Labor Law Reform and Organized Labor: A Comparative Historical Sociology of Unanticipated Outcomes

Reflexive Statement

As both an observer and supporter of organized labor movements in the United States and elsewhere, I have long recognized the pivotal role that state-initiated labor law projects play in either hindering or facilitating the agency of working people as they collectively pursue their economic and political interests. Raised in a strong union family and attending graduate school in the late 1980s, it was perhaps not very surprising that my dissertation research focused on how the Reagan and Thatcher administrations each tried to use legal means to undermine organized labor in their respective countries. While both neoliberal governments certainly had deleterious effects on labor unions, Reagan through the administration of existing labor law via NLRB appointments and Thatcher through a step-by-step enactment of a new legislative framework, their attempts at the instrumental manipulation of labor law were certainly problematic at times. That research led me to examine labor law reforms in other capitalist democracies, with a particular eye toward what types of changes can help to empower organized labor, or at least uncover what strategies unions might adopt when confronted with less than favorable legal environments. This line of inquiry has led to the current article, the findings of which I hope can offer some valuable lessons to labor activists as they pursue the reform of U.S. labor law in the twenty-first century.

Introduction

Union density in the United States stood at 10.7 percent in 2016, down from 11.1 percent in 2015. Though the current rate represents a relatively modest decline from a decade earlier, with union density standing at 12.5 percent in 2004, that decrease is part of a more persistent long-term trend. Specifically, union density in this country has been close to halved since 1983, when the rate stood at 20.1 percent (U. S. Bureau of Labor Statistics 2017). With density levels estimated to be about 35 percent in the mid-1950s, a period which represented the apex of unionization in the United States, future prospects for the organized labor movement certainly look unwelcoming if this pattern of diminution continues much longer.

Factors that have contributed to union contraction during the past half of a century are many and have been well-documented. These include industrial and technological changes, free trade agreements, employer resistance in many forms, and the institutional practices of unions themselves (e.g., Chaison 2006; Goldfield 1987). Another factor contributing to union decline frequently pointed to is the federal labor law system. Though the original legal framework established by the National Labor Relations Act (NLRA) in 1935 served to legitimize and facilitate unionization, later changes to the law, most notably the Taft-Hartley Amendments of 1947, made it more difficult for organized labor to increase membership and to enforce existing collective bargaining agreements. Changes in the composition and staffing of the National Labor Relations Board (NLRB) have further hampered union efforts by failing to conduct timely representation elections and by expanding the legal domain of managerial prerogatives (e.g., J. A. Gross 1995). Overall, of all the variables that affect the fortunes of the organized labor movement, labor law policy and its attendant institutions are clearly of pivotal significance since they provide a large part of the structural foundations and parameters for the agency of individual workers, unions, and employers.

The growing recognition that existing labor law practices have become detrimental to union interests has led to several piecemeal attempts by organized labor supporters to have provisions of the NLRA repealed or amended (Burton 2015). Some have even called for the wholesale scrapping of the entire NLRA and for the development of an entirely new labor law framework (e.g., Brody 2004). Nevertheless, such political projects, whatever their intent and content, are often fraught with difficulty. As with any social policy initiative, indeterminacy and contradiction abound in both the formulation and implementation of labor law. In fact, such efforts can sometimes go completely awry and produce outcomes fundamentally contrary to what was intended.

With the above in mind, my goal in this article is to utilize a comparative historical sociological approach to highlight possible contingencies that might present themselves in any future attempts at labor law reform in the Unites States. Central to this effort is a comparison of two case studies wherein labor law policies targeting organized labor resulted in outcomes antithetical to what was intended by state actors. The cases in question are Germany’s conservative labor law project in the 1950s and France’s Socialist labor law project in the 1980s. Drawing lessons from these two instances of unanticipated outcomes, I conclude by broadly assessing possibilities for labor law reform in the United States, particularly with an eye toward stemming the decline of the organized labor movement while also trying to anticipate any unanticipated consequences that may result.

Sociology and the Unanticipated

Conceptual Issues

Though “anticipating the unanticipated” has an oxymoronic ring to it, many sociologists and social scientists have either implicitly or explicitly pursued such an agenda as they have tried to grapple with the unforeseen in social life. Indeed, over three-quarters of a century ago, Robert Merton (1936:894) forcefully noted that, in one form or another, “the problem of the unanticipated consequences of purposive action has been treated by virtually every substantial contributor to the long history of social thought.” Early social thinkers pointed broadly toward this phenomenon, using singular terms such as “fate” (Machiavelli) or “providence” (Vico) or more colorful phrases such as “the cunning of Reason” (Hegel) or the “invisible hand” (Smith). Major theorists of sociology’s classical tradition often dealt with the issue more specifically. Marx, for example, focused on the unforeseen results of class conflict in particular historical contexts. For his part, Weber demonstrated how the religiously motivated conduct of spiritually anxious Calvinists produced unexpected economic transformations throughout Western Europe. And Simmel uncovered the unexpected in small group dynamics as well as in the broader cultural tragedy bequeathed by modernity (M. Gross 2003; Merton 1936; Portes 2000).

