How the internal flexibility of plants in the U.S. paper industry was increased and what it tells us about the effects of employment security.
There is a widely held view to the effect that internal flexibility is most reliably secured through the provision of employment security. The research reported here provides a test of this theory. Interviews were conducted with managers and, to a lesser extent, union officials in a judgment sample of 21 plants in the U.S. paper industry. With limited and specific exceptions, since the early 1980s paper mills have provided generally secure employment. Despite this, unions vigorously opposed management attempts to increase flexibility. In response, management used threats and intimidation to increase flexibility, with some considerable success. The theory that employment security produces internal flexibility requires some revision. (c) 2000 Elsevier Science Inc. All rights reserved.
Keywords: Employment security; Pulp and paper industry; Flexibility; Work effort; Labor relations
Sociologists' interpretation of the effects of employment security tends to differ from that of economists. Standard economic theory treats employment security as a limit on employer ability to adjust production costs to market conditions, with negative effects on profits and investment, as a barrier to shifts from less to more productive employments (e.g., Donges, 1985), and, especially if legislated, as a disincentive to hire (e.g., Hogan and Ragan, 1995).  It is recognized that the effects of employment security need not be wholly negative. Within a framework of compensating differentials, it may be offset by a lower wage rate. If product demand is fairly stable, any efficiency cost of employment security may be minimal. Government sector wage rates can be interpreted along these lines (Gunderson, 1995). Or employment security may be necessary to induce the acquisition of employer-specific skills (Williamson, 1975, p. 62). Still, the presumption within standard treatments is that employment security, pa rticularly if legislated, is harmful to the performance of the organization providing it, unless evidence to the contrary is provided (see Buechtemann, 1993, p. 9-12).
In contrast, sociologists tend to assert efficiency advantages from the provision of secure employment. It is argued that those whose employment is secure are more willing to acquire employer-specific skills (Hollingsworth, 1997, p. 272-273), to be flexible with respect to tasks (Hirst and Zeitlin, 1997, p. 226), and "to identify more closely with the firm as a community of fate and find it in their interest to contribute to its prosperity" (Streeck, 1997, p. 197), as a result of which they require much less costly supervision (Gordon, 1996).
In this paper, I focused on the issue of flexibility and employment security. My starting point is that, although cogent reasons have been advanced for expecting employment security to be associated with greater flexibility, the confidence with which the assertion is made greatly exceeds the quality of the evidence so far adduced in support of it.
2. Theory and evidence
Many labor contracts are distinguished by their indeterminacy (Marx, 1967, chapter VII; Braverman, 1974, p. 54-58; Offe, 1985, p. 20-25). If the quantity and quality of work cannot be contractually specified, there are likely to be gains in cooperation (Miller, 1992). Employers who push home any bargaining advantage are likely to create a resentful work force that responds in kind. This will involve, at a minimum, unenthusiastic work. Whenever the bargaining advantage shifts to employees, they extract contractual provisions that limit employer latitude. There are, it is argued, good reasons for employers to try to engage the loyalty of their workforces.
The most important method for doing so, the literature suggests, is the provision of employment security. Employers who establish a credible commitment not to layoff their workers are rewarded with employee collaboration in pursuit of improved productivity, including an acceptance of the need to modify job tasks as required by changing technology or other commercial considerations. Conversely, if employers layoff at will, employees respond by withdrawing their cooperation and building institutional protections-most saliently, industrial internal labor markets (Osterman, 1988, p. 63-65). These include detailed job descriptions, the aggressive defense of the demarcations implied by those job descriptions, and the enforcement of seniority in promotion and (in reverse order) dismissal. According to Osterman (1988), "For workers the system has the overwhelming value of creating security in the face of an insecure environment" (p. 65). Part of the appeal of this interpretation is, surely, that it is counterintuiti ve. The substantial literature on incentive pay systems (reviewed by Miller, 1992), for example, assumes that effort is directly tied to immediate (or near term) gains and losses rather than to employment security.
What evidence is there for the hypothesis that employment security increases effort, including flexibility? There are two kinds. First, countries, such as Germany, Sweden (e.g., Osterman, 1988), and Japan (Cole, 1979; Koike, 1988), are thought to provide more employment security, as part of a broader approach that relies on the "carrot," rather than the "stick," to secure compliance (Gordon, 1996, p. 65). It is claimed that this approach has produced economic performance that is superior to that of the United States, which is viewed as the paradigmatic case of employment insecurity. The second type of evidence is that those firms within the United States that have provided employment security seem to have a record of better than average performance (e.g., Kochan and Osterman, 1994; Miller, 1992, p. 118; Osterman, 1988, p. 67).
