Article body

Brazilian labour unions have an excellent reputation in international trade union circles, often being depicted as one of the most dynamic and militant labour movements today. In Brazil, however, many scholars criticize the unions for having abandoned the struggle taken up in the late 1970s against the industrial relations system and the limits it imposes on their actions. Since its creation in the 1930s, the authoritarian corporatist labour relations system has, in fact, proven to be extremely resilient. It survived two periods of intense political upheaval—the trade union movement having played a predominant role in both—and radical change to the country's economic model without undergoing major transformations. The adoption of a new Constitution in 1988 introduced only minor alterations, leaving the pillars of the corporatist system intact until today.

In this article, we put to the test a widespread, determinist vision of Brazilian trade unionism, according to which unions maintain a paradoxical yet atavistic relationship with the heavy body of laws that provides them with advantages while limiting their freedom. In order to reflect on this, we performed field enquiries on two trade unions carefully chosen for their archetypical profile. Our research activities were conducted in the midst of the debate (from June 2005 to July 2006) on the ambitious project to reform the industrial relations system, initiated by former trade union leader and current President Luis Inacio “Lula” da Silva. This gave us the opportunity to discuss with trade unionists the pros and cons of the proposed reform.

We will begin with a brief overview of the history of the trade union movement in Brazil and its interactions with the corporatist labour system. We then describe in more detail the methodology used for this study. Next, we present the results obtained for each one of the unions. Our subsequent analysis allows us to discuss these results in light of the recent attempt at reform and what they reveal about Brazilian trade unionism's relationship with the labour legislation.

History of the Brazilian Labour Movement and the Corporatist Labour Relations System

As in other Latin American countries, the Brazilian labour movement began in the late 18th century under the leadership of anarchists and communists of European descent. Since its inception, the movement experienced successive periods of expansion and decline, as its actions were often met with severe State repression. Unionists had not yet succeeded in consolidating the movement when a coup d'état brought Getulio Vargas to power in 1930.

Soon after assuming office, Vargas began to pass a series of laws with the goal of co-opting the movement and winning the support of the working class.[1] He conceded many social benefits (paid holidays, maternity leave, minimum wage, etc.) considered advantageous in those days. As for trade unions, the laws recognized many union prerogatives, provided that they submit to strict controls over their activities. The official union system, as defined by the 1943 Consolidação das Leis do Trabalho (CLT)—which is the compilation of labour laws passed between 1930 and 1943—functioned according to the following rules:

  • legal recognition of unions: the Ministry of Labour awarded a carta sindical or “union card” to unions who agreed to comply with a series of legal requirements, such as submitting annual budget estimates and union leaders' biographies to the Ministry and requesting permission to elect new leaders;

  • unicidade sindical: the power to represent all workers—union members and non-members—of a given socio-professional category in a given territory was granted to only one union. Thus, workers could not choose their union; they could only decide whether or not to sign up with the one representing their category in their municipality. Labour federations and confederations were also to be organized by socio-professional category. The State drafted the definitions of these categories.

  • imposto sindical or union tax: all workers—affiliated and non-affiliated—had to pay an annual union tax, which employers were obliged to deduct from workers' salaries. The Ministry of Labour was responsible for collecting the tax and distributing it among the different levels of the official union system, retaining 20% of the funds for the Ministry itself;

  • explicit integration of trade unions into the social welfare system:[2] unions could only use the funds from the imposto sindical to finance social services they were required by law to offer to their members, and not for other purposes (e.g., printing flyers, strike funds, etc.);

  • right to strike: in order for a strike to be “legal,” unions had to meet certain requirements, namely quorum at the assemblies held to deliberate over strike action;

  • State intervention in trade unions: in addition to possessing arbitrary powers to award, refuse or revoke a union's carta sindical, the Ministry of Labour could attend union meetings, replace elected union leaders with its own agents and freeze union assets. The restrictions on the use of funds from the imposto sindical allowed the State to survey union finances;

  • conflict resolution and the collective-bargaining process: labour and employers organizations were required by law to engage in collective bargaining. If they reached an impasse during negotiations, either party could request the intervention of the labour courts, known as the Justiça do Trabalho. The tribunals could also intervene out of their own initiative (Gacek, 1994);

  • union representation in the workplace: there were no provisions for this;

  • public servants: the CLT did not apply to public servants. Government employees could not form professional associations or unions.