Of course, it was Merton (1936, 1957, 1998) himself who systematically laid the conceptual groundwork for developing the sociology of the unanticipated. In addition to providing an overview of previous thinking along these lines, his 1936 article “The Unanticipated Consequences of Purposive Social Action” made important contributions on two other fronts. First, Merton specifically narrowed his analysis to the unforeseen effects of what he called “conduct” or “action which involves motives and consequently a choice between various alternatives,” thus excluding consideration of the unexpected consequences of less intentional behaviors (Merton 1936:895). He further noted that such “conduct” can result from either the “unorganized” actions of aggregated individuals or from more “formally organized” goal-directed associations of individuals (Merton 1936:897). It is this latter type of purposive action that will be the focus of this article, specifically state-initiated labor law reforms directed toward the role of labor unions within capitalist democracies.

Second, Merton also provided an initial inventory of the various causal factors commonly implicated in the generation of unanticipated consequences. Inadequate knowledge about all relevant variables impinging on the successful realization of purposive actions was one. A second cause of unforeseen effects is simple human error, whether it is in the conception, development, or execution of a conscious plan of action. What Merton called the “imperious immediacy of interest,” wherein longer term consequences are not considered given the pressing nature of current conduct, constitutes a third factor. Similarly, conduct motivated for the affirmation of certain values, such as the economic activity of Weber’s Calvinists, may also produce unanticipated long-range results. Finally, Merton also noted how the reflexivity inherent in the human condition also created unexpected outcomes, since once purposive action is announced or pursued in a particular social setting it serves to change the conditions of that environment (Merton 1936:898-904).

As is well known, Merton’s (1957:51) later work on this topic took on a more structural functionalist hue, with the concepts “latent functions,” “latent dysfunctions,” and “non-functions” representing a further specification of the more general term “unanticipated consequences.” In this particular light, the present article could in some respect be depicted as a comparative historical analysis of the latent dysfunctions of labor law policy reforms. Specifically, the two cases examined here reveal unintended effects that were directly contrary to the goals of state actors, and hence could be deemed “dysfunctional” from their respective standpoints.

Methodological Issues

Several considerations drove my selection of the German and French instances of labor law reform for comparison. First, following Merton, both historical cases represent definitive examples of how “purposive social action” can produce striking “unanticipated consequences.” As I will describe, conservative-led legal changes in Germany in the mid-twentieth century were enacted with the objective of undermining labor unions in that country, but had the ironic effect of strengthening organized labor over the long-term. Similarly, Socialist-initiated labor law changes in France during the 1980s, implemented with the goal of empowering organized labor, ultimately had the opposite impact. Overall, it is marked unexpected outcomes such as these that Alejandro Portes (2000:10) feels come “closest to Merton’s (1936) original treatment of unintended effects” and are “arguably the most important” for sociological analysis.

Another consideration is more theoretical in nature and stems from the contrasting substantive intentions of state actors and their use of labor law in the two countries. Though both policy projects fundamentally failed in achieving their desired results to either strengthen or weaken organized labor, I feel that a comparison of these two cases can be theoretically fruitful in at least two ways. At a specific level, I hope to begin to uncover and unpack common variables between the two episodes that impinge upon and hamper the successful realization of policy initiatives, whether in terms of labor law or other social arenas. In this regard, I am utilizing a method of “structured, focused comparison” for inductive theory building, in the hope of developing a more analytical explanation for policy failures (George and Bennett 2005:x-xi).

At a more general theoretical level, the selection of the German and French cases serves to help exhaust logical possibilities that exist when it comes to the purported intentions of labor law reforms vis-à-vis organized labor movements and whether or not such projects have their expected or unexpected consequences. ¹ For example, as noted, the French case serves to illustrate a situation wherein efforts to strengthen labor unions by state actors had an opposite and unexpected result. In direct contrast, a well-known and directly pertinent instance where government efforts to strengthen organized labor had its anticipated results, at least initially, is the implementation of the NLRA in the United States in 1935. The German case, as described below, represents another logical possibility: Labor law reform enacted to undermine organized labor had the opposite and unintended effect of empowering unions. While the Taft-Hartley Amendments of 1947 can certainly serve as a counterpoint to the German instance, whereby state actors succeeded in achieving their objectives of weakening unions, the labor law project of the Thatcher government in Great Britain during the 1980s is perhaps a more fruitful example. As I discuss in the concluding section, analysis of the British case also points to the importance of the key variables affecting the labor law policy outcomes uncovered in the German and French cases.

On a more practical level, I am hoping that this comparative historical analysis of unanticipated outcomes can provide lessons on how to best implement labor law policy reforms in the United States and elsewhere. Given the current political economic context both nationally and globally, labor unions specifically and working people more generally are in desperate need for, as David Brody (2004:15) has phrased it, “A Labor Law for the 21st Century.”

