Chapter 6

Steel Labor Relations

The steel industry has a long and difficult history of labor relations issues. The systemic inability of the two parties to resolve their differences at the bargaining table and the resulting series of strikes in the 1950s and 1960s, in particular, is now seen to be the tragic failure of the American integrated steel industry, opening the door to minimills and importers to undermine product and pricing mechanisms that constituted ground zero for Big Steel. For that reason, this chapter gives extended coverage to the mechanisms and issues that lay at the heart of the conflict.

Perspective on Steel Industry Labor–Management Relations

As mentioned above, the steel industry has been one of the icons of the mass production industrial age and industrial unionism. Huge facilities and huge investments in capital equipment have been the characteristics of the business and at times the bane of its existence. Skills within steel companies were basically organized around a tight hierarchy of engineers and executives at the top. Very few CEOs of steel companies in the 20th century did not come from the engineering staff. Beneath this hierarchy, there was a mass of unskilled and semiskilled industrial workers and laborers.

Frederick Taylor, who wrote the famous Principles of Scientific Management in 1911, did his original work in the steel industry and sought the complete decomposition of jobs into simpler, less skilled components. The steel industry took this philosophy to heart and tried to implement it more systematically than any other major industry. The eventual codification of the whole system of jobs in the steel industry, the Co-Operative Wage Study (CWS) system, gave skills and wages in steel a uniquely hierarchical and fragmented character that still largely defines the industry today.

On the other side, the cycles of collective bargaining in the steel industry include a number of false starts within a long-range objective on the union side to move away from its classic adversarial role and gain the right of the union to a voice in the industry. The USWA leadership thought that they were sold a bill of goods by the companies in the steel crisis bargaining of the mid-1980s. Though later they largely made those concessions whole, the parties reentered the change cycle in the early 1990s when on-going difficulties in the industry compelled a new, different, and union-led agenda for collective bargaining. It was partially successful. The early adoption of these new directions took place in the smaller sites installed as part of the US–Japanese co-ventures but there was very little change in core steelmaking operations of the major steel companies. It was only after the 2001 crisis and a new cycle of bankruptcies that the opportunity for more union voice arrived, with additional contract changes and roles being forced on to both the labor and management groups.

The Labor Process in Steel

The steel industry in the nonunion era was dominated by the impact of technological change on workers and the production process as the industry entered the mass production age: The mechanization of steel production process in the early 20th century altered skill requirements in important but often unexpected ways. Machines replaced human muscle power in moving materials through the production process and thus speeding up the process. But, on the whole, instead of breaking down old crafts, machines had simply replaced some older, heavier manual routines. Blast furnaces were much larger, but still involved fundamentally the same chemical process. Open hearth furnaces had surpassed the limited output of puddling furnaces, but not the need for metallurgical knowledge. Rolling mills of massive proportions could handle much more metal, but still needed the careful attention of experienced workers. As a result, the old dichotomy of craftsmen and laborers gave way not to a mass of mindless machine-tending jobs, but to a new hierarchy of jobs with less of a gap between the least and the most skilled and with considerable demands for responsibility and judgment from workers filling them, perhaps with greater occupational homogeneity.

Previous analysis of these developments that emphasizes simply the stripping away of shop floor discretion misses the point.1 Close-knit, co-operative work teams at the center of the production process headed by a few key workers like blast furnace keepers, open hearth melters, and head rollers exercised a range of independent decision making on the job and a freedom from either direct supervision or machine-pacing. They were more similar to the shop floor status of the 19th-century craftsmen. For many more, the range of their discretion was much more narrowly circumscribed but nonetheless crucial to the most efficient operation of the machinery. Overall, given the disappearance of so much unskilled labor, the average skill level in the new steel plants probably went up, not down.2 Operators and rollers saw themselves as unique and more closely identified with their Mill or machinery than with the Company. This would present a large hurdle to be overcome when Statistical Process Control (SPC) and quality measures were introduced in the 1980s and 1990s.

In the first decades of this century, gone were most of the shovels, wheelbarrows, and small army of brawny laborers and in their place were a few handfuls of men who manipulated gears. Under the old style it took 150 men per 24 h to operate a 200-ton furnace and the output was 1.33 tons per man. Under the new system, it took only 60 men per 24 h to operate a 550-ton furnace and the output was 9.17 tons per man.

The companies, however, had done more than simply change individual work processes. They brought them together in huge sprawling complexes that tightly linked each phase of production with the others. Individual departments and processes within departments nonetheless retained some distinctive rhythms and occupational requirements. Some processes were naturally linked, notably the coke ovens, blast furnaces, and open hearths, while others especially the smaller rolling mills were based on batch production. Some, like blast furnaces and rolling mills required steady, routinized feeding of furnaces or machinery, while others required more erratic bursts of frantic exertion, especially tapping furnaces. There was, in short, a great variety of work processes in a steel plant that did not necessarily lead to a common occupational experience for workers in the industry. There was an informal pecking order in the mill and Steelmaking was King.

