March 12, 1998
By Tim Costello & Jeremy Brecher
In the past quarter-century, capitalism has undergone a revolutionary restructuring in the United States and worldwide. Meanwhile, the American labor movement retains the basic structure it established more than half a century ago. Organized labor will have a future if and only if it can redesign itself to cope with the restructuring of capitalism. Neither the new leadership of the AFL-CIO nor its critics on the left and right have engaged in more than a desultory discussion of the changes organized labor must undergo to meet the restructuring of global capitalism. We don’t intend, here, to analyze the full range of changes needed in the labor movement, but to jump-start the discussion of union structure.
American capitalism in the industrial revolution era was dominated by individual entrepreneurs and partnerships producing primarily for a local market and employing at most a few dozen workers. By the beginning of the 20th century, this structure was replaced by one in which a handful of national corporations employing tens or hundreds of thousands of workers dominated each industry. Each corporation sought vertical integration, controlling every phase of production from raw materials to delivery of the finished product to the consumer. And each sought horizontal integration, providing every possible product and/or service within the boundaries of its industry. With some national variations, each major industrial nation possessed a roughly similar structure.
The past two decades have seen a radical restructuring of corporations worldwide. This restructuring has been referred to by many different terms, including downsizing, outsourcing, lean production, and the virtual corporation. Whatever the terms used to describe the change, corporations have moved away from horizontal and vertical integration within an industry and toward a new “core/ring” structure: corporations retain a small core of essential functions within the company, while contracting out the rest to suppliers around the world. Meanwhile, the corporation itself cuts across industrial and national boundaries, further blurring their own boundaries as they join in strategic alliances and multiple forms of partnerships, even with companies with which they are also competing. The result is what might be called the “dis-integrated” corporation. This is far from a return to a I9th century-style free market of small producers. As Bennett Harrison writes in Leon and Mean: The Changing Landscape of Corporate Power in the Age of Flexibility (1994), the “signal economic experience of our era” is not “an explosion of individual entrepreneurship,” but rather “the creation by managers of boundary-spanning networks of firms, linking together big and small companies operating in different industries, regions, and even countries.” Big firms “create all manner of networks, alliances, short- and long-term financial and technology deals with one another, with governments at all levels, and with legions of generally (although not invariably) smaller firms who act as their suppliers and subcontractors, “But the locus of ultimate power and control “remains concentrated within the largest institutions.” The result is what Harrison describes as an “emerging paradigm of networked production” based on concentration of control combined with decentralization of production. While corporations have pioneered, this structure, “privatization” of government functions represents a parallel core/ring” structure in the public sector. The staffing strategy corresponding to this new structure divides the workplace into a shrinking number of core workers and an expanding group of workers in contingent jobs – part-timers, temps, contract workers, day-laborers, and the like. Workers in the core group usually have a basic benefit package, work under standard whether the economic restructuring of the past couple of decades is really a question of “globalization” or rather a question of “lean production ‘ ” This is a false dichotomy; both phenomena are part of the same dynamic.
How Labor Met the Challenge of the Integrated Corporation
To gain some perspective on how organized labor might respond to this transformation, it may be useful to take a look at how it responded to the change from local craft production to the national industrial corporation. While workers need and have formed many kinds of organizations, labor unions have a common trait that distinguishes them from labor parties, social reform organizations, and other labor institutions: they organize workers who would otherwise be in competition with each other. Craft unions emerged in the early 19th century as an increasing number of craft workers were transformed from self-employed artisans to wage laborers. Tailors, cord-wainers, and tinsmiths saw employers cut wages and, if these workers refused to work for less, replace them with others who would. The result was what we today call a “race to the bottom,” in which wages and working conditions tended to fall toward those experienced by the poorest and most desperate. Craft unions linked those who were subject to this competition: workers in the same craft. They created standards for wages and working conditions that they insisted be applied to all workers in the craft. Initially the unions enforced these standards, not through contracts with the employer, but by a refusal to work under conditions that violated union “legislation.” The effect they aimed for was to eliminate labor costs as a factor in their employers’ competition. When production became centralized in large, vertically and horizontally integrated corporations, individual craft unions became virtually powerless.