Germany’s Conservative Labor Law Project

Postwar Germany’s dual system of labor representation, wherein labor unions operate at the industry level and works councils and supervisory board representation function at the firm level, is well known in the comparative sociology and industrial relations literature. Legally structured and protected, the dual system has resulted in a fairly resilient “social partnership” between workers and management (Turner 1998:17-18), at least up until Germany’s reunification in the 1990s. In particular, centralized bargaining practices between industrial unions and employer associations, coupled with negotiations and consultations inside particular workplaces, produced relative labor stability (e.g., Silver 2003:43-44) and prosperity for capital (e.g., Streeck 2009:77-89) for several decades. Overall, this institutional model is frequently credited for producing a more humane form of capitalism.

Perhaps less well known, however, is the fact that this arrangement pre-dates World War II. Indeed, it was just after World War I that legal protections were first put in place regarding labor’s rights of association and collective bargaining. And it was the Works Councils Law of 1920 that originally established the second institutional prong of the dual system (Markovits 1986:33). For several reasons, however, this early incarnation was not very effective in addressing worker concerns. Though the works councils were explicitly subordinated to unions by this early legislation, the labor movement was sharply divided between radical revolutionaries and moderate reformists. This lack of a united front thwarted a greater realization of worker interests under this version of the dual system. Furthermore, as the 1920s progressed, many German employers, who had initially agreed to the creation of works councils to stave off labor’s more revolutionary demands, were able to return to more active resistance to their actual implementation (Thelen 1991:66-72).

This early rendering of the dual system was fully dismantled during the era of National Socialism. After the defeat of the Nazis and the Allied occupation, Konrad Adenauer’s government began the task of reconstructing Germany’s political economy. Looking toward the earlier Works Councils Law as well as Allied provisions as broad models, the conservative government implemented the Works Constitution Act in 1952. In contrast to the post–World War I legislation, however, there were significant differences in the new law. For example, the 1952 act had “narrowed the range of rights works councils had previously enjoyed…by specifying their responsibilities, rights, and obligations more precisely” (Thelen 1991:74). The most important feature of the new legislation, and arguably the one that generated the greatest union opposition, involved the specific relationship between labor unions and works councils. As described by Thelen (1991:75-76):

While the 1920 law emphasized the subordination of the works councils to unions …the 1952 law stressed the autonomy of the works councils and especially their countervailing loyalty to the welfare of the company.

In effect, the intent of the Adenauer administration was to sever works councils from the organized labor movement and make them more like localized employee associations beholden to the firm. Several provisions of the new legislation pointed toward this objective. Article 49 (Paragraph One) states that “[t]he employer and the works council shall cooperate in complete confidence…for the well-being of the establishment and its employees with due consideration for the general welfare” (Windmuller 1953:409). Paragraph Two of the same Article goes on to specifically prohibit “industrial conflicts” between employers and works councils (Windmuller 1953:409). More generally, while labor unions were by no means excluded from the shop floor, their activities were limited and subject to approval by the works councils (Thelen 1991:76).

The conservative government’s intentions in fostering works council autonomy vis-à-vis the labor unions were further apparent in the provisions pertaining to election procedures. Article 10 of the 1952 law states that “[w]age earners and salaried employees must be represented in accordance with their numerical strength on the works council”, though a minimum floor is set in place for the “minority group” that allows them greater representation (Windmuller 1953:406). In most cases, it would be salaried employees who were in the minority and who would also be less likely to be union members. In addition, Article 13 provides for separate elections for wage and salaried employees, unless “both groups decide before the new election by separate and secret ballot upon a joint election” (Windmuller 1953:406). Overall, as Thelen (1991:76) indicates, in light of these provisions put in place by the conservative government “[t]heir hope was presumably to promote union-neutral or even anti-union candidates in works council elections.”

The implementation of the Works Constitution Act of 1952 was clearly a defeat for Germany’s labor unions. While retaining their relative position outside of the workplace, the presence of unions in the workplace was in danger of being seriously curtailed. Organized labor’s largest and leading union, IG Metall, “condemned the law as an ‘open challenge for battle’” (Thelen 1991:77) and hoped to have the legislation repealed or substantially revised. One broad effort was thus aimed toward getting the more labor-friendly Social Democratic Party elected, but given the postwar climate, “it was clear that the political route to a new codetermination law would be a long and arduous one for the unions” (Thelen 1991:77). With Adenauer and the Christian Democratic Union retaining power for the rest of the decade and beyond, it was evident that organized labor was going to have to rely on methods that worked within the structures of the new law.

IG Metall aggressively pursued a two-pronged strategy to “ensure a union presence in the plant as a check on the autonomous (and potentially rogue) works councils” (Thelen 1991:78). One line of response was the aggressive expansion of shop steward committees that would operate parallel to the works councils within establishments. This effort proved quite successful, with not only the numbers of shop stewards increasing markedly throughout the 1960s and early 1970s but also union membership levels increasing. A second strategy, and arguably the most effective in furthering worker interests within the framework of the law, entailed active efforts to get union members elected to works councils. IG Metall, as well many other unions that followed their lead, was even more successful with this tactic. By the late 1950s, and in the ensuing decades, the union occupied 80 percent or more of the works council seats in the industries it covered (Thelen 1991:80). Parallel proportions of work councilors who were also union members were soon found throughout the German economy.