In many ways, the sum of all the technological innovations was greater than its parts. Besides the changes in individual stages of production, it was the overall coordination of the plants that was so remarkable. Every plant was a maze of tracks for numerous railways, cranes, and conveyors. Raw materials moved along these tracks to coke ovens and blast furnaces; liquid pig iron was swept off to the open hearths at the end of giant cranes; steel ingots were shunted off to the rolling mills, where cranes and conveyors carried steel forward. In contrast to past practices, there was far less remelting and reheating of the metal as it moved through the plants. Early 20th-century steelmaking did not involve an assembly line, but there was definitely an integrated flow-through. For the most part, the plants ran continuously through the year, rather than on the seasonal basis that had characterized much of 19th-century production. In fact, several departments ran nonstop 24 h a day and 7 days a week. Mechanization therefore brought not only greater volume of production from the new facilities, but also greater speed and intensity and, for the workers, greater pressure to keep up.

The Steel Industry Wage Structure

People with only a little knowledge of steel industry labor contracts are stunned by the detailed and complex wage and job structures they observe. Few need more convincing that “rigid” labor contracts are a major problem in the industry than to read the wage clause of the collective agreement. In the steel industry, this whole system is known as the Cooperative Wage Study or CWS as it is called. It is now regarded by most as a union construct; however, in fact, the wage and job structures originated with the company management prior to being incorporated into collective agreements.

The origins of CWS go back to a World War II US labor law decision to address wage disparity disputes. Contrary to widespread belief, the Labor Board did not order the parties to develop a joint job evaluation program. In fact, neither the board nor the steel panel expected that the job evaluation route would be used by the parties. There is even less in the record to indicate that systematic job evaluation was a practical possibility in view of the union’s hostile attitude to any such approach.

The Cooperative Wage Study was set up in Pittsburgh by a group of 12 steel companies in 1943. It was exclusively comprised of company representatives, assisted by private consultants, the American Associated Consultants Incorporated. The stated objective was to “determine the wage-rate situation in the companies; determine what it should be; and determine ways and means by which to bring about such corrections as were found to be necessary.”3 Again, contrary to widespread belief later, the union was not a part of the original design of CWS; in fact, it was generally opposed to such undertakings as a preemption of collective bargaining.

The basic objective of CWS was to discover the underlying wage structure of the steel industry—not of any particular plant or company, but of the industry as a whole—and to construct a manual that would reflect this structure and identify deviations from it. These variances would be considered as inequities from the underlying wage structure, to be marked out for elimination through processes agreed upon in collective bargaining with the union. From the outset, the CWS approach was predicated on the assumption that there existed a general wage structure for the basic steel industry, which, in varying degrees, would reflect job relationships in the individual steel plants and companies. The individual plant and company rates for a large number of jobs, when assembled, supported the thesis that there was a general wage structure for the industry and indicated that the companies were on the right track in working toward an overall industry program rather than individual company plans to eliminate wage inequities. In contrast to the usual approach to job evaluation, which tries to “correct” a wage structure through the application of a manual with preconceived weights, CWS architects believed that designing a wage structure was beyond the scope of job evaluation. They preferred to find and use the weights for various job factors that had been developed in the steel industry through the impact of the common labor market, the ups and downs of business cycles, hundreds of thousands of individual judgments and individual bargaining as well as collective bargaining.4

Factor weighting is considered to be a trial and error process, with a given set of weights subject to testing with benchmark jobs. Adaptation of existing weights to a particular configuration of benchmark jobs or even modification of the maximum points allotted to factors is not uncommon and is considered good job evaluation practice. Once the factors and weights have been determined and the benchmark jobs evaluated, the remaining jobs are interpolated by reference to them.

As a matter of preference, it is not necessary to use given factor weights at all in job evaluation. In the factor-comparison method of job evaluation, no fixed weights are assigned to individual characteristics of a job. Factor weights are developed from the benchmark or key jobs so that the accepted total rate for each such job is distributed among the factors in terms of cents per hour rather than points. A feature of this plan is that there need not be a range or a limit set for any job characteristic or factor and predetermined degrees need not be defined. When the factors have been assigned their money weights for the given job, it is necessary only to add the amounts for all factors and the sum becomes the money rate for the job. An advantage of the factor-comparison method for an industry like steel, where there is an unusually wide spread in rates, is that it allows unlimited room at the top for any exceptional factor. Jobs are ranked in each factor by comparing their relative worth to that of the key jobs. This is a flexible method that permits significant differences in the relative weights of factors for different applications of the plan. Thus, the CWS approach was not unique in its refusal to accept fixed factor weights in the evaluation of steel jobs. It was unique, however, in the method it used to arrive at relative factor weights that would best reflect the steel industry wage structure.

Most other job evaluation plans had been constructed around the wage structure typically found in light manufacturing industry. Such plans typically allocated 50% or more of the maximum points available to skill factors and relatively little weight to responsibility, effort, and working conditions. This distribution of factor weights in other industries resulted in highly skilled jobs such as tool and die maker, pattern maker, and machinist being ranked at the top of the job hierarchy, which is generally in accord with their position in the plant wage structure across general manufacturing.