There are problems with each kind of evidence. Comparisons of macroeconomic performance depend, crucially, on the existence of a clear difference in the degree of employment security between Sweden, Germany, and Japan, on the one hand, and the United States, on the other. But the contrast has been overdone. Legislation notwithstanding, Swedish and German employers manage to layoff workers when they need to (Flanagan, 1987, p. 193-199). Employment security exists on a fairly limited basis in Japan (Koike, 1987, p. 308 -313; Taira and Levine, 1996, p. 146-149). By the same token, employer contributions to unemployment insurance in the United States contain a larger experience rating component than they do in Europe. And the legal protections against unfair dismissal, and the magnitudes of damages where a tort is thought to have been proven, are much greater in the United States. There may be a difference in employment security between European countries and the United States, but it is less stark than some of the literature suggests.
Assume for the sake of argument that the asserted differences in employment security are real. Clearly, Sweden, Germany, and Japan are distinguished from the United States by more than their degree of employment security, for example, in the structure of wages between and within firms, the organization of financial markets, fiscal and monetary policies, and so on (see Lawrence and Schultze, 1987). None of the comparisons that attribute international differences in economic performance to differences in employment security can remotely be described as using a multivariate method of analysis. Moreover, when "catch-up" through convergence of productivity levels is taken into account, relative U.S. economic performance looks a lot better (Baumol et al., 1989). And in the 1990s, the relative macroeconomic performance of the United States has improved considerably, while that of Germany and Sweden has worsened (Johnson, 1995).
The international comparisons, then, are interesting, but not conclusive. What of the data on the relative performance of firms within the United States? The number of cases reported is small. Only sketchy information on them is provided (e.g., Gordon, 1996; Kochan and Osterman, 1994). The performance of the firms that combined superior performance with secure employment subsequently declined markedly. (IBM is a good example of this.) By the early 1990s, the employment security they provided proved to be much more contingent and limited than it had seemed in the mid 1980s. The firm case studies typically cited are as inconclusive as the international comparisons. At the same time, some other cases studied suggested a payoff to management from "forcing" tactics, an element of which is deliberate increases in employment insecurity (Cutcher-Gershenfeld et al., 1995).
All this is to say that, given the importance of the issue in recent writings in economic sociology (and industrial relations), there is an urgent need for research that examines the effects of employment security more directly than has so far been the case.
3. The paper industry as a case study
Industrial internal labor markets exist in their most extreme forms in capital-intensive industries with strong unions (e.g., automobiles and steel). There are reasons for this. First, unions tend to be stronger where there is a substantial capital investment. A large investment effectively forecloses the option of closing a plant until its capital costs have been amortized (Simons 1948, p. 131). Second, industrial, internal labor markets were adopted in response to union pressures (Jacoby, 1985, p. 243-250). Third, process technology and its maintenance create a job structure conducive to the construction of a sequence of skills likely to be drawn into a job ladder or to an intense preoccupation with job demarcation issues or both.
Resistance to flexibility might be expected in the paper industry because it is highly capital intensive and has powerful unions (by United States' standards). At the same time, productivity growth is important. Between 1973 and 1994 output per hour grew by about 2.5% per year--about 75% over the entire period.  Managements are intensely concerned with improving processes. Other industry characteristics might be expected to produce employment insecurity. Prices fluctuate widely, and the importance of foreign competition has increased. From 1972 to 1988 the share of imports almost doubled--from 6% of industry production to over 10%--while the share of exports fluctuated with no trend until the mid 1980s, after which the export performance of the industry improved. The paper industry was losing ground to its foreign rivals until the mid- to late-1980s, after which its relative performance improved markedly. 
Like automobiles and steel, the paper industry produces commodities that have to meet world competition, has strong unions, and is capital intensive. These are the conditions thought to produce insecurity and inflexible work organization. In fact, work organization rigidity has been common in the paper industry. Like other capital-intensive industries, it contains two main sorts of manual jobs: operators, about two thirds of the labor force, who run paper machines, pulping, bleaching, and power control processes; and maintenance employees, who are responsible for the upkeep of the plant and equipment, electricians, pipe fitters, millwrights, welders, carpenters, and so forth.
In most mills, operator jobs have had these characteristics: 1) organization into lines of progression, each based on a technical process and/or machine (a paper machine, power recovery, pulping), with increasing skills from the bottom to the top of the line; 2) progression up the line on the basis of seniority rather than merit; 3) bumping arrangements between jobs that govern layoffs; 4) training mainly through example and explanation by the person in the job above; 5) strict definition of job tasks, with no obligation to perform any task other than those attached to the current position; 6) a refusal by operators to perform maintenance tasks; and 7) wages attached to jobs rather than individuals, so that individual merit cannot easily be rewarded.
Maintenance jobs have shared some of the preceding characteristics: 1) reverse seniority has applied in layoffs; 2) wages have been attached to jobs rather than individuals. But there are not the same lines of progression. Maintenance employees are either hired directly from the external labor market or complete apprenticeships within their employing plant. There are separate and distinct trades: those hired into a trade do not perform the tasks of other trades. In principle, people in individual trades do not assist each other. For example, if a mechanic fixes a pump, the electrician waits until the mechanic has finished and then reconnects the pump to the power source, rather than providing assistance. In practice, there may have been some informal cooperation.