In retrospect, this particular outcome was clearly unanticipated by the conservative government when they enacted the Works Constitution Act of 1952. Indeed, not only were the results unexpected, they were directly contrary to the intentions of the legislation’s authors. Instead of labor unions being directly marginalized on the shop floor, and by extension indirectly weakened throughout Germany’s industrial relations system, the conservative’s plan “backfired” and organized labor’s position was ultimately strengthened both inside and outside the workplace (Kerbo and Strasser 2000:57). As described by Thelen (1991:17), “[u]nion dominance in the works councils and the subsequent evolution of the relationship between them and the unions [had] turned the ‘logic’ of the dual system on its head.”

Later history offers further evidence for how Germany’s conservative labor law project was eventually turned “on its head.” The ascension of the Social Democratic Party to power in the early 1970s finally gave organized labor the political opportunity to have the Works Constitution Act repealed or substantially overhauled. The 1952 act was indeed replaced by the Works Constitution Act of 1972. However, “replaced” is clearly too strong a characterization, for as Andrei Markovits (1986:48) states, “[t]he 1952 and 1972 acts are on most points identical.” The differences between the laws are twofold. First, the 1972 legislation gave works councils expanded influence over more shop floor issues. In the area of personnel policy, for example, works councils would now have “full codetermination rights on issues such as working time arrangements in the plant, short-time work, overtime, work breaks, the establishment of vacation times, plant wage systems, and the setting of piece rates” (Thelen 1991:100-1). Second, the 1972 law strengthened the role of labor unions within the workplace. In addition to union officials being granted greater access to establishments, “the revised law permits works councilors to engage in union activities within the plant, thus de facto legalizing open union recruitment on the shop floor” (Markovits 1986:49).

In many respects, it could be argued that German labor law had come full circle. The 1972 legislation, with just slight modifications from the 1952 act, looked a little more like the Works Councils Law of 1920 that labor unions strongly supported. Whereas the labor movement was too divided to realize the full potential of the initial legislation in the early twentieth century, however, much of the latter part of the century was a different story. And a law that unions once saw as an “open challenge for battle,” is now lauded. Lowell Turner (1998:37), for example, describes how during the period of reunification in the 1990s, union leaders would “carry around the Works Constitution Act” as they organized the workers of eastern Germany and set up the works councils that they once feared would undermine their power. Overall, much to the chagrin and directly contrary to the intentions of post–World War II era political conservatives, works councils had been effectively encapsulated by organized labor and were used to the advantage of working people.

France’s Socialist Labor Law Project

France’s socialist labor law project began in 1982, approximately one year after Francois Mitterrand was elected president, and the Socialist Party had gained control of the National Assembly. Though in many policy areas the ensuing “Socialist experiment” primarily took the form of “redistributive Keynsianism” rather than a revolutionary restructuring of the country’s political economy (Hall 1986:191-93), France’s labor law regime was the target of a more “radical agenda” (Howell 1996:147). Indeed, eventually one third of the French Labor Code, originally put in place in 1950, was rewritten with the explicit objective of giving workers greater power and representation (Howell 1992:181). Underlying this extensive reform was the ideology of autogestion, which claimed to be “a new kind of socialism” (Smith 1987b:46) entailing democratic self-management by workers and employers within workplaces and by labor unions and employers at the industry-wide level. Overall, the dual system of German industrial relations clearly served as an inspiration and a broad model for the French reforms.

The Mitterrand labor law project entailed the enactment of five different pieces of legislation, collectively called the “Auroux Laws” after the Labor Secretary who authored them, between 1982 and 1983. The initial legislation was titled Liberties of Workers within Enterprises (“Auroux Law I”) and was adopted on August 4, 1982. While much of it simply expanded earlier provisions already in the 1950 Labor Code to smaller firms and to public employers (Glendon 1984:452), there was a significant portion reflecting autogestionnaire ideology. Specifically, every French firm was to create what were essentially employee participation groups that allowed workers to express concerns and grievances regarding working conditions to management representatives without fear of reprisal. Importantly, however, the particular form that this “direct and collective right of expression” (Smith 1987b:49) was to take was never specified in the legislation. Rather, the particular characteristics of the workplace “expression groups,” such as their “frequency, size, duration and organization” (Smith 1987b:49), was to be negotiated by the interested parties (management included).

Auroux Law II was adopted on October 28, 1982. Titled Development of Employee Representative Institutions, changes embodied in this piece of legislation addressed the place of trade unions within the workplace as well as two other already established institutions representing worker interests in French industrial relations: employee representatives and shop committees. Despite broad similarities in the goals of these three entities, they each played different roles within the workplace. Employee representatives were elected by workers to deal with firm-level grievances. Shop committees, comprised of employees as well as union representatives, were involved in a wider array of plant issues but only played a consultative role. The changes enacted by Law II pertaining to these mechanisms were “only minor” and “relatively modest” (Glendon 1984:461-62) and for the most part simply extended these institutions to smaller firms (Smith 1987b:49). This second law made much more significant changes with respect to labor unions, however. The right to form a union section was now extended to all firms where earlier it had been restricted to those with 50 or more employees. In addition, the workplace rights of unions were expanded in several areas, such as giving union officials the ability to collect dues during working hours and allowing increased protections against dismissal for their activities (Glendon 1984:466-69; Smith 1987b:49). In light of these particular provisions, one observer of French industrial relations was compelled to comment that the law “was designed and seems likely to appreciably strengthen the presence of unions within companies and to promote unionization generally” (Glendon 1984:466).