Steelworkers in the mills however had a great deal of responsibility for materials. Executives themselves said that 90% of the operational problems in steel mills involved the flow of materials. Workers also still required major physical exertion and labored in unusually difficult conditions of heat, dust, and danger.

However, the steel industry was also characterized by highly interdependent processes performed in a group rather than on an individual basis and involving the use of heavy and expensive equipment. Therefore there are a number of jobs that rank higher than skilled craft occupations in the steel plant age hierarchy. The main difference between such steel jobs as Roller—Bloom Mill; First Helper—Open Hearth; Blower—Bessemer and Heater—Plate Mill, as compared with typical skilled maintenance jobs, is that in addition to a high degree of skill these jobs also involve considerable responsibility for the continuous flow of operations and for expensive tools and equipment, which is not normally required of tradesmen. Under conventional job evaluation plans, such production jobs will hit the ceiling provided by the maximum points for responsibility factors without attaining any significant differential about highly skilled maintenance jobs. These jobs would be adjusted out of line on the high side and their rates red-circled under most plans, despite the fact that there is general agreement among the union, management, and the worker themselves that top skill production jobs deserve to be ranked well above skilled trades jobs in the plant. The suspicion that radically different factor weights were needed to reflect the steel industry wage structure led to the search for an entirely new manual rather than an attempt to modify existing plans to fit the needs of the industry.

As stated, a distinctive factor in the CWS manual is “Responsibility for Operations.” A steel plant is typified by a highly interdependent series of processes, most of which are performed by crews working with large machines and equipment. The smooth flow of work from one process to another is, therefore, of the greatest importance, because faulty operation on one piece of equipment has serious effects on processes both before and after that particular operation. This can lead to underutilization of both equipment and labor. This responsibility is apt to center on the top working member of the crew whose skill, knowledge, judgment, and motivation is typically the effective determinant of productivity. Roller, Heater—Soaking Pit, and First Helper—Open Hearth are illustrative. To meet this element, “Responsibility for Operations” was introduced. Later, as a result of the introduction of a new level in this factor during negotiations with the union, all crew members of production units having a share in the responsibility for operations received some credit under this factor. Below is an example of a CWS Job Description and its Classification from the 1950s.

In this ranking scale, in a normal manufacturing plant laborer and machinist jobs sit at opposite ends of the scale. These jobs are found frequently in both light and heavy industries and should be evaluated properly by any plan. Jobs with exceptionally high skills, like tool and die makers, are favored by the light industry scale, while heavy industry jobs with high responsibility do not get proportional credit for the responsibility factors The net result is that jobs like Roller and First Helper are rated about the same or even lower than highly skilled craft jobs by light industry plans even though they have always earned far more than those jobs in the steel industry. In steel, the high end production jobs went to the top of the hierarchy.

The 1959 Steel Strike

The critical moment for steel labor–management relations was the 1959 round of negotiations, which saw the most pivotal of many strikes. The core was a dispute about the authority of management and work rules on the shop floor.

The 116-day strike by the USWA against U.S. Steel and the other large steel companies was the most important strike in the history of the industry. It has been seen as the all-time great strike over “restrictive work rules.” Observers on all sides characterized the conflict as the classic industrial relations struggle over workers versus managements rights and authority on the shop floor. The fundamental issue was article 2-B of the collective agreement concerning local practices. However, the most comprehensive study of the 2-B provision reveals an ironic twist to the otherwise classic positioning of labor and management over employment security.5

Section 2-B of the agreement was first negotiated in the 1947 contract. At the time it received little attention from both the union and the employers. They were both more concerned about postwar wages and capacity utilization in recessionary times. However, by 1959, it had become the core issue in the struggle over shop floor rights and management prerogatives.

Section 2-B was a six-paragraph clause in the contract recognizing the existence of “local working conditions” either “practices or customs,” “written or oral.” It protected these local work practices and provided only two circumstances in which they could be changed: by mutual agreement between management and the union; or, unilaterally by management if the “basis” for working conditions changed. Subsequent arbitration cases made the explicit formulation that the meaning of “basis” for change had to be technological change or formal changes in work methods. For instance, arbitrators ruled that crew sizes were covered under Section 2-B. Management could not defend crew reductions solely by citing its management rights clause, work load levels, and time studies. Cost reductions alone could not justify a crew reduction. Management was however given permission to eliminate a job as long as it could prove that the basis for the job had changed, either by the introduction of new technology or by an alteration in the production process. The union frequently grieved in order to document a paper trail of working conditions and staffing levels. Gradually it created a written record of what staffing levels were at each level of operation at every plant across the whole industry, which could be invoked as a precedent later. The union argued that management had no right to eliminate or change a job solely on the basis of economic factors where no changes in technology or the production process were involved.