These characteristics of operator and maintenance jobs embody all the rigidities lamented by Osterman (1988) and have been seen by industry management as an obstacle to achieving competitive productivity levels. Promotion on the basis of seniority sometimes delivers employees of doubtful quality into positions of some responsibility. Tying pay to seniority makes it hard to implement compensation systems that provide incentives for superior performance. Rigid demarcations mean that time that might be used productively is often, from management's point of view, wasted.
Despite this tradition of inflexible work organization, there have been significant changes during the last decade, or so. In a number of plants, flexibility has been increased in one or more of the following ways. Multicrafting means that craft employees take on a broader range of tasks. One or another less demanding skill (standard welding, oiling) may be assigned to the more demanding crafts (pipe fitter, millwright). Alternatively, it involves elaborate programs in which craft employees are trained in a range of skills, often classified as primary and secondary (e.g., the requirement that a certified millwright acquire basic skills in, say, pipe fitting). Mutual assistance means that craft employees are required to assist each other in completing tasks rather than waiting for other employees to complete their tasks before getting involved in whatever maintenance is underway. In many mills, the separate trades of electrician and instrument technician have been combined into a single classification. In add ition to increased craft flexibility, contracts had been changed in some mills to add minor maintenance to operator jobs. Finally, the rigidity of the line of progression has been replaced with broader operator responsibilities, pay tied to skills acquisition, work rotation, and teamwork.
4. Hypotheses, data, and method
In the context of the description of the paper industry provided above, the arguments on the effects of employment insecurity on flexibility of Osterman (1988), Miller (1992), and others suggest the following hypotheses:
1. The high degree of inflexibility that has characterized the bulk of the industry should have been produced by severe employment insecurity.
2. Success in increasing flexibility has been a result of the provision by management of enhanced employment security.
Several methods of data collection are available for addressing this problem. One is to study, in detail, a single or a few plants. This method's advantages are obvious. It has yielded some of the most valuable findings in industrial sociology and has been used to effect in studies of paper mills (Vallas and Beck, 1996; Zuboff, 1988). But it has a disadvantage. Plants in such industries vary markedly in technology, union organization, products, work organization, and personalities. Some paper industry plants were built almost a century ago (but have since been improved). Others are either newly built or have been massively upgraded. The market for specialty papers is different from the market for a basic commodity such as newsprint. The number of unions per plant varies, and some have no union at all. In the Northwest, the main union is the Association of Western Pulp and Paper Workers, reputed to pride itself in its militancy and formed from a faction that broke away from the major industry union, the Unite d Paperworkers International Union (UPIU). There is variation in forms of work organization and in the climate of labor relations. Although studying a very small sample yields greater detail, there is a risk of unrepresentative results; a risk that is compounded where all the plants are part of the same company. A method that better addresses the problem of representativeness (assuming adequate response rates) is mall surveys containing standardized items. This method has also generated valuable findings. But it raises problems of respondent conscientiousness (see, for example, Phillips, 1995, p. 155-156).
As a compromise between these two methods, the research reported here was designed to assure reasonable representativeness of the sample and to allow an in-depth discussion of issues of work organization. A judgment sample of 21 plants was constructed, designed to include a range of ages, union structures, and products. Plants were sampled from four different regions: New England (four plants in Maine and Vermont), the South (six plants in South Carolina, Tennessee, Alabama, and Georgia), the Northwest (five plants in Oregon), and the Midwest (six plants in Michigan and Wisconsin). In general (but there are exceptions), mills in New England and the Midwest are older than their Southern and Northwestern counterparts. All but one of the plants sampled manufactured pulp, paper, or both. The single exception specialized in the calendaring of paper produced in a nearby plant owned by the same company. The mills produced the standard range of products in the industry: market pulp, newsprint, fine paper, paperboard , specialty papers, and tissue. The sample included plants from 16 different companies. Interviews were conducted between December 1995 and February 1997, with one or more managers in each plant. They lasted between 1.5 and 2.5 hours. In several mills, union officials participated in the interviews or were interviewed separately. In about half the mills, the interview was followed by a plant tour, which allowed brief discussions with some plant workers at their work sites. In all, a plant visit would normally take about half a day. Clearly, this method does not provide the detail yielded in the research of Zuboff (1988) or Vallas and Beck (1996). Nonetheless, the semistructured interviews did allow an exploration of issues that the respondents defined as important.
Of the plants visited, three had been fairly or significantly flexible from the time they first opened. Most of the rest had increased their flexibility. Of the 18 mills without significant flexibility in the past, the contractual language had changed to allow the following: multicrafting, 11 mills; mutual assistance, six mills; skill consolidation, six mills; operator maintenance, eight mills; less rigid lines of progression (widened tasks or payment on the basis of certified skills rather than seniority), six mills. It should be noted here that I have been discussing provisions of the collective agreement. In several mills contractual flexibility provisions were either erratically enforced or not enforced at all. This was most clearly the case for operator maintenance, which was a very sensitive matter with the craft employees. Sometimes, both supervisors and operators avoided confrontations by ignoring the terms of the contract. I return to this issue later.