Auroux Law III appeared to confirm the intentions evident in the previous piece of legislation. Titled Collective Bargaining and the Regulation of Labor Conflict and enacted on November 13, 1982, periodic negotiations between unions and employers were now required at both the firm and industry levels. While there was no legal obligation to reach a collective agreement, the parties faced criminal fines and imprisonment if they did not meet in good faith. Two objectives underlay this particular law. One quite obviously was the expansion of collective bargaining. Related to this was the hope such periodic meetings would “encourage a movement away from confrontation and negotiation under crisis conditions toward a continuing dialogue between what it calls ‘the social partners’” (Glendon 1984:472). This aim was designed to mitigate the more dramatic and widespread industrial conflicts that have long characterized French industrial relations (Jeffreys 1996:509).

The next piece of legislation passed by the Socialist government was titled Committees on Health, Safety, and Working Conditions (Auroux Law IV). Enacted on December 23, 1982, it established health and safety committees in firms with more than 50 employees. Although not directly related to organized labor, Auroux Law IV did serve to protect worker interests on the shop floor by giving these committees the power to monitor, inspect, and take action to improve health and safety conditions. The final piece of legislation, titled Democratization of the Public Sector (Auroux Law V), was passed on July 26, 1983. This act extended provisions of some of the earlier legislation to public sector employees. In particular, the expression groups that were established in Auroux Law I were now mandated for public-sector enterprises. Auroux Law V even went a little further by setting minimum requirements for when the expression groups should meet: “at least once every two months and for at least six hours a year” (Glendon 1984:482). This last piece of legislation also extended employee representation on the governing boards of public enterprises and required negotiations with public-sector unions (Glendon 1984:481-82).

In retrospect, there were three key objectives evident in the Mitterrand administration’s labor law project that was designed to empower workers and labor unions (Smith 1987b:53). These were the creation of workplace expression groups (Auroux Law I), the strengthening of existing mechanisms for employee representation (Auroux Law II), and the facilitation of more routinized collective bargaining (Auroux Law III). The Socialist government can be deemed to have been successful in partially realizing its intentions with respect to only one objective. With respect to the other two goals, there were significant unanticipated consequences.

As intended, collective bargaining negotiations did increase throughout the French industrial relations system as a consequence of Auroux Law III (Ross 1987:212; Smith 1987b:56). Yet, this goal was arguably not realized to the extent that the Socialist government had hoped for. By mid-decade, “mandated negotiations were held in only 60 per cent of eligible firms,” and “there was much resistance to the new law, especially within smaller firms” (Smith 1987b:56). And despite the required negotiations, relatively few collective bargaining agreements were ever established, with only about 10 percent of the labor force covered by such contracts (Smith 1987b:56). Interestingly, and perhaps unexpectedly, of the relatively few agreements established, most were at the firm rather than the industry level (Moss 1988:325-27). Some observers thus speculated that if the trend continued, the French organized labor movement would be slowly transformed into a less effective type of “enterprise unionism” (Howell 1996:153).

Auroux Law II had even more unanticipated consequences. The provisions of this piece of legislation had two effects that would serve to undermine French labor unions. First, the strengthening of employee representatives and shop committees, coupled with the new expression groups, made union membership appear less necessary for many workers (Smith 1987b:55). Second, as one observer has noted, “the reinforcement of representative institutions in the workplace…enmeshed already overtaxed union militants in deeper levels of bureaucracy” (Ross 1987:212). Mobilizing for collective action, a long-standing and pivotal strategy for French unions, thus became more difficult (Smith 1987b:55).

All of the above aside, it is perhaps the unintended consequences of Auroux Law I that had the most significant impact on organized labor. Though by the mid-1980s worker expression programs existed “in one half of the applicable firms” (Moss 1988:325), the primary problem stemmed from who had established them and how they operated. Unions and workers “were timid and skeptical” about the groups, and thus took little initiative in establishing them (Ross 1987:212). It was the managers that took the lead in creating the expression groups, and they were the ones who “trained staff in leadership techniques, scheduled the meetings and explained their purpose” (Moss 1988:324).

More important than their establishment was how the expression groups were made to function. As noted by W. Rand Smith (1987b:53), “employers…largely ‘contained’ expression groups and…harnessed them to a broader management policy of building employee identification with the firm.” Specifically, the expression groups were used to “undercut the unions’ traditional role as mediating agent between employees and management by ‘individualising’ the worker-management relationship” (Smith 1987b:53). Not surprisingly, in light of this experience, employers’ initially hostile attitudes toward the “right of expression” provisions of the new legislation changed to increasing approval for expression groups as the decade of the 1980s unfolded (Moss 1988:326).

