 and then declining to a level in 1960 almost the same as that in 1945.
It is not surprising, therefore, that almost all of the empirical research on the relative wage impact of unionism in the United States was done after 1945. Two other factors help to explain the concentration of the research in the recent period. First, the supply of wage and other data required for research in this area also expanded greatly during the 1930s and 1940s. Second, command of the techniques of statistical economics was uncommon among economists until after 1945.
The empirical findings summarized below are based on evidence presented in 21 of the recent studies of unionism and wage differentials in the United States (see Lewis 1963, chapters 3–4, for reviews of thesestudies ). Since a considerable part of the recent research has been reported only in unpublished papers and doctoral dissertations, the summary undoubtedly is not exhaustive.
Nine of the studies cover large segments of the U.S. labor force and use the Variant 2 or Variant 4 wage comparison procedures to estimate the relative wage effect of unionism. Thus the summary numbers that emerge from these global studies all tend to be estimates either of the effect of unionism on the average wage of all union workers, relative to the average wage of all nonunion workers at a particular date—in short, the average union/nonunion relative wage effect—or of the change in this effect between two dates. These studies provide little or no information on the dispersion of the relative wage effects among different groups of union workers.
The other 12 studies deal with relatively small parts of the U.S. labor force as follows: production workers in basic steel manufacturing (Rees 1951); hotel employees in large cities (Scherer 1956); production workers in rubber tire manufacturing (Sobel 1954); skilled craftsmen and, separately, common labor in building construction in large cities (Sobotka 1953); production workers in bituminous coal mining (Greenslade 1952); employees in selected establishments, localities, and occupations, separately, by industry, in the manufacturing of paints andvarnishes, wooden furniture, footwear, cotton textiles, hosiery, automotive parts, and women’s dresses (Maher 1956); barbers in large cities (reported in Lewis 1963, chapter 3); production workers in the manufacturing of men’s and boys’ suits and coats (Rayack 1958);commercial airline pilots (Sobotka et al. 1958); street railway and bus motormen (Lurie 1961); seamen in east coast ocean transportation (Rapping 1961); and physicians (Friedman & Kuznets 1945; Lewis 1963, chapter 3).
These detailed studies provide a check on the estimates of the average union/nonunion relative wage effect drawn from the global studies. More importantly, it is only from such studies that the dispersion in the relative wage effects of unionism among detailed industries and occupations can be gauged.
Variation of influence over time
The clearest finding emerging from the 21 studies is that the size of the effect of unionism on the average union/nonunion relative wage in the United States has varied greatly and in a systematic fashion over time. The largest effect apparently occurred about 1932, during the rapid downturn of the depression. In 1932 the average union/nonunion relative wage effect may have been 25 per cent or even larger. In the recovery period from 1933 to the outbreak of World War II the effect dropped sharply to a level that was probably between 10 and 20 per cent in 1939.
The decline in the effect continued through World War II and into theperiod of rapid inflation that followed the war. At the trough, which occurred in the period 1945–1948, the effect of unionism on the average union/nonunion relative wage was approximately zero. (There is some evidence that the union/nonunion relative wage effect also was near zero at the peak of the inflation following World War I.)
The very low relative wage effect observed during 1945–1948 did not persist, however, in the following ten years. By 1957–1958 the effect of unionism on the average wage of union labor relative to the average wage of nonunion labor had risen to about 10 to 15 per cent. It should be noted in this connection that the rate of inflation of the U.S. general price level during 1949–1958 was only about a third as large as that during1939–1948.
In the late 1950s the extent of unionization of the U.S. labor force was approximately 25 per cent. Therefore, about one-fourth—or threeto four percentage points of the 10 to 15 per cent effect of unionism on the average union/nonunion relative wage estimated for 1957–1958—consists of a decline in the average wage of nonunion workers relative to the average wage of all workers. The remainder, 7to 11 per cent, is the estimated effect of unionism on the average wage of union workers relative to the average wage of all workers.