In fact, management across the steel industry in the 1950s had been eliminating thousands of jobs—employment in the industry peaked in 1953— by combining job functions or doubling up jobs, and not through technological innovation. The union believed management only wanted to slash jobs and speed up production, and management believed that the union wanted to interfere with all efforts at improving productivity.6

Section 2-B neither protected steelworker jobs from automation, nor limited management’s ability to introduce new technology and machinery that would result in a reduction of staff or a reorganization of the production process. Section 2-B did, however, limit management’s right to cut staffing levels and reorganize jobs without a corresponding introduction of new equipment or technology. The latter was precisely what companies like U.S. Steel sought to do in their older mills in the 1950s and why Section 2-B assumed such importance.

Having rejected the new technologies of the BOF and continuous casting, during the 1950s the steel companies were seeking to modernize by “rounding out” their existing facilities. In many cases, new equipment could not be introduced without seriously disrupting the product flow. Therefore, steel management went on an active campaign to reduce costs by reducing existing staffing levels and adding new responsibilities to jobs without introducing new equipment or technology. Section 2-B prohibited this and the union used the clause to block job slashing and production speed-up. As a result, management concerned about their increasingly inefficient and aging mills came to view the section as the disastrous unintended result of the 1947 bargaining, which not only limited management’s authority to determine such critical shop floor issues as crew sizes and force reductions, but also undermined its cost reduction strategies.7

U.S. Steel negotiators tried in the 1959 bargaining to get a provision in the industry-wide contract to overrule local practices and informal agreements. Older mills had thousands of work practices. A national contract could not possibly codify all the individual work practices and informal agreements at all these mills. Even at the level of the individual mill, all local working conditions could not be summarized in an agreement, because local work practices and particular understandings were too numerous.8

The USWA leadership turned the strike into a crusade against a return to the conditions of the 1930s—complete unilateral authority of management on the shop floor and the end of job security. Eventually the union triumphed. They received a modest wage increase and the iconic Section 2-B remained.

With hindsight, there are two incredible ironies in the story of the 1959 Strike and Section 2-B. First, if management had indeed embarked on a major technology innovation program and pursued the BOF–Concast route they would not have been inhibited by the union contract. Failure to innovate would eventually lead to the companies’ demise. Second, looking backward, the post-1993 contracts now explicitly require management to engage with the union and the workers over technological change and changes to production methods, the exact things that management rights in the mass production era were attempting to block.

The 1959 Strike had another political legacy. The next time steel bargaining came around in 1962, the parties again came to an impasse. John Kennedy intervened to avoid another major economic disruption. He convinced the union to accept a zero wage increase. He asked the companies in return to constrain prices. They initially rejected his demand outright. Then, under threat of a government takeover, the steel companies reluctantly agreed and a settlement was reached. However, the bitter taste and deep distrust between the steel industry and government persisted for years to come.

Centralized Bargaining in Steel

Centralized bargaining has always been the gold standard in trade union circles. The principal objective of centralized collective bargaining was to “take wages out of competition.” Wage increases are more readily and securely achieved, and companies are forced to compete on grounds of efficiency and innovation, not on downward wage pressure.

The postwar steel industry was an outstanding period of high Wagnerism, named after Senator Wagner of New York, who introduced the New Deal labor legislation in the US Senate in 1935. The labor and management parties were to pursue their competing self-interests by negotiating complex collective agreements with multitudinous work rules. Both sides wanted it that way. Distributive bargaining and compliance-based employment relations were the norms of the day.

The steelworkers union was also strongly positioned for industry-wide bargaining because of the national-level jurisdiction of the federal National Labor Relations Board.

Mangum and McNabb9 analyze the dynamics of the rise and fall of centralized bargaining in the U.S. Steel industry in terms of the relationship of product markets and labor markets. According to the authors, the turn of the century steel industry was different from other oligopolies such as automobile manufacturing, in that steel had a higher degree of product homogeneity. Steel was steel and consumers were unconcerned with the source as long as it was cheap and in plentiful supply. There was also a substantial degree of technological homogeneity. Steel was produced throughout the industrialized world by almost identical coke ovens, blast furnaces, open hearth furnaces, and finishing mills, all staffed by workers with an almost identical complement of skills. In addition, because of its weight and consequent transportation costs, the product for steel was regional.

The skills required were virtually identical in every mill but the companies did not compete for workers within the same labor pool. Typically access was intergenerational. The skill requirements were industry specific, making steel could be learned nowhere but in a steel mill. There was some mobility only in maintenance, transportation, and clerical skills. The skill increments between jobs were so small that they could be learned on the job by substituting during illnesses and vacations. Hiring on at an unskilled level and advancing by seniority made sense both to management and to labor.

When the United Steelworkers began unionizing the industry in 1937, wage-rate inequalities between workers performing the same work in different plants became a serious issue in bargaining. Unionization of the entire industry came in the early 1940s and with it, as described above, the introduction of the Cooperative Wage Study which eliminated wage-rate differentials. The standard job classification system and accompanying wage-rate structure had a unique complexity. There were dozens of specialties eventually compressed into 32 job classes. The auto industry assembly plants by contrast had only assemblers plus some skilled maintenance personnel on each assembly line. In a steel mill, each division had many different occupations ranging through all job classes.