Overall, while a variety of other factors certainly played a role, the labor law project of Mitterrand’s government contributed to the weakening of French unions in the 1980s. This outcome was directly contrary to the intentions of the Mitterrand government and the Socialist Party’s broader objectives. Nonetheless, this unanticipated result was the product of the laws themselves and how they were responded to by employers and unions (Smith 1987a:3-6). In this regard, the labor law project of Mitterrand and the Socialist party, despite its very different content, has many similarities with the experience of German conservatives described above. In the next section, I will analyze some of the commonalities between the two cases in the hope of shedding insight into how the unanticipated consequences of labor law reform might be anticipated.

Commonalities in the German and French Cases

A comparison of the German and French labor law projects points toward three sources of contingency that frequently contribute to the unanticipated outcomes of planned political interventions. One source leading to contrary outcomes is the internal character of the legislation being implemented. Any particular legislative initiative may have conflicting, if not contradictory, potentials embedded within the same law. As Christopher Howell (1992:182) put it, “the paradox is that there is a certain plasticity of law which goes beyond simple unintended consequences and the failure to achieve a set purpose. Laws rarely carry within them only a single logic.”

Howell’s (1992, 1996) systematic analyses of the Mitterrand administration’s legislation cogently illustrates this point. Specifically, he maintains that the Auroux Laws “contained two distinct and coherent, but incompatible, logics” (Howell 1996:148). One of these logics had the potential of generating more widespread collective bargaining in France with labor unions having a “critical role [as] privileged representatives of the working class” (Howell 1996:148). The other logic, which as history has shown was realized, would have the expression groups and other employee representative institutions emerging as “alternatives to union organization” and producing a “microcorporatist” enterprise unionism (Howell 1996:149; emphasis in original).

Similarly, the conservative government’s 1952 Works Constitution Act, designed to make works councils more autonomous institutions within German industrial relations, also embodied different potentials. In this case, however, it was only one piece of legislation rather than several that led to unforeseen results. The particular logic underlying this political endeavor was to formally separate the works councils from the influence of the organized labor movement, perhaps turning them into institutions more amenable to employer influence. The legally structured independence of the works councils, however, did not necessarily guarantee such an outcome. Indeed, as I recounted above, German unions strategically worked within the law to take control of the autonomous works councils, a potential that was not anticipated by German politician or employers alike.

This particular facet of the German experience points to two other factors relevant to the generation of unintended consequences for state labor law projects. These latter factors tend to be more external, rather than internal, to the law itself. Specifically, both the political and economic structural conditions surrounding the legislation’s implementation need to be accounted for, as does the strategic agency of pertinent parties. As George Ross (1987:212) described, in particular regard to the France’s Socialist labor law project but certainly applicable to any case:

Legislation can only partially manipulate industrial relations practices, of course. The final outcomes of reforms like the Auroux Laws depend on the relative strengths and intentions of labor and capital over the long run.

Overall, while there is often a strong interdependence between the two dimensions of structural context and strategic agency, there is a complexity that goes beyond simple correspondence and determination.

The German case demonstrates this complexity. Recall that the initial reaction of German labor unions was to work for the repeal of the 1952 Works Constitution Act. This was not politically viable in the postwar climate, so alternative strategies within the framework of the law had to be developed. In addition to the promotion of shop floor committees led by IG Metall, Germany’s organized labor movement set out to capture the legally autonomous works councils by having union members elected to them. Their efforts were remarkably successful, despite occurring in a structural context where workers were disadvantaged vis-à-vis the government and employers. As quoted earlier, Thelen (1991:17) described this feat by stating that “the unions [had] turned the ‘logic’ of the dual system on its head.”

If German unions can be characterized as having taken over the works councils, it may be said that French employers captured the Socialist-initiated expression groups. This accomplishment can in part be attributed to both structural context and strategic agency, for “[t]he Auroux Laws were attempted at a time of unprecedented employer strength and militancy” (Howell 1992:192). The reverse side of this was a lack of strategic response by French unions. As I noted above, French workers were hesitant and suspicious of the new expression groups, so it was employers who initiated them and harnessed them toward company goals. One can only speculate as to what the outcome of the Socialist labor law project would have been had French unions taken a more proactive role and pursued a strategy similar to that of German unions a few decades earlier. Perhaps the one “logic” inherent in the Auroux Laws that empowered unions would have come to fruition, and not the “logic” that served to partially displace unions.

Reforming American Labor Law

Ever since the Republican-led Congress amended the NLRA in 1947, members and supporters of organized labor have sought to reform the legislative framework that structures the conduct of industrial relations within the United States. After the failed attempt in 1949 to repeal the Taft-Hartley amendments in toto, four later bids to reform the NLRA were to be more measured and piecemeal in terms of what legal provisions they proposed. For example, the repeal of Section 14(b), which allowed states to implement “right to work laws,” was the focus of a major reform effort under the Johnson administration. A second attempt at labor law reform during the Carter presidency aimed primarily at protecting union organizing by increasing penalties against employers for unfair labor practices. During the early years of the Clinton administration, industrial relations reform focused on banning the use of permanent replacements during strikes. Similarly, during the early years of the Obama administration, reform focused on allowing for “card check” certification processes that were thought to facilitate union organizing. As is fairly well known, all four of these reform endeavors met with failure, even though labor-friendly Democrats controlled both the executive branch and at least one chamber of the legislative branch during all these efforts (Burton 2015:13-15; Kahlenberg and Marvit 2012:104-5).