In most of the plants studied, then, as in the industry as a whole, there had been a history of work organization rigidity. But a small number of plants had always been substantially flexible, whereas the majority of the rest had secured a marked increase in internal flexibility.
5.1. Hypothesis 1: Was employment in the paper industry insecure?
Price instability, increased foreign competition, and rapid productivity growth suggest that employment in the paper industry should have been highly insecure. From 1980 to 1995 employment in paper mills [Standard Industrial Classification (SIC) 262] fell from 133,000 to 126,000, and in board mills (SIC 263) from 51,000 to 39,000. The reductions in employment in board mills were concentrated at the beginning of the period. Employment levels were much more stable at its end. Moreover, the amplitude of employment fluctuations in the paper industry has been less than in the manufacturing industry as a whole. 
More insight into the security of employment in the paper industry is provided by the plant data:
1. Since the early 1970s, there had been layoffs of one kind or another in nine of the 21 plants.
2. However, in only four of them had layoffs been permanent. And in the four plants that had laid off workers permanently, the numbers involved had been small, a score or so, in all but one plant.
3. In all plants, layoffs were governed by seniority. So even if, say, 100 out of 800 employees were to be permanently laid off, the bulk of the workforce was not at risk. Only those with the least seniority had a probability of being laid off that warranted worry.
4. Limited use of permanent layoffs took place despite the fact that 16 plants had downsized their labor force. They had downsized by attrition, sometimes encouraged by early retirement packages. This had been true in the plants that laid off permanently too. They had been obliged to lay off workers permanently because of the sheer volume required.
5. The four plants that had laid off workers permanently were distinctive. Three were located in New England and one in the Midwest. Each was an old plant with old technology. Consequently, they did not constitute immediate precedents for the labor force in other, younger, plants. Old technology indicates a problem. Employees recognize that. Employees in plants with new, or reasonably new, technology need not consider their jobs to be at risk of permanent loss. This lack of concern showed up in interviews with union officials in newer plants.
6. Employment insecurity has been linked to the shift to precarious contractual forms (Cappelli, 1995, p. 580-581). Temporary employees were used at eight of the 21 plants. In two, they had been used to facilitate a process of attrition. For example, at one plant labor requirements were going to fall when an older paper machine was taken out of operation after a new machine had been assembled. To avoid later layoffs, temporary appointments were made during the period leading up to the shut down of the old machine. But in most cases temporary employment was a probationary status en route to permanent employment. The number of temporary employees had sometimes been quite high in two of the plants. The largest use of temporary employees other than to allow attrition to continue while new equipment was being brought on line was in a mill in the Midwest with over a 1000 hourly rated employees and 85 mill trainees whose jobs were formally nonpermanent (in practice, they were rarely laid off). Clearly, those tempora ry employees had reason to feel insecure about their jobs. Conversely, their presence increased the security of the remainder of the labor force for whom the temporary workers were a buffer.
7. Subcontracting is an important element of work at paper mills. It is also identified as precarious employment. In practice, it is not a threat to permanent mill employees. Subcontractors were used where special skills were needed (e.g., construction), for major maintenance, and for security and janitorial services. Amounts of subcontracting varied and were sometimes large. But contracts and legal precedent prevent the use of subcontractors to displace permanent employees. My interviews suggested that subcontracting was an issue because it threatened the membership base of plant unions rather than because it threatened the security of current employees.
What, then, can we conclude about the level of employment security in the paper industry? The main conclusion is that only a relative handful of employees in the industry have had reason to worry about their employment. These results tend to disconfirm hypothesis 1.
5.2. Hypothesis 2: What produced increased flexibility?
The sample contained two nonunion mills. These had the most labor flexibility. Each had multicrafting, mutual assistance, consolidated electrical trades, operator maintenance, teams, and pay for skills in operator jobs. This suggests that the avoidance of unionization had been the most reliable route to flexibility. What about plants that had been rigid but had secured contracts that allowed greater flexibility (setting aside, for the moment, whether or not the contractual provision was actually applied)? Consider, first, the role of the general industry environment.
Most of the plants in this study had offered financial incentives to induce greater flexibility. But if the cost of the wage increase matches or exceeds the productivity gains from reduced rigidity, management is no better off. This seems to have been the experience in the International Paper Company (IPC). The company's first response to a history of poor performance was to negotiate broader job classifications in exchange for higher rates of pay (Birecree, 1993, p. 349). By the early 1980s, however, it was clear that the productivity gains yielded were not sufficient to address the company's problems. Poor performance in the 1960s and 1970s combined with the perception that future sales growth, in part, would have to be concentrated in highly competitive foreign markets led management to seek more substantial and quicker improvements in labor productivity through multicrafting, team production among operators, and a reduction in the barriers between craft and operator jobs.  The UPIU vigorously resisted , and the IPC responded with lockouts and replacement workers in particular disputes. It also successfully challenged the UPIU's bargaining tactics at the National Labor Relations Board.  IPC seems to have won a comprehensive victory.
