Employers and the politics of skill formation in a coordinated market economy: Collective action and class conflict in norway

Bowman John R. Queens College and the CUNY Graduate Center 12 2005 33 4 567 594 This article uses a case study of vocational training in Norway to explore the conditions under which employers will cooperate to increase the skill level of their workforce. It generates two sets of insights into the political economy of training in coordinated market economies. First, by demonstrating that cooperation among employers was a recent achievement that required the creation of specific, targeted mechanisms, it suggests that a cooperative outcome is difficult to attain, even amid the generally hospitable institutional environment characteristic of these economies. Second, it demonstrates that the employer collective action problem with respect to the initial training of apprentices differs from that posed by the additional raining of experienced workers. This means that even within a single economy, there are several varieties of training politics, each characterized by different patterns of cooperation and conflict; this further complicates the task of creating a coordinated training regime. employers vocational training coordinated market economies varieties of capitalism collective action sagemeta-type Journal Article search-text 10.1177/0032329205280925POLITICS & SOCIETYJOHN R. BOWMAN Employers and the Politics of Skill Formation in a Coordinated Market Economy: Collective Action and Class Conflict in Norway JOHN R. BOWMAN This article uses a case study of vocational training in Norway to explore the condi- tions under which employers will cooperate to increase the skill level of their workforce. It generates two sets of insights into the political economy of training in coordinated market economies. First, by demonstrating that cooperation among employers was a recent achievement that required the creation of specific, targeted mechanisms, it suggests that a cooperative outcome is difficult to attain, even amid the generally hospitable institutional environment characteristic of these econo- mies. Second, it demonstrates that the employer collective action problem with respect to the initial training of apprentices differs from that posed by the additional raining of experienced workers.This means that even within a single economy, there are several varieties of training politics, each characterized by different patterns of cooperation and conflict; this further complicates the task of creating a coordinated training regime. Keywords: employers; vocational training; coordinated market economies; vari- eties of capitalism; collective action 1. INTRODUCTION This article uses a case study of vocational training in Norway to explore the conditions under which employers will cooperate to increase the skill level of their workforce. Training policy has emerged as a central topic among compara- tive political economists not only because of the intrinsic contemporary impor- tance of creating a high-skilled workforce, but also because the decision whether or not to train workers presents employers with an important coordination prob- POLITICS & SOCIETY, Vol. 33 No. 4, December 2005 567-594 DOI: 10.1177/0032329205280925 © 2005 Sage Publications 567 lem whose solution sheds light on the institutional makeup of advanced capitalist economies. All firms benefit from a skilled workforce, but individual firms may choose not to incur the costs of training workers, hoping to "free ride" by hiring workers trained by other firms. In line with the classic collective action problem, the result of individually rational training decisions is likely to be a workforce with a low level of skills. Differences in the way in which this problem is resolved are one of the indica- tors that underlie Hall and Soskice's important distinction between "coordinated market economies" and "liberal market economies"1. In coordinated market economies, like Germany and Norway, employer coordination problems, such as whether or not to contribute to the training of the labor force, are generally resolved through nonmarket mechanisms. As the result of what Hall and Soskice describe as "institutional complementarities," nonmarket coordination in one arena, such as wage determination or corporate governance, makes nonmarket coordination in other arenas, such as employment training, more likely.2 My analysis of recent training politics in Norway generates two sets of insights into the political economy of training in coordinated market economies. First, it demonstrates that the hospitable institutional environment and institutional comp- lementarities characteristic of these economies are not in themselves sufficient to generate a cooperative training outcome. Although it is in many respects a coordinated market economy, Norway's regime for providing initial training to young workers was not always marked by much cooperation among employers. Increased employer cooperation in the provision of training after 1994 was the result of collective appeals by employer organizations, institutional innovation, and increased selective incentives. Even in coordinated market economies, then, cooperation among employers may be a contingent achievement that requires the deliberate creation of additional, more targeted, institutional mechanisms. Second, I demonstrate that the employer collective action model that domi- nates the literature provides an incomplete picture of the training decisions faced by individual firms and generates a distorted analysis of training politics. In par- ticular, the politics of training is not only about achieving increased cooperation among employers in the filling of apprenticeship places. It is also about the level of additional training that experienced employees receive. This kind of training generates different patterns of conflict and cooperation than the training of initial apprentices. In Norway, employers who cooperated