This set the stage for industry-wide wage bargaining. The management side was also interested in wage equality within regional markets in order to gain wage as well as price leadership. Price uniformity was more easily maintained when the biggest single input cost, wages, was also uniform. There was no product heterogeneity to justify price differentials and no technological heterogeneity to diverge production costs.

Industry-wide bargaining did not become a reality until 1955. Demand for steel was high and expected to remain so for many years. The dominance of U.S. Steel and oligopolistic pricing reinforced the union’s strategic interests, though the latter was coming under pressure from imports.

In response to the perceived foreign threat, the US labor and management parties in the early 1970s negotiated a qualitatively new deal, the Experimental Negotiating Agreement (ENA). The union gave up the right to strike for an extended period in exchange for automatic wage increases—3% for productivity plus cost-of-living (COLA). The drop in steel demand and escalation of inflation in the 1970s’ OPEC Crisis doomed the generous terms of the ENA to failure.

Eventually the ENA was not renewed and industry-wide bargaining itself was abandoned by the companies’ refusal in 1985 to bargain any longer through the Coordinated Committee Steel Companies (CCSC).

Pattern bargaining was the new form of centralized bargaining begun in 1986, where a “lead company” was chosen by the union to set the industry pattern from one bargaining round to the next. This sustained a substantial degree of wage-and-labor-cost uniformity across the industry, mostly due to the union’s efforts.

New Directions for Bargaining in the 1990s

Ironically, the implosion of U.S. Steel companies in the 1980s saw leadership in the industry’s collective bargaining approach coming from the union side in the “New Directions” bargaining policy led by Lynn Williams, the Canadian who became International President of the USWA, beginning in the 1993–1994 negotiations. This has ushered in an unprecedented level of labor–management cooperation in the steel industry and significant innovation in local workplace practices.

The Union summarized its principles as the Twelve Points of Light for the industry:

1.Full Union Involvement in all strategic decisions including the development of the company’s business plan, acquisitions, design, and implementation of technological change, environmental issues, and access to all information pertaining to these issues. This included participation from the shop floor to union nominees on the Board of Directors.

2.Employment security guarantees.

3.No concessions in wages and benefits, and full and continuing access to the company’s financial books and records.

4.Neutrality in union organizing campaigns and card check recognition.

5.Revitalized apprenticeship and training programs.

6.Investment commitments for plant modernization.

7.Health care cost reductions through managed care without penalties or cost shifting to employees.

8.Funding of legacy costs (unfunded retiree benefits).

9.Corporate guarantees to secure pensions and other benefits in case of company mergers, takeovers, sales or joint ventures, including successorship protection.

10.Settlement of contracting out issues.

11.Joint public policy agenda including rebuilding the nation’s infrastructure, industrial investment, national health care, trade policy, and labor law reform.

12.Long-term (six-year) agreement.10

The above points constituted perhaps the most forward looking and ambitious bargaining program any union has ever tabled and successfully negotiated. However, the results were mixed due to external events and internal organizational dynamics.

Through the Cooperative Partnership Agreements (CPA) negotiated first in 1993, the United Steelworkers of America attempted to introduce its vision and model for more participative and productive work systems throughout the basic steel industry. The CPA structured joint labor–management planning, problem-solving, and decision-making processes at every level of the organization—shop floor, department, plant, division, and corporate, extending to union appointees for each company board of directors. It also prescribed participants from both the union and management at each level and included substantial training for all participants. In addition to improving relations and performance the USWA sought to increase its access to business information and influence over strategic as well as operational decision making.

Research was subsequently done to assess the effectiveness of the industry-wide contractual language and the centralized union-driven approach to workplace reform at the local level.11

The reasons for the US industry’s problems included antiquated technology, an industry-wide bargaining structure that separated wage increases from productivity improvements, and a bureaucratic system of job classifications and work rules designed for an environment of market stability and the promotion of labor peace. This Taylorist system was composed of narrowly defined jobs, individual incentives, standardized procedures, strong managerial controls, and extreme specialization.12

Earlier, during the 1980 negotiations, the USWA and the industry attempted a new approach toward dealing with their productivity and quality problems. Much industry and public commentary at the time drew attention to Japanese Quality Circles and other team-based problem-solving practices. The new effort in steel was called Labor Management Participation Teams (LMPTs) and a two-page LMPT Experimental Agreement outlined a set of organizing principles.

While the Experimental Agreement was nationally negotiated, recognition was given to the fact that many of the industry’s productivity and quality problems existed at the plant level in the organization, and process improvement through widespread workforce participation was required to solve them. Teams typically met weekly to work on problems ranging from quality and productivity to safety and the work environment, and union members along with their supervisors received up to 40 hours of training in subjects such as problem solving, statistical process control, conflict resolution, meeting skills, and group dynamics. Some companies also had plant- and department-level union–management committees to administer the team process and solve problems at a higher level, although the emphasis of the LMPT processes was at the team or shop floor level.