The defeat of the UPTU by the IPC focused the minds of managements and unions in the rest of the industry. At the plants I visited, several managers emphasized the recent reasonableness of the union, a reasonableness apparently provoked by the concern that aggressive behavior on their part would lead to loss of members through plant closures, decertifications, or the permanent replacement of striking workers. (Management's view of the Association of Western Pulp and Paper Workers was less sanguine.) For example, a bitter strike at the Arkansas plant of one firm I visited had been terminated by union officials when they learned that management had a replacement labor force trained and ready to go into the plant the next day.  Managements had, in any case, been adopting a more aggressive stance in labor relations. They were emboldened in doing so by the IPC experience.
In the plant interviews, I asked how (if) internal flexibility had been increased. It turned out that aggressive tactics and threats had been major parts of management's armamentarium.
1. Only one manager (in Maine) emphasized, at length, the importance of the (largely Southern) nonunion sector of the industry in changing employee assumptions. But most managers said that relative mill performance entered into their discussions with the union, and the sample against which performance was compared usually included mills from the South.
2. At another mill contractual concessions had been secured after the company threatened closure of all the mills in its Western division.
3. The manager of a mill located close to an IPC plant at which conflict had been particularly bitter emphasized the role of that dispute in shifting the views of his own labor force.
4. There is a quite lively market in paper mills. From time to time companies decide to rid themselves of one or another mill, usually because it specializes in a product line outside of the main focus of the company. The mills sold, however, are rarely the jewels in the crowns of the companies selling them. Typically, they are old mills and are sold at a low enough price to make profitability possible. Sales of this sort are forceful signals to the workforce that there is something seriously wrong with mill profitability. These sales are also usually associated with a new, more forceful, management approach to labor relations through which any established pattern of indulgence is terminated or modified (cf. Gouldner, 1954). Four of the mills in the sample had changed hands within the decade or so preceding the interviews. In three, the change of ownership was followed by concessions on flexibility. In the fourth, the takeover of the single-plant company that owned the mill was recent. But, it became clear du ring the interview that the new owners were pressing for a more forceful approach to labor relations. A fifth plant, which was in very serious difficulty, had not been sold. But, during the crisis, and layoffs, that produced some contractual concessions, it had been rumored that the plant would be sold to a Canadian company known for its extremely aggressive approach to labor relations.
5. In the early 1980s, one Southern mill had secured substantial contractual concessions after a bitter strike. At this mill, the interview began with a group of three managers, but we were later joined by the presidents of the mill's four union locals. This joint interview indicated that, when the interview took place, relations between management and labor were reasonably amicable and cooperative. But the context for this cooperativeness had been set by the strike. The union leaders made it clear that the strike had been an appalling experience for all of them. They had, so speak, seen the abyss, drawn back from it, and were resolved to stay away from it in the future, if at all possible. Flexibility at this plant was a product of a very hard line taken by management. At no other plant in my sample did a strike seem to have played such a large role in changing work organization. But at a Midwestern mill, concessions had been secured after the company unilaterally imposed more flexible work arrangements, the union responded with the threat of a wildcat strike, and in response the company made it clear that it would fire and replace any worker who participated. Finally, although there had been no major strike in another Midwestern plant I visited, it turned out that the other Southern mill of this two-plant company had changed work organization after threatening to permanently replace strikers.
6. In 11 of the mills, managers reported using the threat of a decision to invest elsewhere, or to shut down the plant in whole or in part, as central elements of the process of extracting concessions on flexibility.
Overall then, aggressive tactics had played a major role in securing contractual concessions on flexibility from the union in most of the plants previously characterized by rigidity. There were, however, three exceptions. Two were in the West, where plants are generally newer. At one, the manager interviewed said that the plant's strong profit performance would have made such threats hollow. At the other, a new plant manager was bringing in a less paternalistic management style, and it appeared that the approach would become more aggressive when the tactical situation seemed suitable. The third plant in which the use of threats was denied was in the South, was clearly intensely paternalistic, and its manager contented himself with saying that his workforce was aware that they had to adapt to a competitive environment.
The research reported here was designed to test the widely held view that employment security increases the likelihood that employees will accept a more flexible organization of work. The results tend to contradict this view.
6.1. Employment security
The paper industry has a history of work task rigidity even though, I have argued, for the bulk of the workforce there has been, and there remains, very considerable security of employment. Permanent layoffs have been infrequent, and the proportion of the workforce vulnerable to them is small. Temporary layoffs have been more frequent, but where used on a fairly regular and predictable basis, seem not to constitute a huge problem for the employees involved. Temporary employees buffers the permanent workforce. Contracts and the law more or less effectively prevent management from using subcontracting to lay off and replace permanent employees. However, that the relative security of employment currently provided by U.S. paper mills did not produce flexible work might be explicable in ways that save the theory.