in the training of apprentices 568 POLITICS & SOCIETY This research was supported by a City University of New York PSC-CUNY Research Award. I am grateful to Arvid Fennefoss and Sveinung Skule for providing me with valuable insights on training policy and politics and for their generous sharing of research material. Of course, they bear no respon- sibility for the analysis presented here. I also thank Erling Barth, Alyson Cole, Arvid Fennefoss, Geir Høgsnes, Irving Leonard Markovitz, Mieke Meurs, Erik Olin Wright, and the members of the editorial board of Politics & Society for their comments on an earlier draft of this article. Finally, I am grateful to Fredrik Engelstad and the staff of the Institute of Social Research in Oslo for their continued hospitality. fought a bitter strike in opposition to a proposal to guarantee experienced workers a funded right to leaves of absence to pursue additional training. This contrast between initial and continuing training is significant because in a world in which workers can expect to change jobs several times in their work career, where production processes within individual sectors are becoming more dynamic, and where "lifelong learning" has become a new buzzword, the relative importance of further training can only increase.3 If further training is driven exclusively by the short-term needs of individual firms, as it seems to be in Nor- way, then the institutional foundation of the collective, long-term outlook that is allegedly characteristic of coordinated market economies may be weakened. The Norway case also demonstrates that the challenges involved in creating institu- tions that move control of continuing training beyond the individual firm are con- siderable. These issues are masked by the dominant model of training politics, which focuses almost exclusively on the initial training of workers entering the labor force and implicitly conflates the logic of initial training with that of continuing training. I begin the article with a critical examination of the literature on the political economy of training. I describe the logic of the collective action problem faced by employers and provide a brief look at some of the ways in which it can be resolved, drawing heavily on the German case. I also elaborate the contrast between the problems of providing initial training and further training. The second section of the article analyzes apprenticeship in Norway, focusing on the changes effected by Reform 94, a major reform of secondary education, which transformed the incentives faced by employers and increased the level of cooperation. The third section analyzes the politics of further training in Norway. I distin- guish two situations. The first of these involves training in which the employer has considerable control over the content of the training and the identity of the work- ers who receive it. I argue that Norwegian employers have been generally sup- portive of providing training under these conditions. It is more likely to conform to employer needs, and it contributes to worker morale in a context in which cen- tralized wage setting makes it difficult to reward individual workers with pay raises. However, when these conditions change and employer control over further training is reduced, the logic of the training decision faced by employers' changes. The main empirical focus of this section is on the fate of labor movement demands for a statutory right to financed leaves of absence to pursue training, which first surfaced in Norway in the mid-1990s. Under this proposal, all workers would have a right to pursue further training. The decision about which workers would receive training and what kind of training they would receive would no lon- ger be made exclusively by employers. Unlike the initial training provided to apprentices, this training was not viewed by employers as a collective good that would benefit the labor market. Rather, the proposal for an employee right to fur- JOHN R. BOWMAN 569 ther training was seen as a challenge to important managerial prerogatives. These demands eventually succumbed to a combination of united employer opposition and lukewarm labor support. Their fate demonstrates some of the obstacles in the way of instituting a reform that increases coordination in the provision of further training, even in settings where initial training is relatively coordinated. Most of the analysis of contemporary developments is based on extended interviews with officials in Norway's employer associations and labor unions. I conclude by suggesting that the Norwegian case reveals three distinct train- ing decisions and three different patterns of training politics. The decision to take on the initial training of apprentices conforms to the conventional employer col- lective action model. In Norway, this arena is marked by cooperation both among employers and between the labor unions and employer organizations. However, this cooperation among employers was contingent on recent institutional changes. Second, the decision to provide additional training to selected experienced work- ers is not experienced as a collective action problem by employers. Employers appear willing to provide this training on the basis of the interests of their firm. It also does not appear to be the site of significant labor-capital conflict. The third variety of training politics was in response to the proposal for employers to finance the kind of "lifelong learning" that may not be tied to the short-term inter- ests of individual firms. In this case, the only employer coordination visible was the organized defense of their managerial prerogatives. 