The symbol of this powerlessness – a sort of 19th century equivalent to the defeat of the air traffic controllers in the 1981 PATCO strike- was the defeat of the steelworkers at Homestead by the Carnegie Steel Company in 1892. Thereafter craft unions, with relatively small and temporary exceptions, were excluded from most major American industrial corporations for nearly half a century. “Craft autonomy” was the basic principle on which Samuel Gompers had built the dominant American labor center, the American Federation of Labor. The radical Industrial Workers of the World – the “Wobblies” -scorned this structure as the “American Separation of Labor.” An article written by Wobbly activist William Trautmann around 191 1 called “Why Strikes Are Lost” described a pathetic series of strikes in which groups of workers went down to devastating defeats while unionized workers in the same company belonging to other craft unions continued to work. The Wobblies and many others over the next half century advocated one or another version of “industrial unionism,” in which all workers in a particular industry would be organized into one union. That was the only way workers could remove labor costs as a competitive factor among their employers. Only when the AFL’s Committee on Industrial Organization began to ignore craft distinctions and begin to organize the steel, auto, and other basic industries on an industrial basis, did organized labor reestablish itself in the core of American industry. The success of American workers in gaining a share of the economic growth that began with World War 11 was largely rooted in their ability to eliminate labor costs as a factor in competition within an entire national industry through industrywide and “pattern” bargaining. Of course, this “industrial” structure was not the only one in the labor movement. Some craft unionism remained, especially in construction. The public sector unions, which expanded rapidly in the 1960 and 1970s, developed organizational patterns largely shaped by the governments with which they bargained. Union structure reflected not only unions’ own policies, but also the bargaining units delineated by the National Labor Relations Board.
The New “Separation of Labor”
As corporations diversified, globalized, downsized and subcontracted, and as contingent labor became more common, the labor movement’s response was almost entirely ad hoc. Unions amalgamated shrinking locals and signed up any new members they could find. Diversified national unions merged, thereby becoming still more miscellaneous. The fundamental principle underlying union organization -to unite those who would otherwise be in competition -appeared almost forgotten. This response might be characterized as “general unionism” – the tendency of unions to incorporate workers even if the work they did was completely unrelated. (This should not be confused with the current union merger movement in Germany and Holland, also referred to “general unionism,” in which a country’s unions are consolidated into as few as three to six unions covering such broad sectors as manufacturing, public services, and regulated utilities.)
For example, the United Auto Workers (UAW) has incorporated freelance writers and university staffers about as far as imaginable from its previous core constituency of auto and other metalworkers. Equivalent examples could be found for most large unions. General unionism has become a widespread practice in American unions, even in those which espouse industrial unionism in theory. General unionism represents in part a response to the breakdown of traditional boundaries among industries in the era of “networked production.” But general unionism also represents an abandonment of the core union goal of uniting all those workers who are in competition with each other. While the UAW was organizing writers – itself something to be applauded – a rapidly growing proportion of auto and auto parts plants were becoming nonunion, with devastating results for auto workers and the UAW itself. General unionism does not solve the problem of linking workers who would otherwise be in competition. In the era of the “dis-integrated corporation,” we face a new version of the “Separation of Labor” on a colossal scale. The structure of unions and bargaining units established in the era of integrated national corporations divide workers who need to be connected with each other, even when they are members of unions.
The 24,865 union members at the GeorgiaPacific Corporation are divided among the Oil, Chemical (OCAW), Paperworkers (UPIU), Carpenters (UBC), Graphic Communications (GCIU), Office and Professional (OPEIU), Machinists (JAM), Utility Workers (UWUA), Electrical Workers (IBEW), Service Employees (SEIU), Teamsters (IBT), and Steelworkers (USWA). The 14,947 Gannett Company union members are divided among the Communications Workers (CWA), Newspaper Guild (TNG), IBT, GCIU, SEIU, Operating Engineers (IUOE), and JAM. The 3,761 unionized workers at the Archer Daniels Midland Company are divided among the Grain Millers (AFGM), Bakery, Confectionery and Tobacco (BCTW), UPIU, lUOE, USWA, JAM, IBT, Longshoremen’s and Warehousemen’s (ILWU), Laborers’ (LIUNA), OCAW, Retail and Wholesale (RWDSU), Auto Workers (UAW), IBEW, and Food and Commercial Workers (UFCW).