It could be argued that ambient insecurity in the labor market undermines any protections available to pools of privileged employees, such as those in the paper industry. Especially relevant here is the fact that, unlike the Sweden (in its pre 1991 heyday) described by Osterman (1988), those in the United States who do lose their jobs have negligible protections. But even if we set aside Flanagan's (1987) view that employment security in Sweden may have been exaggerated, were this argument correct it would make nonsense of the use as exemplars by Kochan and Osterman (1994) of U.S. firms that have provided secure employment and, in their view, drawn benefit from having done so.
Alternatively, it could be argued that the evidence of willingness to dismiss strikers where the law allows them to do so offsets the management's commitment to security of employment in nonconflictual periods. This may be so, but it should be borne in mind that a major element in the disputes has been the quest for the greater flexibility that employment security is supposed to provide. The strikes have represented, in part, the decision by some workforces to refuse cooperation, despite a plant history of secure employment.
However, it is possible that the degree of employment security since the early 1980s may be less important in determining worker willingness to be flexible than does a history of earlier insecurity. Employment insecurity in the 1930s and 1940s shaped the preferred contractual forms of organized labor in America. This is an important point, and I return to it.
6.2. Threats and intimidation
At several plants, it was reported that flexibility was to some degree provided on an informal basis, without contractual obligation. Informal cooperation of this sort might be seen as a gauge of the existence of a job security-based engagement in work. However, almost none of the managers interviewed regarded the amount of flexibility provided in this way as satisfactory. This is understandable. Informal cooperation can be withdrawn at the discretion of workers if something offends them. Moreover, the nature of the division of labor in a paper mill means that cooperation depends on the relations between different groups of workers as well as on relations with management. Operator willingness to perform minor maintenance depended on the forbearance of the relevant craft workers. The same thing applies to mutual assistance between trades or any limited excursion into the tasks of other crafts. Now, paper workers in a single plant often belong to different locals of the same union or to different unions altoge ther. Consequently, whatever noncontractual understandings might be developed are vulnerable to disruption from the evolving political relations of union groupings, whose mutual dislike, I gathered from interviews with union officials, is often as great as their dislike for management. Informal understandings could not provide to management a satisfactory avenue to the level of flexibility that it thought it needed.
It remains the case that, as noted above, the contractual right to demand flexibility was sometimes not exercised. Most commonly, operators declined to provoke craft workers by conducting maintenance, and supervisors opted for a quiet life by refraining from forcing the issue. In several plants, some craft employees that were trained in skills outside their craft specialty rarely exercised them.
Changing a contractual provision does not necessarily produce a corresponding change in behavior. But the importance of the contractual changes should not be underestimated. Provisions in the contract for multiskilling had usually produced a change in behavior: the unions took the changes very seriously, and the problem of supervisory reticence was itself in the process of being addressed through training or replacement of supervisors. Managements saw real progress in increasing productivity. The trends in trade referred to earlier (recognizing that other factors, particularly exchange rate changes, were at work too) suggest that their view was correct.
All this is to say that the data reported here indicate that there had been a real increase in internal flexibility in the industry. It is also clear that it was largely secured by threats and intimidation rather than though the provision of employment security. To what degree does this contradict the widespread theory that it is employment security that produces flexibility?
The paper industry may be atypical. Subsequent research can settle that matter. But there is no prima facie reason for regarding it as less typical than, say, the automobile industry, on which much research and generalization on flexibility rests (e.g., Hollingsworth and Boyer, 1997).
It is possible that the flexibility secured in the U.S. paper industry by using threats and coercion is less than is present in the other countries typically treated as models in this respect--Sweden, Germany, and Japan. But we have little detailed information on the degree of internal flexibility in those countries. Osterman's (1988) appraisals, for example, seem not to have involved fieldwork. Moreover, the forms of flexibility in those countries may depend on factors other than employment security. For example, the broad job classifications in Sweden emphasized by Osterman (1988) may have been mandated by the shift to centralized bargaining initiated by Swedish employers during the 1940s as a way of dealing with a major problem of wage drift (see De Geer, 1982, p. 138-141, 234-235).
It is also possible that there is more than one route to flexibility. For example, one of the plants in the sample seemed to have increased flexibility in all thee of the areas of work organization emphasized here (skilled worker tasks, lines of progression, operator maintenance), without conspicuous coercion. The change apparently came after a group of managers and union officials secluded themselves in a hotel for a period of time and hammered out a set of institutional changes. This plant's experience was distinct. But it remains the case that part of the context of its decision making process was competition from nonunion plants and IPC's hard line.
With all these qualifications in mind, it still remains the case that over the last decade or so many U.S. paper mills made considerable progress in increasing the flexibility in their use of labor. Those that did so appear to have depended heavily on coercion rather than employment security.
Gordon (1996, p. 61-65) contrasts, albeit briefly, the relatively virtuous, carrot-oriented approach to labor relations of Ford with the unambiguously vicious, stick-based approach of Caterpillar. He argues that, in the long run, the former is more efficient than the latter. It may be. But the evidence examined in this paper suggests that greater flexibility need not be a result of the "carrot" strategy. Quite the contrary.