2. THE LOGIC OF VOCATIONAL TRAINING Underlying almost all analyses of the politics of training is the idea that the decision whether or not to train workers presents employers with a collective action problem. This strategic aspect of a firm's decision to invest in training was first identified by A. C. Pigou: "[S]ince workpeople are liable to change employers . . . socially profitable expenditure by employers in the training of their workpeople . . . does not carry a corresponding private profit."4 In this view, firms face a collective action problem--a "poaching externality."5 They are better off if they can draw from a pool of skilled workers, but either the fear that other firms will "poach" their workers or the hope that they themselves can engage in poach- ing leads them to refrain from investing in training. The outcome is a workforce that possesses a suboptimal level of skills. At the heart of the poaching externality problem is the fact that employers can- not capture all of the gains from the training that they provide workers. Workers are able to use the additional training to raise their skill profile and to seek higher wages on the external labor market. Thus any market conditions or institutional factors thatreduce the abilityof other employers either to use or to recognize these skills, or that otherwise constrain workers'abilityto seek alternative employment, also reduce the danger of poaching and increase the incentive of employers to invest in training. 570 POLITICS & SOCIETY One important variable affecting employer incentives to invest in training is the nature of the skills themselves. The more specific the skills are to the training firm, the less they will be valued on the external labor market, and the less likely the trained worker will be able to translate the new skills into a better job with another firm. At the extreme case, when skills are almost entirely firm specific, all of the gains from the additional skillsare captured by the training firm, and there is no poaching externality at all.6 In addition to the nature of the skills, several other factors can generate labor market imperfections that may provide incentives for firms to invest in skills. Under conditions of low turnover, adverse selection effects may make employers reluctant to hire new workers. The very fact that they are seeking new employ- ment may be interpreted by other employers as an indication of their low quality.7 In fact, any information problems that make it difficult for workers to change employers will reduce the danger of poaching and increase the employer incen- tive to invest in skills because other employers will be unable accurately to gauge the skill levels of workers who seek new employment.8 When wage compression exists, for example, when increases in real wages lag behind increases in the mar- ginal product of workers and when this gap expands with increases in marginal product, firms not only capture some of the gains from skills, but also have an individual incentive to increase the level of skills. Under conditions of wage com- pression, then, "general skills [become] de facto specific skills."9 The German apprenticeship system provides the best-known example of how institutions and labor market conditions can interact to overcome the employer collective action problem produced by the poaching externality. Germany's "dual system" of apprentice training combines formal vocational education in state vocational schools with workplace training in enterprises. Costs are shared among the state governments; the apprentices, who accept low wages during the period of their training; and the firms, who provide the machinery, trainers, and wages to trainees.10 Why do German firms cooperate in a system that exposes them to the possibility that they will invest in training workers who will be hired by other firms? Part of the answer lies in the firm-level benefits of cooperating. A firm that takes on apprentices provides itself with the additional flexibility created by workers to whom it does not owe a long-term hiring commitment.11 Participat- ing firms also save on recruitment expenses and reduce the costs incurred when new workers have to learn to adapt to their new jobs.12 There are also considerable public relations benefits from serving as a training firm.13 Cooperation in the German system cannot be fully explained, however, with- out incorporating institutional factors. David Soskice argues that the role of insti- tutions in promoting investment in training is mostly indirect, operating through what he and Hall would later describe as "institutional complementarities."14 The German financial system allows firms to adopt a long-run perspective, which, in turn, "makes it sensible" for firms to pursue a product market strategy requiring a JOHN R. BOWMAN 571 skilled workforce. A high-quality education system reduces the cost of training apprentices. A centralized wage-setting system makes it difficult to compete for skilled workers on the external labor market. A labor market characterized by low turnover increases the adverse selection hazard faced by potential employers. And significant internal labor markets increase the incentive for both firms and trainees to participate in apprenticeships.15 Centralized wage-setting institutions are also more likely to produce a compressed wage structure, increasing the incentives of individual firms to invest in training.16 Soskice argues that in the German case, the direct impact of employers'associ- ations and chambers of industry on the achievement of a cooperative training out- come has been minimal. By this, he means that although they possess "significant informal sanctioning ability over companies," they have not used formal or infor- mal sanctions to promote an investment in training among their member firms.17 However, it doesn't follow that these institutions have no direct impact on cooper- ation. One thing that employers'organizations (in association with unions) do to promote cooperation is to facilitate agreement on the content of training. A stan- dard curriculum with certification at the end of the training period helps to guar- antee that the high start-up costs incurred by training firms will eventually be compensated for in the form of the productive labor of skilled