While there is no prediction when or if political opportunities will arise again for the incremental, or for that matter even the wholesale, reform of the NLRA, there are other options that can be pursued for protecting or enhancing the rights of working people within existing legal frameworks. One possible line of action, for example, entails what some comparative historical social scientists have called institutional conversion. Such transformation results when “actors are able to redirect institutions or policies toward purposes beyond their original intent” (Hacker, Pierson, and Thelen 2015:180). The “capturing” (Haipeter 2013:117) of works councils by German unions in the 1950s and 1960s, as discussed above, serves as one example of how such “conversion” can occur. Recall that the original intent of the Works Constitution Act of 1952 was to circumvent union power by establishing works councils as a firm-level alternative to industry-level worker organization and interest mediation. Strategic agency by Germany’s organized labor movement, wherein it was able to seat large numbers of its members on works councils, resulted in the transformation of the councils into a de facto vehicle for union interests. This institutional conversion was, of course, historically contingent, as evidenced by the recent realignment of works councils with firm-based employer interests and the simultaneous diminution of union presence at that level (e.g., Jackson 2005).

Prospects for “institutional conversion” within the framework of U.S. labor law have recently been documented with respect to the rights of nonunionized workers. In a case study of primarily immigrant female workers in New Mexico, Garrick (2014) describes how female hotel housekeepers, working in tandem with the staff of a local worker center, used Section 7 and Section 8 of the NLRA to obtain redress for grievances. In response to an intensification of the labor process, the housekeepers stopped working and demanded a meeting with management, which responded with an ultimatum to get back to work or be fired. Refusing to go back to work the housekeepers were fired, but upon reviewing the case lawyers at the worker center came to the realization that the “housekeepers had—without knowing it—mobilized part of their Section 7 rights, which states that workers “can engage in other concerted activities for the purpose of collective bargaining or mutual aid or protection” (Garrick 2014:497, emphasis in original). Their firing thus represented an unfair labor practice by the employer under Section 8, and it was subsequently decided in court that the housekeepers were legally eligible for reinstatement and back pay. After reviewing other cases that successfully protected nonunion organizing within the framework of existing labor law, Garrick (2014:508) concludes that “repurposing the National Labor Relations Act” may be an important strategic opportunity for empowering working people.

While Garrick’s research demonstrates that the NLRA contains potentialities that may have heretofore been largely unrecognized by many labor activists, Kahlenberg and Marvit (2012) make a similar point with respect to the Civil Rights Act of 1964. They make a compelling argument that reframes union membership and related activities as individual civil rights protected by this widely accepted and fundamentally effective legislative framework. While at first glance it might appear somewhat of a stretch to place one’s union identity on par with one’s gender, racial or ethnic identities, Kahlenberg and Marvit (2012:91) note “that the Civil Rights Act has long covered mutable traits that can involve individual choice, such as religious affiliation or pregnancy.” Other volitional acts are covered by the act as well, such as membership in the armed forces, bankruptcy status, and whistleblowing. Overall, if such a legal tack were pursued, some instances of unfair labor practices by employers under the NLRA could be recast as instances of employer discrimination under the Civil Rights Act wherein individual employees could file claims in civil courts.

Of course, Kahlenberg and Marvit are fully cognizant that this potential legal strategy for individual union members has its limits. Most notably, unfair labor practices by employers that primarily target unions as collectives would still fall under the provisions of the NLRA, such as using permanent replacements during strikes or bad faith in collective bargaining violations. Nevertheless, there is one key tactical advantage for the organized labor movement. As Kahlenberg and Marvit (2012:108) point out, “the civil rights approach would take a very substantial first step by removing what evidence suggests is the most powerful weapon an employer has to end a union organizing drive: firing a few employees who support the union and terrorizing the rest into submission.”

Overall, both the ideas of “repurposing” existing labor policy and the reframing of union membership as a “civil right” open up further strategic possibilities in the use of law to protect the rights of working people as they pursue their economic and political interests. The potential success of each tactic, however, is contingent not only on the resources afforded organized labor by structural economic conditions (Silver 2003) but also by the degree of interpretive realignment union leaders and members can muster in articulation with extant policy frameworks (Campbell 2012). With respect to the latter condition, perhaps the comparative historical analysis of the German and French cases discussed above can also help labor activists in the imagining of alternative tacks vis-à-vis the current legal context. Whether these or other strategic possibilities to revitalize the organized labor movement in the United States are undertaken, and to what extent, remains to be seen.

Conclusion

State labor law projects, whether emanating from the political right or left, can be risky endeavors. The German and French cases recounted above show how legal reforms initiated for one purpose can have unintended effects directly contrary to a government’s original goals. These types of circumstances occur in other policy domains as well. For example, drawing in part from his own research on immigration policies, Portes (2000:13) has acknowledged that a keen awareness of how far awry legal reforms can go from primary purposes “can lead to paralysis in both policy and theory.” This awareness also has implications for sociologists and other social scientists.