Role of money wage rigidity. That the effects of unionism on the relative wages of union workers tended to be greatest during periods of rapid deflation of the general price level and lowest during times of unexpectedly rapid inflation was observed and commented upon by the authors of several of the unionism wage studies. Albert Rees, Milton Friedman,Walter A. Morton, and numerous other economists have attributed the apparent negative correlation between the rate of inflation and theunion/nonunion relative wage effect of unionism to wage rigiditiesintroduced by collective bargaining: the union contract running for aperiod of a year or more, the reluctance of employers of union labor toagree to unusually large money wage in-creases during periods of unexpectedly great inflation lest the inflation not continue, and the similar opposition of unions to wage decreases during periods of declining prices.
One of the studies (Lewis 1963, chapter 6) has tried to measure the sensitivity of the average union/nonunion relative wage effect of unionism to variations in the rate of inflation and the rate of unemployment in the U.S. labor force during the period 1919–1958. The results of the study tend to confirm the large variability over timeof the relative wage effect observed in other studies and suggest thatthe variability is more largely ac-counted for by negative correlation with the rate of inflation than by positive correlation with the rate of unemployment.
Effects on the interindustrial relative wage structure
The effect of unionism on the average relative wage of the workers employed in a particular industry is a weighted average of corresponding relativewage effects for (a) the union labor and (fc) trie nonunion labor in theindustry, where the relative weight for the union labor is the extent ofunionization of the workers in the industry. For example, if the extentof unionization were 50 per cent and if unionism had raised the relativewage of the union workers in the industry by 20 per cent and lowered therelative wage of nonunion workers by 2 per cent, the effect of unionism on the average relative wage of the industry would be a plus 9 per cent. Therefore, the greater the dispersion among industries in the extent oftheir unionization and in the effects of unions on the relativewages of union workers and of nonunion workers, the greater will be thedispersion among industries in the effects of unionism on their averagerelative wages.
Since 1944 roughly three-fourths of the persons engaged intransportation and about half of those in mining, construction,manufacturing, communications, and electric and gas utilities have beencovered by collective bargaining arrangements. However, within each of these industry divisions there is considerable dispersion in extent ofunionization among detailed industries.
In contrast, the extent of unionization has been close to zero in agriculture, trade, finance and insurance, and government (excludinggovernment enterprises). The same is true in general of the serviceindustries except for hotels and eating and drinking places in large cities, the entertainment industries, and some of the personal service indus-tries. About half of the labor force is employed in the agriculture, trade, finance and insurance, service, and government industry divisions.
Estimates of extent of unionization by industry, with considerableindustry detail, are available for all of the major industry divisions in the United States. On the other hand, the 12 detailed”industry” studies of unionism and wages provide estimates of the relative wage effects of unionism for only a short list of industries. Nevertheless, the general character of the distribution of the relative wage effects among all U.S. industries can be gauged by combining the results of the 12 studies with the data on extent of unionization by industry. In the late 1950s the majority of workers probably were employed in industries whose average relative wages were raised orlowered by unionism by no more than 4 per cent. The data also indicate thatthe distribution was positively skewed with relative wage effects in someindustries equal to or exceeding 20 per cent. It is unlikely, however, that the payroll of these industries was more than a very small fraction of the aggregate compensation of the whole labor force.
The industries in which the relative wage effects of unionism were the largest algebraically tended to be those with above average relative wages. Therefore, unionism has been a factor making the relative inequalityof average wages among indus-tries larger than it otherwise would be. In 1958the coefficient of variation of average wages among industries was about 6 to 10 per cent larger than it would have been in the absence of any effects of unions on relative wages.
Effects on relative wages by occupation
Fragmentary dataindicate that the dispersion of extent of unionization among occupations isprobably at least as large as that among industries. Within industries characterized by industrial unionism— unionism of wage earnersregardless of occupation —there is, of course, relatively littledispersion in extent of unionization among the wage earner occupations. Onthe other hand, in both these indus-tries and others, white-collar(supervisory and other nonproduction) employees typically have not beenorganized by unions. In addition, in industries in which unionism followscraft (occupation) lines, such as printing and building construction,there are some substantial differences in extent of unionization among thewage earner occupations.