trainees.18 Employ- ers'associations also help to persuade firms to cooperate by providing them with "analytic"information about the causal connection between cooperative behavior and a beneficial outcome.19 They also facilitate the sharing of "relational" infor- mation concerning the likely behavior of other firms. In their information, advis- ing, and monitoring roles, employers' associations have an advantage over state institutions in that they enjoy the trust of employers. "Although companies may be willing to share . . . information with one another or with a representative asso- ciation,they willnot willinglyshare itwith the state."20 Thus, Soskice writesthat it is of the greatest importance that in Germany the institutions concerned directly with advising and monitoring companies are primarily the chambers and the sectoral employer associations and secondarily the works councils (with whom companies generally have high-trust relations).21 The main problem with this approach to the political economy of training is that initial vocational training and further vocational training are conflated.22 This is especially evident in the literature that discusses the German case, which focuses almost exclusively on the apprenticeship system and contains almost no discussion of the system of further training, which, institutionally, bears almost no resemblance to the highly coordinated dual system of apprenticeship training. Surprisingly, the authors who cite the apprenticeship system as evidence for their claims that Germany embodies a coordinated form of employment training are not bothered by the fact that Germany's system of continuing training is "institu- tionally chaotic,"23 "more a 'jungle'than a 'system,'" and characterized by the low 572 POLITICS & SOCIETY quality of training offers, the lack of transparency in the training market, and uneven access to further training.24 The problems of providing training to secondary school students and present employees share some features in common, but they are also marked by signifi- cant differences. From the perspective of the firm, it is necessary to distinguish two continuing-training situtations. First, training so that an established worker can increase his productivity in the firm is less likely to be viewed through the lens of the poaching externality. Even if the new skills are transferable, the totality of worker skills are likely to have a large component of firm-specific skills. Thus, it is unlikely that workers will be able to capture the benefits of this kind of training in the external labor market. Retraining workers to take advantage of new technol- ogies is likely to be more closely linked to a firm's product market strategy than is the training of apprentices. We can expect, then, that faced with training needs connected to their basic competitive strategies, firms will view the additional training of selected employees not as a public good subject to poaching on the external labor market, but as a necessary investment in the firm's productivity. In short, the collective action problem that has absorbed analysts of initial training assumes a very different form in this case. Unlike the intitial training case, we can expect employers to invest in training for self-interested reasons. The poaching externality declines in relevance. The incentives faced by firms are likely to be different in the case of further training that is driven not by the needs of the individual firm but by the needs of workers to develop their human capital, to become the kind of "lifelong learners" who will be able to adapt not only to changing technologies within an individual firm but to the kind of structural change that transforms individual firms (and whole sectors), perhaps even rendering them obsolete. Here, in the institutional setting characteristic of most economies, in which the protections and guarantees thatare builtinto successful initialtraining programs are not present, the poaching externality emerges with a vengeance. Employers have very little incentive to invest in further training whose primary purpose is to increase the transferability of worker skills. This problem is compounded by the fact that employers may view control over which employees receive training and the kind of training that they receive as a managerial prerogative that they are reluctant to sacrifice. In this case, the collective action problem derived from the poaching externality is probably more difficult to overcome than in the case of initial training. There are also differences on the worker side between initial and continuing training. Although the emphasis in the literature has been on employer contribu- tions to skill formation, apprentices also contribute, by accepting reduced wages during the period of their training. Young trainees have a much greater incentive to pay some of the costs of their training because the alternative--greatly reduced labor market options--is so unattractive. In this respect, secondary school stu- dents constitute a captive audience. This is not the case with experienced workers, JOHN R. BOWMAN 573 who are likely to have a lower incentive to participate in further training, except when it is necessary to carry out their present job. This incentive will be further reduced if wages are compressed; higher skill levels are less likely to produce significant wage increases. Finally, the institutionaland politicalcontexts of initialtraining and continuing training are likely to differ. Initial training, particularly in a cooperative market economy, functions in tandem with the formal education system. In effect, initial training policy is a subset of education policy. It is a national policy domain in which the legislature and educational authorities can expect to be deeply involved. Trainees are student workers whose relationship with their employers is embed- ded in a very different institutional framework than the relationship between adult workers and