In the public sector, government workers may be divided among a dozen or more unions dealing with the same state or municipality. Within many workplaces there are multiple unions, plus a far larger number of workers who are not union members and who may not be eligible for union membership according to laws governing bargaining units. Many so-called local unions are not really located in the workplace, but rather are amalgamated locals representing dozens of different bargaining units in many different industries. One small Teamster local, for example, has 500 members spread throughout eastern Massachusetts in over fifteen bargaining units each with its own contract. A statewide SEIU local in Massachusetts has 10,000 members in 125 bargaining units, each with its own contract. Meanwhile, workers who produce the same kinds of goods or services are often divided up among many different national unions. And workers who work for the same company in different countries are almost entirely cut off from each other and therefore easily made to compete. The “American Separation of Labor” has been replaced by the “Global Separation of Labor” and the global “race to the bottom.” This structure allows workers once again to be put in competition with each other. If the labor movement is to fulfill its function – to do the things that lead workers to support and make sacrifices for it – it must facilitate cooperation between those workers capital puts in competition. Finding a structure to make that possible is essential for labor’s future. There are many examples of efforts to develop structures that connect workers across the lines of the “separation of labor,” We describe a few, not as models, but -rather to suggest what attempts to address he “separation of labor” might look like.
Organizing
The AFL‑CIO, with its new emphasis on organizing, has made some efforts to counter the ‘separation of labor.” For example, it gives challenge grants to unions that cooperate in organizing campaigns. Its “Union Cities” program attempts to draw many unions in a community into a cooperative organizing effort. However, these efforts face a problem once they begin to succeed. While unions may work together in the organizing process, ultimately the workers recruited are often parceled out among different unions, (This is not a new problem: AFL craft unions funded a massive campaign to organize the steel industry in 1919 but then undermined it by demanding that the hundreds of thousands who signed up be divided among the various crafts.) One interesting effort to address this problem comes from the United Health Care Workers of Greater St. Louis (UHCW), a network of healthcare workers formed to promote worker-to-worker communication and organization in the industry. Despite several attempts in recent years to organize particular facilities, the 58,000 workers in the St. Louis area’s 38 hospitals have no union representation. So the UHCW has turned to a sectoral strategy based on “Campaigns for Justice” in a growing number of area hospitals. The UHCW holds hospital meetings, organizes issue campaigns, publishes a newsletter, maintains a web site, and reaches out for cooperation to the rest of the labor movement and a wider coalition of community allies. They are also developing a strategy to keep themselves from being divided among different unions. In January, 1998 their steering committee proposed to local labor unions: “To avoid destructive competition and a likely ‘no union’ outcome, a consortium of supporting labor organizations might be formed under the St. Louis CLC. This UHCW support base could include any unions claiming a direct interest in the organization of healthcare workers and those simply in support of building a stronger St. Louis area labor movement. The employers are adversary enough. Inter-union warfare can cripple, if not destroy the real potential of this project. Those unions participating in the consortium could serve on an advisory committee, help recruit community support, and help establish the basis for a full-blown organizing program.”
Second Channels
The labor movement includes far more than just unions. Organizations like jobs with justice and the Labor Party provide “second channels” outside of official union structures through which workers can communicate. So do ethnic and racial networks, labor education programs, health and safety groups, community-labor coalitions, and many other elements of the wider labor movement. Such channels cannot replace, but can pave the way for direct relations among workers’ recognized representatives. Beyond that, an important part of restructuring the labor movement is to recognize organizations other than unions as central parts of the labor movement and to use these organizations to address the problems that unions themselves are having difficulty dealing with.
Local Level
In many workplaces, workers are divided among different bargaining units and different unions. One innovative approach comes from the collaboration of a Teamsters and a Steelworkers local near Chicago. They recently jointly organized nearly 800 workers at Ryerson-Tull Steel Services near Chicago. According to The Wall Street journal, “The Teamsters will represent 525 workers and the Steelworkers will get 272. The unions will jointly negotiate a contract and will share responsibility for it.” This represents in some ways a step toward the kind of shop steward councils that coordinate different unions in British workplaces.