These differences in extent of unionization by occupation, together with the estimates of relative wage effects of unionism drawn from the studies that have dealt with occupations, suggest that the dispersion in relative wage effects of unionism among all occupations in the United States may be as large or larger than that among all industries.Furthermore, the interoccupational distribution of relative wage effects ofunionism, like that for industries, probably is positively skewed, with asmall fraction of the labor force employed in occupations in which therelative wage effects were larger than 20 per cent
Differing effects among union workers
The 12 studies thatdealt with fairly small groups of union workers indicate that thedispersion across union jurisdictions of the effects of unions on therelative wages of union workers probably was substantial except in theperiod near the end of World War II. For the period 1954–1957, for example, the esti-mates of these effects, drawn from the six studies thatprovided figures for this period, ranged from zero to about 50 percent.
Unfortunately, although the 12 “industry” studies covered 18 different groups of workers, there is no single date at which numerical estimates are available for each of these groups. For example, for the period 1936–1939 there are estimates for seven groups:(1) bituminous coal miners (30%); (2) skilled building craftsmen (25% ); (3) wage earners in men’s clothing manufacturing (16%);
(4) wage earners in rubber tire manufacturing (14%); (5) motormen in local transit (8%); (6) unskilled building labor (5% ); (7) hotelemployees (0%) . For the later period, 1954–1957, there are estimates for six groups, only three of which are on the preceding list: (1) bituminous coal miners (50% ); (2) commercial airline pilots (27%); (3) east coast seamen (20%); (4) barbers (19% ); (5) motormen in localtransit (12% ); (6 ) wage earners in men’s clothing manufacturing (0% ).Furthermore, some of the differences among the studies in the estimates atthese and other dates consist more largely, no doubt, of errors of estimationthan of real differences in effects, and in some other instances estimationerrors may have hidden real differences. For these reasons any ranking ofthe 18 groups according to the size of the relative wage effect at aparticular date is subject to considerable uncertainty.
Nevertheless, it can hardly be doubted that there were some largedifferences in relative wage effects from one union jurisdiction to anotherin the latter part of the 1950s. What factors account for thisdispersion? In the literature on labor unions, there is much evidence that the relative wage effect for the union workers in an industry will tend to be greater the larger the ratio of the output of union firms in the industry to the corresponding output of nonunion firms. The studies of unionism and wages also support the view that the relative wage effect will be larger, other things the same, the greater the rate of increase in the demand for labor in the industry.
That these two factors may fail to explain a substantial fraction of the dispersion among unions in relative wage effects for union workers is indicated by the fact that, except for 1945–1948, the largest effects estimated in the 12 studies consistently were those for bituminous coal miners. These miners, to be sure, have been relatively highly unionized; but so were many of the other groups covered in the studies. Furthermore, since World War II the relative demand for labor inbituminous coal mining has been declining.
Some observers of unionism in the United States have attributed the large relative wage effects for unionized coal miners in part to an unusual lack of concern by the United Mine Workers, the union representing most of the unionized coal miners, for the reduced employment of union miners resulting from high relative wage effects. It may be true that differences among unions in the extent of their concern for the employment of their members are important in explaining the differences in their relative wage effects. Until measures of such “taste” or “choice” differences among unions are developed, however, the quantitative significance of these differences for relative wages cannot be gauged.
Numerous other factors have been suggested by economists to explain the dispersion among unions in their effects on the relative wages of union workers: the ease of substituting other productive services for the services of union labor in production; the elasticity of supply of these other productive services to employers of union labor; the ratio of the cost of union labor to the cost of the other productive services; the degree of concentration of output (“degree of monopoly”)among firms in unionized industries; the ease of entry of new firms into unionized industries; the extent to which union control of the supply of labor in its jurisdiction is facilitated by governmental licensing of occupations and trades; and other factors. For each of the enumerated factors it is relatively easy to find cases that apparently support its inclusion among the variables explaining interunion dispersion of relative wage effects.