employers. Policies concerning continuing training have little con- nection with national educational policy making. Rather, they concern the kind of product market and labor market policies that will fall into the domain of collec- tive bargaining or firm-level employment relations. Disputes are more likely to take the form of labor conflicts than educational policy disputes. Moreover, the stakes of conflicts are likely to differ. More so than apprentice training issues, dis- putes over the skill development of experienced workers are more likely to take the form of conflicts over the way in which production is carried out and over the nature of product market strategies. This contrast between the politics of initial training and the politics of continu- ing training is further explored in the following case study of Norway. I begin by analyzing the institutional mechanisms responsible for increased cooperation among Norwegian employers in the country's apprenticeship system. I then turn to the issue of continuing training. The Norwegian case confirms the willingness of employers to invest in further training when it meets the needs of their firms. However, it also demonstrates the difficulties of instituting a continuing training regime that defines training goals more broadly as employee rights. Firms that cooperate in the apprenticeship program and that willingly invest in continuing training when it meets their needs were strongly resistant to invest in this more broadly conceived "lifelong learning." 3. INITIAL TRAINING IN NORWAY: THE ACHIEVEMENT OF EMPLOYER COLLECTIVE ACTION 3.1. Workplace Training before 1994 Prior to 1994's Reform 94, which reinvigorated Norway's apprenticeship regime, upper secondary education in Norway took place primarily in the public schools. State regulation of apprenticeships dated back to the early nineteenth century, but by the latter part of the twentieth century, apprenticeship training was not a significant part of the educational system. In short, Norway's coordinated market economy, with its highly centralized wage-bargaining system and devel- 574 POLITICS & SOCIETY oped welfare state, entered the 1990s with a distinctly uncoordinated system of skills provision. From the perspective of the problem of the poaching externality that was dis- cussed above, perhaps the most notable element of the Norwegian training regime was the persistent complaints from employers that workplace training repre- sented a significant burden for which the state provided insufficient compensa- tion.25 A 1978 commission that was convened to examine the apprenticeship sys- tem viewed this problem as being so serious that it recommended that if the support provided to training firms overshot budgetary bounds, then the state should find the necessary money by taking it from the schools and thereby sup- pressing the growth of school-based vocational training.26 A 1980 law on skills training in the workplace (Lov om fagopplæring i arbeidslivet) contained several features that were designed to expand the scope and quality of the apprenticeship system, including state subsidies to employers of 15 to 25 percent of the wages paid to apprentices.27 The pot was further sweet- ened for employers when, as a result of a 1985 amendment, they were released of the obligation to pay apprentices while they were in school.28 The act contained two other initiatives aimed at increasing the attractiveness of apprenticeships for small or specialized firms. It offered firms the possibility of forming "training cir- cles" or "training offices." A training circle (opplæringsring) is a group of firms that cooperate to provide training for an apprentice. In this arrangement, a single firm whose small size or specialization prevents it from offering a full curriculum of training experiences enters into agreements with other firms to share the burden of training. A training office (opplæringskontor) serves a similar coordinating role. It serves as an intermediary between the apprentice and training firms. When a single firm or a group of firms uses a training office to coordinate their training, it is the training office that signs the contract with the apprentice.29 It then assumes the task of administering the contract and arranges with other firms to provide the full range of training required by the national authorities. Few firms took advan- tage of the possibility of forming such collective organs during the 1980s, but, as we shall see below, training offices received a new lease on life in the mid-1990s and became a major component of the Norwegian training regime. In spite of these and other reforms, Norway approached the beginning of the 1990s amidst growing dissatisfaction with its occupational education system. On one hand, there was a sense that the competitiveness of the national economy would be more dependent than ever on Norway becoming a "knowledge-based community based . . . on the effective organization and use of technology." On the other hand, there was a perception that the country had not succeeded in fully exploiting the population's talents.30 Among the causes for concern was a contin- ued shortage of apprenticeship places.31 In 1990, only 4 percent of the student cohort who entered high school in 1988 had apprenticeship contracts. Twenty- three percent had dropped out; 26 percent were drifting horizontally through the JOHN R. BOWMAN 575 system as a result of changing courses of study midstream. Fewer than half of the students (47 percent) were in an upper-level course of study.32 One study con- cluded that "most young people who have not gone on to higher education will receive general or vocational education in upper secondary school and will not be provided with training under a formal contract with an employer."33 3.2. Reform 94 Reform 94 was a comprehension restructuring of secondary education in Nor- way. Its main provisions were the provision of a statutory right to three years of secondary educat