National Level
As corporations developed a multi-industry, conglomerate structure in the 1960s, some unions turned toward multi-union coordinated bargaining. The AFLCIO’s Industrial Union Department (IUD) at one time had 80 coordinating bargaining committees in various companies, and in some cases, industries. While many of these became inactive over time, some continue: for example, OCAW, the Chemical Workers division of the UFCW, and Needletrades, Industrial and Textile Employees (UNITE) recently jointly negotiated a new contract with the Merck pharmaceutical company. (Most of the remaining IUD coordinated bargaining committees will be unable to function if, as seems likely, the forthcoming merged Machinists (IAM)/Auto (UAQ)/SteelWorkers (USWA) union do not allow its members to participate in them. The merging unions say they will create coordinated bargaining councils of their own, but whether councils sponsored by a single union can effectively overcome organizational rivalries remains to be seen.) Such bargaining committees or councils have no legal standing in labor law, and are often opposed by employers (who like workers divided into different unions) and by some union officials (who consider them a potential threat to their power). Nonetheless, they could point the way toward structures for dealing with “disintegrated corporations:’ especially if they were industry-wide, international, and integral to the life of member unions.
Transnational Level
Two sets of institutions have linked unions internationally. The best known are the two international federations that have contested since the end of World War 11 and the onset of the Cold War. The World Federation of Trade Unions is largely composed of left and Communist-led unions and has affiliates in 116 countries. The International Confederation of Free Trade Unions (ICFTU) includes unions that were strongly anti-Communist: it has 125 million members in 141 countries. Each federation tended to subordinate international labor cooperation to the struggle between Communism and anti-Communism. With the end of the Cold War, the World Federation of Trade Unions has largely faded from sight along with its Communist sponsors. The ICFTU has lost the Communist threat that provided its principal focus. Meanwhile, militant new unions from Brazil, South Africa, Korea, and other newly industrializing countries are joining or considering joining. So the ICFTU appears to be on the way to becoming a single global trade union federation representing most of the world’s unionized workers. Whether a transformed ICFTU can create a new role for itself remains to be seen.
Multiple, Overlapping Structures
What little discussion there has been of structural change in the labor movement has tended to polarize between those who emphasize the need to globalize unionism and those who emphasize localism. An increasingly popular catchphrase is “global unionism * ” The underlying concept here is that, since corporations have gone global, so should unions. This is envisioned as coming about either through unions in one country, organizing across borders in other countries, or through alliances among unions in the same industries in different countries which gradually grow toward merger. (Some US unions have considered simply signing up members and organizing locals in other countries – with little thought to the imperialistic implications of such a course.) So far global unionism is only a somewhat hazy vision.
Starting in the late 19th century, unions in related industries began to create less well-known international structures that are today known as International Trade Secretariats (ITSs). Until recently these were quite weak organizations, mostly holding formal meetings among top officials. In recent years, however, several of them have begun to develop important transnational campaigns to support workers being victimized by global corporations. For any given campaign, one ITS or an affiliated national union is designated the coordinating center. Within this structure, all unions representing workers in the target corporation are welcome to cooperate on an equal basis. Examples include global councils created for Bridgestone/Firestone and UPS. Such councils may hold the potential to extend the coordinated bargaining council model to a global level. While the goal of transnational labor cooperation is essential, there are serious problems in achieving it. Companies cut across industry lines, so that if each industry were organized by one union, workers in many companies would be divided among many unions. Collective bargaining structures are rooted in national labor law and practice in ways which make a transnational merger of existing unions extremely difficult. Existing national unions are already far too remote from local workplaces, and a global unionism that merely extended or merged existing structures would be likely to compound the problem of union bureaucratization and lack of participation.
The dis-integrated corporation can only be policed by workers who are able to organize themselves effectively at the worksite. The concept of “solidarity unionism:’ forcefully articulated by Staughton Lynd, which has been characterized as an idea which emphasizes local autonomy and community-level organization, opposes bureaucratic domination, and regards egalitarianism as a core value. This approach, although it advocates worker solidarity worldwide, concentrates on the local level. It suggests no structure for realizing that solidarity beyond the local context. Some unionists fear that this approach will only increase the tendency toward enterprise unionism. “These are the forms we should be building:’ Lynd writes in Living Inside Our Hope: A Steadfast Radical’s Thoughts on Rebuilding the Movement (1997): “Shopfloor committees, on the one hand, and parallel central labor bodies, on the other.” And even a local solidarity that extends beyond any one company is vulnerable to a global capital which plays off localities against one another.