At least since the 1970s, the paper industry has provided secure employment to the bulk of its labor force. But doing so did not produce a willingness on the part of its several unions to concede to changes in work practices. Among the 21 mills sampled here the main route to flexibility lay through either nonunion status or a combination of threats and intimidation. These data are not consistent with the theory that says that employment security leads to flexibility. But they may indicate that the theory requires qualification rather than outright rejection.
As Hollingsworth (1997) has emphasized, the institutional context--the embeddedness--of a set of practices may limit the capacity to change them. In this case, a history of employment insecurity in the past, and the structure of unions and of union preferences shaped by that history, mean that simply shifting from less to more employment security will not change union practices. In effect, in part of U.S. industry there is union custom, practice, and rhetoric, that limits cooperation. In this context, force may be the only route to change.
It is worth noting that several examples of the performance advantages thought to be produced by employment security involve plants at which a background of vicious conflict preceded the more cooperative relationship. This was true of the General Electric plant discussed by Osterman (1988, p. 75). It was also true of the Southern plant discussed earlier, at which management and the unions had rebuilt cooperative relations and increased flexibility only after a bitter labor dispute in the early 1980s. Osterman (1988) interprets the General Electric case as evidence of the improvements in performance produced by employment security and cooperation as opposed to confrontation. I am raising the possibility, instead, that the foundation for the cooperation may have been laid in the conflict that preceded it. Where an institutional logjam (so to speak) prevents change, the course of action with the highest probability of success available to management may involve coercion and intimidation that induces unions to m ake concessions, after which some effort can be deployed-including employment security--to rebuild community. This may not be a particularly congenial conclusion. But I think that there is a good chance that it is correct!
(*.) Corresponding author. Tel.: +1-514-398-6849; fax: +1-514-398-3403.
E-mail address: email@example.com (M.R. Smith)
(1.) The 1993 arguments are outlined in a general form in Buechtemann (p. 9-12).
(2.) The figure is for SICs 261, 262, and 263--that is to say, pulp, paper, and paperboard mills. The data were downloaded from http://stats.bls.gov/news.release/prin/t01.htm.
(3.) These data are by commodity (Group 341) and cover the whole industry, including converting plants. This underestimates trade in the section of industry of interest here, paper and board mills. The source is The OECD Standard Data Base for Industrial Analysis 1974-1993.
(4.) See Bureau of Labor Statistics Bulletins 2445 and 2481.
(5.) Birecree (1993, p. 348) reported a judgment of a former IPC director of employee relations to the effect that, in the 1960s and 1970s, the demand for paper in the United States had been relatively price inelastic--this allowing generally congenial management-union relations.
(6.) In a dispute with Champion International, the UPTU delayed contract settlements in all seven company mills so it could initiate a company-wide strike. The company rescinded its demands for concessions. In response to the same tactic, IPC took the union to the National Labor Relations Board, and won.
(7.) How the labor force was found was not clear in the interview. Birecree (1993, p. 356), noted that at least one consulting firm developed a capacity to provide contract labor during a strike.
Abraham, K. G., & Houseman, S. N. (1993). Job Security in America: Lessons from Germany. Washington, DC: Brookings.
Abraham K. G., & Houseman, S. N. (1994). Does employment protection inhibit labor market flexibility? Lessons from Germany, France, and Belgium. In Rebecca M. Blank (Ed.), Social Protection Versus Economic Flexibility: Is There a Trade-Off? (pp. 59-94).Chicago, IL: University of Chicago Press.
Baumol, W. J., Batey Blackman, S. A., & Wolff, E. N. (1989). Productivity and Amen can Leadership: The Long View. Cambridge, MA: MIT Press.
Birecree, A. (1993). Corporate development, structural change, and strategic choice: bargaining at International Paper Company in the 1980s. Indust Rel 32, 343-366.
Braverman, H. (1974). Labor and Monopoly Capital: The Degradation of Work in the Twentieth Century. New York: Monthly Review Press.
Buechtemann, C. F. (1993). Introduction: employment security and labor markets. In C. F. Buechtemann (Ed.), Employment Security and Labor Market Behavior (pp. 3-66). Ithaca, NY: ILR Press.
Bureau of Labor Statistics Bulletins 2445 and 2481. (1997). Washington, D.C.
Cappelli, P. (1995)., Rethinking employment. Br J Indus Rel 33, 563-602.
Cole, R. E. (1979). Work, Mobility, and Participation. Berkeley, CA: University of California Press.
Cutcher-Gershenfeld, J., McKersie, R. B., & Walton, R. E. (1995). Pathways to Change: Case Studies of Strategic Negotiations. Kalamazoo, MI: W. E. Upjohn Institute.
De Geer, H. (1982). Job Studies and Industrial Relations: Ideas About Efficiency and Relations Between the Parties of the Labour Market in Sweden 1920-1950. Stockholm, Sweden: Almqvist & Wicksell.
Donges, J. B. (1985). Chronic unemployment in Western Europe forever? World Economy 8, 353-372.