However, the absence of data on several of these factors, the small number and likely selection biases in the sample of groups of union workers for which estimates of relative wage effects of unionism are available, the differences among the estimates in the dates to which they pertain, the errors in the estimates, and the need to control for the effects of other factors while examining those of each particular factor impede, at present, evaluation of the relative importance of these factors or even the estimation of thedirection of their effects. Such evaluation will require much additional research on the factors themselves and on the impact of unionism in industries and occupations for which relative wage effect estimates are unavailable for recent years.
H. GreggLewis
BIBLIOGRAPHY
Friedman, Milton; and Kuznets, Simon 1945 Income From Independent Professional Practice. National Bureau of Economic Research, General Series, No. 45. New York: The Bureau.
Garbarino, Joseph W. 1950 A Theory of Interindustry
Wage Structure Variation.Quarterly Journal of Economics 64:282–305.
Goldner, William 1958 Labor Market Factors and
Skill Differentials in Wage Rates. Pages 207–216 in Industrial Relations Research Association, Proceedings of the Tenth Annual Meeting. New York: The Association.
Greenslade, Rush V. 1952 The Economic Effects of
Collective Bargaining in Bituminous Coal Mining. Ph.D. dissertation, Univ. of Chicago.
Levinson, Harold M. 1951 Unionism, Wage Trends, and Income Distribution, 1914–1947.Michigan BusinessStudies 10, no.4.
Levinson, Harold M. 1960 Postwar Movement of Prices and Wages in Manufacturing Industries. U.S. Congress, Joint Economic Committee, Study Paper No. 21. Washington: Government Printing Office.
Lewis, H. Gregg 1963 Unionism and Relative Wages in the United States: An Empirical Inquiry. Univ. of Chicago Press.
Lurie, Melvin 1961 The Effects of Unionization on
Wages in the Transit Industry.Journal of Political Economy 69:558–572.
Maker, John E. 1956 Union, Nonunion Wage Differentials.American Economic Review 46, no. 3:336–352.
Princeton University, Industrial Relations Section1960 Wage Behavior in the Postwar Period: An Empirical Analysis. Research Report Series, No. 100. Princeton Univ., Industrial Relations Section.
Rapping, Leonard A. 1961 The Impact of Federal Subsidies and Maritime Unionism on the Relative Earnings of Seamen. Ph.D. dissertation, Univ. of Chicago.
Rayack, Elton 1958 The Impact of Unionism on Wages in the Men’s Clothing Industry: 1911–1956. Pages 674–688in Industrial Relations Research Association, Proceedings of the SpringMeeting. New York: The Association.
Rees, Albert 1951 Postwar Wage Determination in the Basic Steel Industry.American Economic Review 41: 389–404.
Rees, Albert 1962 The Economics of Trade Unions. Univ. of Chicago Press.
Ross, Arthur M. 1948 The Influence of Unionism Upon Earnings.Quarterly Journal of Economics 62:263–286.
Ross, Arthur M.; and Goldner, William 1950 Forces Affecting the Interindustry Wage Structure.Quarterly Journal of Economics 64:254–281.
Scherer, Joseph 1956 The Union Impact on Wages: The Case of the Year-round Hotel Industry.Industrial and Labor Relations Review 9:213–224.
Sobel, Irvin 1954 Collective Bargaining and Decentralization in the Rubber-tire Industry.Journal of Political Economy 62:12–25.
Sobotka, Stephen P. 1953 Union Influence on Wages: The Construction Industry.Journal of Political Economy 61:127–143.
Sobotka, Stephen P. et al. 1958 Analysis of Airline Pilot Earnings. Unpublished manuscript, Northwestern Univ., Transportation Center.