For discussion, we suggest putting together these ideas and some others, we’d suggest, for discussion, the following approach: In a sense, all firms in the world are now in competition with each other, putting their workers in competition as well. Nonetheless, there are some workers who are much more closely connected than others. These generally include: those who work in the same workplace; those who work for the same employer; those who produce roughly the same goods and/or services; those who use the same skills; and those who work in the same geographical locality. The problem is that these connections often cut across one another. For example, those who work in the same company may work in different industries. So any organizational structure that links along one dimension is likely to divide along another. (When a union or an ITS conducts a campaign against a global corporation, it often turns out that unions that are supposedly in many different industries all represent workers in the corporation. Conversely, when industrial unions attempt to organize all the workers in an industry, they often find that many companies have only a small proportion of their workers in the target industry. Of course, there are “lagging” industries, of which the auto industry might be an example, that still maintain a more traditional industrial structure, just as some industries retained craft organization far into the industrial era.) The only way to cut this Gordian knot is to recognize the need for multiple, overlapping structures.
Let’s take a group of workers in any workplace. They need to be able to connect and coordinate with those they would otherwise compete with. That is likely to include workers in the same company, workers producing similar products, workers in the same occupation or using the same set of skills, and workers in the same local or national labor market. The workgroup should of course be free to choose their own representatives. And those representatives should be able to form coordinating councils or similar structures with whichever other workers they need to connect with. Similar structures might well be relevant at every level from local communities to the global economy. Such structures, especially those functioning transnationally, are unlikely to begin as vehicles for full-fledged collective bargaining. But that does not mean they can’t fight in a variety of ways to begin reducing competition among workers. One approach here is to promote and fight for minimum standards for everyone who works in a company, an industry, a locality, or an occupation. Such standards can be embodied in codes of conduct. Both unions and committees of ununionized workers can then fight for these minimum standards through collective bargaining (where it exists), legislation, and direct action.
The international anti-sweatshop campaigns pioneered by UNITE and other unions and allies operate in just this way. (This approach is somewhat like the 19th century “industrial legislation” approach, described above, that unions used before the development of modern collective bargaining.) The nascent global councils established for such corporations as Coca-Cola and UPS could serve as steps in the same direction. Functions such structures might perform include: information exchange; drafting and negotiating codes of conduct; monitoring codes of conduct; establishing agreements with companies regarding human rights; supporting the effort to raise the wage-and-benefit bottom in a company or industry; support resisting downward leveling for company/industry; supporting the movement toward common, collectively bargained standards; joint organizing; and participating in coalitions with other movements and organizations.
From Here to There
While connecting workers who are in competition with each other might seem like a no-brainer, the actual structure of organized labor often serves as an obstacle to such connections. Many local and national union leaderships jealously guard the lines of communication even among their own members and between their members and outside groups. (For example: the international chemical firm, Cytec, recently agreed to company-wide coordinated bargaining under the aegis of the IUD, but a major industrial union refused to allow its Cytec local to participate.) While some company-wide and industry-wide bargaining councils cutting across union boundaries do exist, they are the exception rather than the rule. When we have spoken about creating transnational links among trade unionists, local activists have repeatedly looked at us with skepticism and then blurted out, “We can’t even get in touch with people on the other side of town who work for the same company and belong to the same union.” The changes we have discussed would most likely shift power from national unions toward both the workplace and the global level. But they do not require, and would not be best achieved, by either the creation of a new national labor union center or the kind of split that established the CIO.
What is really needed is a process by which workers who are now separated begin to connect with each other. The structures we have discussed are simply vehicles to that end. While individual unions will no doubt continue to operate within bargaining unit definitions for some purposes, council-type structures can essentially ignore them and act on the need to connect all workers who would otherwise be in competition. Unions have an obligation to help workers set up such structures. Unions should encourage workers to form their own horizontal links by participating in central labor councils and in groups like jobs with justice that bring together workers across union lines. Unions should aggressively encourage institutions where workers – including both union members and members of community groups – meet each other and cooperate. Where unions fail or refuse to facilitate the efforts of workers to connect and coordinate with those with whom they need to cooperate, such links should be part of union reform programs – and workers should go ahead and make the links anyway. Such an approach builds loyalty not just to a particular organization but to the labor movement and the working class as a whole. This is increasingly crucial as workers have shorter and shorter stays with and less and less identification with disintegrated employers. This phenomenon known as “churning” severely weakens unions today: even when workers on a job are successfully recruited to join the union, they are likely to be replaced by new workers with no such commitment. An effective labor movement must equip workers to identify with the movement wherever they may work in the future. In the long run, such an approach requires a transformation in the character of unions in the workplace. There is no way that union officials can control the disintegrated corporation from outside the workplace.