Flanagan, R. J. (1987). Labour market behaviour and European economic growth. In R. Z. Lawrence & C. L. Schultze (Eds.), Barriers to European Growth: A Transatlantic View (pp. 175-229). Washington, DC: Brookings.
Gordon, D. M. (1996). Fat and Mean: The Corporate Squeeze of Working Americans and the Myth of Managerial "Downsizing." New York: Free Press
Gouldner, A. W. (1954). Patterns of Industrial Bureaucracy. New York: Free Press
Gunderson, M. (1995). Public sector compensation. In G. Swimmer & M. Thompson (Eds.), Public Sector Collective Bargaining in Canada: Beginning of the End or End of the Beginning? (pp. 103-134). Kingston, Ontario: IRC Press.
Hirst, P., & Zeitlin, J. (1997). Flexible specialization: theory and evidence in the analysis of industrial change. In J. R. Hollingsworth & R. Boyer (Eds.), Contemporary Capitalism: The Embeddedness of Institutions (pp. 220-239). Cambridge, UK: Cambridge University Press.
Hogan, S., & Ragan, C. (1995). Employment adjustment versus hours adjustment: is job security desirable? Economica 62, 495-505.
Hollingsworth, J. R. (1997). Continuities and changes in social systems of production: the cases of Japan, Germany, and the United States. In J. R. Hollingsworth & R. Boyer (Eds.), Contemporary Capitalism: The Embeddedness of Institutions (pp. 265-310). Cambridge, UK: Cambridge University Press.
Hollingsworth, J. R., & Boyer, R., (Eds.). (1997). Contemporary Capitalism: The Embeddedness of Institutions. Cambridge, UK: Cambridge University Press.
Jacoby, S. M. (1985). Employing Bureaucracy: Managers, Unions, and the Transformation of Work in American Industry, 1900-1945. New York: Columbia University Press.
Johnson, K. H. (1995). Productivity and employment: review of the evidence. In The OECD Jobs Study: Investment, Productivity and Employment (pp. 37-62). Paris, France: Organisation for Economic Cooperation and Development.
Kochan, T. A., & Osterman, P. (1994). The Mutual Gains Enterprise: Forging a Winning Partnership Among Labor, Management, and Government. Boston, MA: Harvard Business School Press.
Koike, K. (1987). Human resource development and labor-management relations. In K. Yamamura & Y. Yasuba (Eds.), The Political Economy of Japan, Vol. 1, (pp. 289-330). Stanford, CA: Stanford University Press.
Koike, K. (1988). Understanding industrial Relations In modern Japan. London: Macmillan.
Lawrence, R. Z., & Schultze, C. L., Eds. (1987). Barriers to European Growth: A Transatlantic View. Washington, DC: Brookings.
Marx, K. (1967). Capital: A Critique of Political Economy, Vol. 1. New York: International Publishers.
Miller, G. J. (1992). Managerial Dilemmas: The Political Economy of Hierarchy. Cambridge, UK: Cambridge University Press.
OECD Standard Data Base for Industrial Analysis 1974-1993 (1995). Paris: OECD
Offe, C. (1985). Disorganized Capitalism: Contemporary Transformations of Work and Politics. Cambridge, MA: MIT.
Osterman, P. (1988). Employment Futures: Reorganization, Dislocation, and Public Policy. New York: Oxford University Press.
Phillips, R. (1995). A comment. In R. J. Adams, G. Betcherman, & B. Bilson (Eds.), Good Jobs, Bad Jobs, No Jobs: Tough Choices for Canadian Labor Law (pp. 148-165). Toronto, Ontario: C.D. Howe Institute.
Simons, H. C. (1948). Economic Policy For a Free Society. Chicago, IL: University of Chicago Press.
Streeck, W. (1997). Beneficial constraints: on the economic limits of rational voluntarism. In J. R. Hollingsworth & R. Boyer (Eds.), Contemporary Capitalism: The Embeddedness of Institutions (pp. 197-219). Cambridge, UK: Cambridge University Press.
Taira, K., & Levine, S. B. (1996). Employment flexibility and joblessness in low-growth, restructured Japan. Ann Am Acad Political Soc Sci 544, 140-153.
Vallas, S. P., & Beck, J.P. (1996). The transformation of work revisited: the limits of flexibility in American manufacturing. Soc Problems 43, 339-361.
Williamson, O. E. (1975). Markets and Hierarchies: Analysis and Antitrust Implications. New York: The Free Press. Zuboff, S. (1988). In The Age of the Smart Machine: The Future of Work and Power. New York Basic Books,
|Printer friendly Cite/link Email Feedback|
|Author:||Smith, Michael R.|
|Publication:||The Journal of Socio-Economics|
|Date:||Nov 1, 1999|
|Previous Article:||Marriage contracts and the law-and-economics of marriage: an Austrian perspective.|
|Next Article:||An integrated model of consumer materialism: Can economic socialization and maternal values predict materialistic attitudes in adolescents?|