Later, the Cooperative Partnership Agreements attempted to provide a framework for: (1) joint decision making at all levels of the corporation, from the shop floor to the board of directors; (2) full and continuing access to business plans, records, and information including products, pricing, markets, capital spending, cash flow, finance, mergers, acquisitions, joint ventures, and new facilities; (3) jointly implementing new work systems and technology; (4) comprehensive education and training to provide the skills necessary for effective problem solving and participation. Specifically, improvements were targeted in quality, service, productivity, competitiveness, profitability, and safety. The agreement also advanced the goal to make the workplace “more equitable, less authoritarian, and less stressful.”13

Finally, the agreement established a joint process for workplace redesign including the implementation of self-directed work teams and a shifting of responsibility for daily operations, planning, scheduling, and administration from supervision to bargaining unit members. It also outlined a process for the implementation of technological change including advance notice and information. If the parties were successful in implementing workplace redesign, then the union received joint decision-making authority over the effects of any technological change including the number and types of jobs required by the new technology, the skill and training requirements, the inclusion of new jobs in the bargaining unit, new work rules or operating procedures, and any health, safety, or environmental initiatives.

Further, this top-down contractual approach to industry-wide workplace reform provides an alternative to both the company- or plant-specific patterns of diffusion of high performance and participative work systems seen in most other industries, and to the earlier LMPT efforts in the steel industry itself.

The CPA called for extensive sharing of business information with the union. The majority of local unions reported receiving information on markets (80%), business plan development (75%), competition (70%), and financial performance (62%). However, information on long-term strategy (40%) was less forthcoming. Workplace redesign efforts had taken place in 42% of the plants, yet only 4% had attempted the provisions for a Joint Technology Change Program.14 The experience of the USWA in implementing Cooperative Partnership Agreements across the US basic steel industry has been mixed. The agreements appear to have generally increased the sharing of business information and to a somewhat lesser degree many local unions reported positive results in the areas of quality, safety, decision-making, productivity, and cost reduction. However, in most cases the contractual provisions for CPA implementation have not been completely fulfilled. In particular, training was far less extensive than anticipated by the agreements, and many plants lack the departmental joint structures and team efforts on the shop floor. Further, provisions for joint workplace redesign and the union involvement in technological change have not been extensively implemented. Comparisons of successful and unsuccessful CPA implementation show significant differences in the extent of training, sharing of information, development of mid-level structures to support shop-floor teams, and managerial involvement. When these components were in place, the CPAs provided positive results for the locals unions involved. Further, three main barriers to more complete implementation of the Cooperative Partnership Agreements were identified. The first was managerial resistance, the second was the resources and priority given to the CPA by the International union, and the third was the nature of the agreement itself.

A challenge for the USWA with the CPA was both to get a commitment through collective bargaining that the parties could not walk away from, and to ensure the resources would be made available to implement the agreement effectively.

These mixed results on the CPA should not be interpreted as failure by the USWA. Rather they are a glass half full, diffusing workplace reform at a rate comparable or better than diffusion in other industries.15 It should also be said that there was a level of reluctance on the part of the local union leadership to accept all of the responsibilities that came with the information flow and involvement with the business side of the steel companies. They intuitively resisted the prospect of accountably for decisions that might be unpopular.

Negotiated Restructuring in the 21st Century

However, much of this effort on both sides of the table encountered major difficulties in the 2001 steel crisis and subsequent bankruptcy of most of the so-called re-constituted mills. The International President of the union, Leo Gerard, took over the mantle of union leadership for the industry by his dramatic new efforts with the International Steel Group (ISG) agreement described below.

The steel crisis of 2001 brought a new wave of restructuring to the steel industry. The basic industry in North America and Western Europe never recovered from the destabilization that followed the Asian Financial Crisis of 1997–98 and the surge of cheap and dumped imports from Asia and Eastern Europe which followed. The most dramatic development has been the USWA agreeing to large-scale consolidation of the industry, as well as new collective bargaining agreements with the new owners, particularly ISG. These new agreements are qualitatively different from their predecessors in the 1980s. The newest deals are instead agreements with the minimill management teams being brought in to now run the remaining integrated mills.

Union President Leo Gerard decided to create a revolution in the U.S. Steel industry. In an audacious move, in the midst of bankruptcy proceedings involving pension rights, he decided that the Union would reorganize the basic steel industry by creating its own steel company, find new investors, and hire a new management. The new entity became for a time, the largest basic steel producer in the United States, the International Steel Group (ISG) comprising the former LTV Steel, Bethlehem, and National until later taken over in turn by ArcelorMittal.

The terms of the new agreement became a reference point for the whole industry. Implementation and administration of collective agreements always present at least as many challenges for the labor and management parties as the negotiation of the contract in the first place. It remains to be seen how the new agreements will play out in practice and over time. However, there is no doubt the USWA-ISG was the most dramatic change in collective agreements in the history of the industry.