The development of subcontracting and of contingent work requires direct counter-power at the worksite. While many trade unionists have been rightly wary of proposals that might legalize “company unions:’workers who are in different unions in the same workplace desperately need something like the British shop steward council or German works council approach that brings together representatives of different unions and of the ununionized within each workplace. Unions must turn themselves over to the members. The staff role should be redirected to providing workers with the skills, knowledge, and connections they need to run their own unions. Such a transfer requires a deliberate strategy. For example, contract demands should include released time for rank-and-file workers to do union work. To avoid having such released time become a management-manipulated perk, it must be combined with ways to ensure that local union officials work under the direct supervision of their rank-and-file. Local unions also need institutions like the “solidarity committees” developed to direct the kind of “in-plant campaigns” that have been used at Staley and elsewhere. This approach will inevitably come into a degree of tension with the collective bargaining structures created by national labor law and past union practice. It involves cooperation among different organizations which traditionally compete for bargaining rights. It also empowers new groups. The goal should be to make this tension constructive. For example, the original CIO split from the AFL would not be a good model today, especially in the context of globalization. Rather, the AFL-CIO itself should be transformed to encourage and support direct linkages among workers, rather than supporting its affiliates’ monopoly of workers’ lines of communication. (A simple starting point might be to make available an on-line database to help workers connect with all other organized workers in the same company, industry, and occupation around the world.) What motivations exist for the actually existing labor movement to move in this direction? First, it is a matter of institutional survival – without some change, the labor movement will simply continue its plunge to oblivion, even if it keeps trying to recruit more members as it falls. Second, just as a minority of AFL leaders at the local and national level supported the development of industrial unionism, so many of today’s union leaders who wish in good faith to represent the interests of workers are likely to recognize the obsolescence of today’s union structures and the need to move beyond them. Third and probably most important, workers who are being rendered powerless by the “separation of labor” have good reason to fight for vehicles to overcome it.
Restructuring of the labor movement is a process that can come from below. Demands for union restructuring – such as the demands for craft cooperation and industrial unionism -have been a frequent theme throughout American labor history. A more recent example comes from France, where “union density” rivals the low level of that in the United States and unions in that country are divided among three rival federations. Many of the mass strikes of the past few years occurred when workers formed local worker assemblies with participation of union members but outside of union control. They successfully pressured the national unions and union federations, often successfully, to stop pursuing their separate interests and begin to cooperate. Local workers can begin to reach out on their own initiative. A 1997 strategy meeting of the Teamsters for a Democratic Union proposed that local unions “Host a meeting of locals with a common employer … Extend steward and VOC [Volunteer Organizing Committee] networks across local union lines … Hold joint educational conferences for members and stewards from several locals.” Local unions could establish special “outreach stewards” to connect with other workers in their company, their industry, and their occupation across both union and national boundaries. After we presented some of the ideas in this article to a group of union activists, the president of a local union in a large corporation said, “We’ve talked from time to time about getting together a council of all the locals in the corporation, but the national union leadership never really wanted to do it. I think I’ll ask them one more time and if they don’t do it, we’ll just invite them all to get together on our own.” New structures are likely to find their prototypes in such “creative ad hocism” Structures aren’t everything. They are just a means for working people to use to meet their needs. Only a social movement in which people cooperate and act on their own initiative can pour life into any structure. Working people need to organize themselves and act on their own behalf. One aspect of that action is to demand that existing structures change to meet the needs of today and tomorrow. For workers, some new structural approach – whether the one outlined here or something else – is necessary to address the basic needs which have always forced workers to organize unions. Without it, they will be nearly as unprotected as they would be with no unions at all. For the labor movement, such change may be wrenching, but it may also be the only alternative to extinction.