Because the dialectics of social life are so complex and everything depends on the specific context in which it is embedded, it becomes nearly impossible to predict how individuals and groups will behave or what outcomes will extend from deliberate policy. The role of sociologists as engineers of the future thus dissolves into the much less attractive role of professional doubters and critics (Portes 2000:13).

Despite such warnings, Portes does not advocate retreat from purposive social action or planned political interventions. Important lessons can often be gleaned from the sociological analysis of the unexpected. Portes points toward two in particular.

One lesson is that “change must proceed in measured steps, with close attention to fortuitous events and pressures from outside forces” (Portes 2000:15). Exemplifying such a “measured steps” approach in the area of labor law would be the reforms implemented by the Conservative Thatcher government in Great Britain during the 1980s. This labor law project also demonstrates the importance of how policy feedback (Campbell 2012) can contribute to the success, or failure, of political initiatives vis-à-vis organized labor. Because this case is instructive and perhaps provides insights for labor law reform in the United states, it bears brief review here.

Consisting of five pieces of legislation over a 10-year period, the Thatcher government took a “step-by-step” approach to labor law reform as a consequence of the Conservative Party’s earlier involvement with similar policy initiatives. Specifically, the Conservative Heath administration had implemented the Industrial Relations Act in 1971 that strictly regulated the ability of unions to engage in industrial action, undermined closed shop practices, and legally hampered unions in a variety of other ways. The wide-ranging and draconian nature of the legislation sparked militant resistance from the British organized labor movement, and the Conservatives were eventually voted out of power and the law was repealed. In light of this experience, prior to being elected to office, Thatcher and the Conservative Party had decided to undermine union strength with more piecemeal legislative reform (Nash 2000). Not only did this approach mitigate union opposition, it created space for policy feedback. For example, the fourth and fifth pieces of legislation, the Employment Acts of 1988 and 1990, addressed some of the unintended consequences that appeared to be emerging from earlier legal changes as the decade progressed. One particular area of concern, among others, stemmed from ballot requirements for industrial action put in place by a 1984 legislative act, which had been premised on the idea that union members were fundamentally less militant than their leaders and thus industrial action was less likely if all were able to vote on it. Somewhat unexpectedly, however, some union members became more committed to industrial action because they were able to vote on its undertaking. It appeared that a logic of union democratization was emerging that was contrary to the Thatcher administration’s overall objectives. Consequently, later legislation was used to implement amendments that were able to stem the extent of these unanticipated outcomes (Mackie 1988; Nash 2004).

A second lesson noted by Portes (2000:15) is that “one must know the actors involved and their actual goals in order to anticipate their reactions to external intervention.” Clearly, sociological analysis is needed here, but as Portes distinguishes, it needs to be of the sort done by “social craftsmen” rather than “social engineers” or “social architects.” In effect, greater attention to cultural meanings, historical experience, and other particularistic dimensions of social embeddedness is needed if the potential unanticipated consequences of legal reforms are to become more manifest. Such analysis is easier said than done, but “[w]ithout this painstaking effort, any organizational blue print, no matter how well devised, is likely to yield unexpected outcomes, thus following the fate of so many failed interventions of the past” (Portes 2000:15).

I would add a third lesson that sociologists and policy makers need to be aware of with respect to politically driven legal interventions: the internal “logic” or potentialities within the law itself. The coherence of a particular legislative initiative is by no means given, and there may often be conflicting, if not contradictory, possibilities inherent in the same law. I noted this issue above in relation to the German and French cases above, as well as briefly addressed it with respect to the British case and the possibility of using civil rights legislation to further union goals. I should also mention the well-known argument of critical legal theorist Karl Klare on the NLRA. As Klare (1978:291) argued, as a piece of legislation the NLRA was “marked by indeterminacy, openness and divergency.” On the one hand, it could easily have been interpreted and utilized to create a robust form of industrial democracy for labor unions and working people. Nevertheless, structural factors, later political interventions, and employer strategies, as well as some labor activities themselves, led to a more contractualist business unionism that would characterize U.S. industrial relations throughout most of the twentieth century.

In closing, all of the above lessons should be kept in mind by American scholars, activists, and political leaders as we proceed into the future. Reeling from the Great Recession and with public sector unionism under attack across the country, working people are desperately in need of what David Brody (2004:15) has labeled “A Labor Law for the 21st Century.” Whatever the form or the content of such legislative changes, should the political opportunity for such a policy project even occur, I hope a keen reflexivity by labor’s supporters will be able to anticipate, and perhaps avoid, any deleterious unanticipated consequences that may arise for organized labor. Of course, even if political fortunes do not change in the foreseeable future and labor law reform does not occur, reason for optimism about the continuing viability and vitality of the organized labor movement in the United States, as well as elsewhere, certainly exists. Existing labor law, and even the implementation of more repressive measures, will contain the strategic space for institutional conversion and a return to the logic of industrial democracy. Furthermore, as Beverly Silver (2003:19) notes, “labor and labor movements are continually made and remade” by the dynamics of capitalist development. Thus, new forms of labor agency, resistance, and empowerment, both within and outside legal structures, can certainly be anticipated as the rest of the 21st century unfolds.