The Role of the Local Union in Steel

In their classic study of the local union, Sayles and Strauss identified the central focus and core activities of local unions as contract administration, grievance handling, and bargaining of local issues under the direction of national and international unions. However, the position and role of local unions are being transformed by efforts to involve employees more directly in problem solving, decision making, and improvements in workplace operations. Many of the principles guiding the structure and governance of local unions that grew up under the mass production Wagner New Deal model of industrial relations had to be changed. The changes are not easy and often do not come about without considerable political debate and internal and local–national union conflict.16

The most important change in the past 50 years has been an expansion of direct local union involvement in firm governance and management, challenging the boundaries of managerial authority established by the Wagnerist industrial relations system. This includes development of business strategy, new investments, product development, choice and introduction of new technology, training, job design, quality assurance, subcontracting, business planning, supplier selection, and work reorganization. Unions have also demonstrated their ability to add value on the shop floor by providing leadership, internal organization, and networking that brings effective coordination and implementation capability.

Other research has traced the differential levels of involvement of unions in the restructuring of the UK and German steel industries and the implications of such involvement on outcomes. The relative lack of institutions to allow for union involvement in restructuring decisions in the UK steel industry is contrasted with the more proactive responses by the German IG Metall union enabled by a much richer set of institutions in the German steel industry.17

For these reasons the forms of employment relations and contractual rules were critical to the evolution and performance of the industry. It is a deeper story than simply the aggregate capital investment and macrolevel labor market indicators relied on by Mangum and McNabb.

In Europe and Japan, stable, secure job tenure encourages reliance on internal labor market practices and creates a strong incentive for employers to invest in and support training activities. Co-determination rules arising from national labor laws confer a degree of authority on shop floor workers who, possessing the necessary skills, authority and long experience, intervene actively to solve operational problems as they arise. Their more extensive training makes it less likely that such problems will arise in the first place. The resulting “industrial culture” is common to both machine producers and users in Germany and facilitates the process of user–producer interaction leading to highly successful machinery design and use. Furthermore, the characteristics of the German workplace come to be reflected in German-designed technologies that assume a well-trained, highly skilled, relatively autonomous operator who will remain associated with the machine for a long period of time. The fact that such conditions do not exist to nearly the same degree in the North American users’ plants—and the resulting institutional distance between German machine tool producers and their North American users—is ultimately responsible for the implementation difficulties that have been documented.

Pension and Legacy Costs

Pensions are perhaps too appealing an item in labor management negotiations in heavy industry. There is a temptation to respond to a high union wage demand by tabling a counter proposal with a modest wage increase but a big move on the pension. It saves both sides of the table the trouble of struggling through the complex processes and sociology of high performance workplaces. It sells well with the union membership. It is appealing for the CEO as well because the costs are amortized over 15 years and he knows he is not going to be around when the bill comes in.

There are two problems with this. First, over time what happens is that the pension plan takes on the whole weight of the economic adjustment process. This is more than it can reasonably be expected to bear. Secondly, there are jurisdictions such as the USA and parts of Canada that have a Pension Benefits Guarantee Fund (PBGF) as an insurance device to guarantee workers benefits in cases of companies going bankrupt. The availability of this mechanism however can become a moral hazard for management. If a steel company blows out the fund by offloading its obligations, then someone is going to ask the moral hazard question and employers may wind up being told to self-insure.

It is not too much to say that, unlike in Europe or Japan where industrial policy imperatives have been the key drivers, in North America it has been legacy costs that have been the fundamental determinants of restructuring in the steel industry. The issues of pensions, particularly with regard to distressed companies and industries, raise fundamental questions about workers' rights, industrial restructuring, and regulation of corporations. Previously little known agencies such as the Pension Benefits Guarantee Fund (PBGF) in Ontario and the Pension Benefits Guarantee Corporation (PBGC) in the United States suddenly arrive on the front pages of the newspaper as central actors in determinations of whether companies are considered viable or not.

Pension and bankruptcy problems in the US industry 20 years ago foreshadowed problems experienced subsequently. Bankruptcy proceedings under Chapter 11 of the bankruptcy legislation in the US have in fact served as the vanguard for workers rights and restructuring disputes in much of industrial America. The leading study of the relationship between industrial restructuring and pensions in the US concludes that pension legislation and regulation simply does not provide adequate protection from economic imperatives because there is no general regulatory framework through which competing claims can be resolved and rationalized. The gap is the absence of an industrial policy, both for steel and more broadly for the economy as a whole.18

The US Employee Retirement Income Security Act (ERISA) of 1974 was designed to ensure that employers’ unilateral or negotiated promises of retirement income were honored and legally protected. The act did not require employers to offer pension plans as part of their employment benefits or to offer a certain level of benefits. What the Act did was provide workers with legal rights to promised benefits while ensuring that the vesting, funding, and management of private pension plans meet a mandated minimum standard of operation. However, as the steel industry cases have made graphically clear, at root of what is in dispute is the appropriate balance between economic imperatives of corporate restructuring and corporations’ obligations to their workers and society.

The 1986 bankruptcy of LTV was the watershed event in the struggle over the pension rights of steelworkers and corporate restructuring in the steel industry. LTV Corporation declared Chapter 11 bankruptcy in an attempt to shift its unfunded pension obligations to the federal government’s PBGC.

At the time, LTV blamed its situation on the precipitous fall in domestic demand in the 1980s, the rise in steel imports and the lack of government policy to facilitate restructuring. Through Chapter 11, the company continued to operate and in the late 1980s regained profitability. The cost advantage of going into bankruptcy, particularly the transfer of unfunded pension obligations to the PBGC, tax concessions granted to the whole industry by the federal government and sustained improvements in productive efficiency and labor productivity through the bankruptcy proceeding significantly improved LTV’s competitive position vis-à-vis other steel companies. The USWA, on behalf of current and former employees (pensioners), argued that retirees deserved more than the PBGC’s mandated and guaranteed minimum monthly pension benefits. The PBGC claimed that the company could still pay most of its pension liability if it were to gain more concessions from union members. To that end, the PBGC applied to the courts to return most of LTV’s pension plan assets to the company. It lost in federal bankruptcy court but won in the Supreme Court. The precedent of the LTV case has become a temptation for the whole steel industry to reduce costs by dumping off their pension obligations.

The company, notwithstanding an operating profit at the time had $4B in debt, some $2.3B of that related to pension obligations. Its pensions were chronically underfunded, unable to meet current or actuarially forecast, contractually agreed-to pension obligations. At the time, the LTV situation was viewed as the classic endgame for declining industries, the inevitable result of management accepting unrealistic wage demands in return for continuity of production. Unions were seen to have significant power in the short run because management was caught without investment options to go elsewhere. By this logic, union wage demands drove LTV into bankruptcy. The single and simplistic metric to support this theory is the often-cited, then as now, average wage for steelworkers versus the average wage for manufacturing or the economy as a whole. An equally if not more plausible explanation is that three combined circumstances in the 1980s were at play: the rationalization and restructuring of production capacity; the use of early retirement pensions by the company as the central means of reducing its workforce; and the rapid penetration of imports into the domestic market. LTV had consolidated Jones and Laughlin, Youngstown Sheet and Tube, and National Steel, reducing its combined workforce from 85,000 to 20,000. Rationalization added enormous long-term debt, largely the result of having to shed so much labor in so short a period of time, to an already indebted company that had absorbed the debts of the separate companies. It can be argued that LTV deliberately used its pension plans as the means of shedding excess labor through early retirement and plant closing pension benefits, believing that only this strategy would mollify the union’s objections to the massive rationalization of employment. LTV acted legally in pursuing its merger strategy, but it did so knowing that the pension costs of this strategy might ultimately be borne by the PBGC.19

This is the case for considering steel company actions as strategic bankruptcy, for LTV then and other companies more recently. In effect, this strategy turns the public agency (PBGF or PBGC) into a creditor of the company. In the process, the LTV case and other current cases risks bringing the public agency to the brink of bankruptcy.

Bankruptcy proceedings are a very crude instrument for dealing with the chronic and complex issues facing the steel industry. Where does one find a decision rule to be fair to the secured creditors and sensitive to the circumstances of the company and its workers and retirees? An industrial policy or corporate restructuring policy might provide such a guide. In their absence, a judge is left to make policy in the context of competing interpretations of the logic of bankruptcy.

Today, the pension side of the legacy cost problem for steel companies has been brought under reasonably manageable control, though not without major controversy and radical redrawing of several of the major pension plans. The other side of the legacy cost issue—health benefits—now constitutes arguably a more daunting problem and will occupy major attention in steel industry bargaining for the foreseeable future.

Steel Unions and Labor Agency

Trade unions have been an ambient part of the economic and cultural environment of the steel industry around the world. They will continue to be so but what actual role they play in the globalized industry is very uncertain.

Academics refer to this as the question of labor agency. At the most general level, there are three potential things steel workers and unions can do. All of these involve having a “voice” function. Unions can advocate for employees around wage, benefit, and safety issues within the workplace, as reflected in their labor agreements. They can be an active voice in broader industry and environmental issues as membership organizations. Where unions are nonexistent or marginal, workers can merge with broader social groups as collective community voices in economic, social, and environmental affairs.

As revealed in the story told in this chapter, the role of unions in the steel industry is not the medieval siege between fixed positions that is often offered up in presentations by the media. Steel industry industrial relations have a lot of moving parts. We have seen, for instance, over 30 years of struggles in the US industry where government regulatory powers, specifically in the pension and benefits area, resulted in the United Steelworkers gaining strategic power outside the workplace as a party in the restructuring of U.S. Steel companies.

Looking forward, the importance of environmental regulation is now the lead driver in the industry’s technology investments for the coming decade. The intersection of union and civil society alliances will be a key determinant of the industry’s horizon and prospects.