Typical of corporatist system, then:

the State intervenes to organize social classes with the goal of integrating groups representing work and capital in vertical and non-conflicting organizations. These groups exist in virtue of the recognition awarded to them by the State. In turn, the State defines their modes of action and the relationship between them in a way that ensured that each part's actions and demands would be restrained and harmoniously integrated for the benefit of the entire organism.

Rodrigues, 1990: 54, our translation

Such labour legislation had long-lasting perverse effects on union practices. First, severe limits on union organizing had important “disorganizing” effects on the movement (Boïto Jr, 1991). Strike regulations, the absence of union representatives in the workplace, the imposition of State-defined limits on unions' organizational bases and the prohibition of horizontal ties between unions made mobilizing workers all the more difficult.

Second, by guaranteeing funding and a monopoly over the representation of a given group of workers (through the imposto and the unicidade sindical, respectively), it was the State that guaranteed unions' organizational survival, not the workers. This allowed for, and even encouraged, the creation and maintenance of labour organizations that held distant, if any, relations to the workers they claimed to represent. Or, what was worse, it allowed for the rise of pelegos[3] —that is, union leaders who occupied official union positions in order to advance their political careers or increase their personal fortunes while doing nothing for workers. Workers were obviously alienated from such organizations.

Another custom developed under the corporatist system was that of removing conflict from the workplace and deferring it until after the labour relation had been terminated. The absence of union representatives in the workplace and in-company grievance procedures, along with a law permitting employers to fire workers without just cause, encouraged workers to keep problems at work to themselves until they were fired. Once dismissed, a minority of them sought compensation in the labour tribunals. Thus, workers became accustomed to resolving their problems by appealing to the Justiça, rather than confronting their employer directly. They rarely obtained results from individual appeals to labour courts due, in part, to the courts' “chronic pattern of excessive delay” (French, 2004).

Although some initially contested the creation of the corporatist system, often denouncing it as fascist,[4]Varguista[5] and communist union leaders eventually opted for gaining control of official unions to avoid repression and to use their resources to mobilize workers (Alexander, 2003). Union activities expanded gradually in the 1950s and 1960s, thanks to the election of presidents with more lenient attitudes towards labour: Juscelino Kubitschek (1956–1961) and Joao Goulart (1961–1964). In the early 1960s, the movement became increasingly radical,[6] which was not tolerated for long. On March 31st 1964, the army staged a coup and established a military government. It then began to conduct massive purges of the labour movement, intervening in the unions to replace their leaders with army officers. It also adopted strict laws to make strike activity virtually impossible and it began to determine all wage increases, thereby depriving organized labour of its reason for existence (Alexander, 2003: 179). Pelegos thrived in this context (Riethof, 2002).

During the ten years following the coup d'état, there was virtually no sign of militant union action.[7] Factory committees began to slowly reappear in the mid-1970s and in the late 1970s, strikes in the automotive plants in the ABC region[8] announced the definite revival of labour militancy. The ABC Metalworkers were soon joined in protest by other “categories” and sectors—namely the public sector, which had been virtually absent from the movement before then. These waves of protest continued well into the 1980s, with public servants often playing a central role in them.

Those observing the revival of the union movement—as well as many who participated directly in it—argued that a “new unionism” or novo sindicalismo was being born. They believed that labour activists had made a break with the past.[9]Novo sindicalistas criticized their predecessors from the pre-1964 era for being distant from workers, collaborating with pelegos and accepting the limits of the corporatist system. They claimed to be focussing their efforts on organizing in the workplace, building closer relations with workers, giving priority to defending workers' immediate demands and fighting against the corporatist system and the State. Union autonomy and freedom were key demands of novo sindicalistas. In 1983, they defied the corporatist structure and founded the first union central, the Central Única dos Trabalhadores (CUT).

Although new unionists made some progress in expanding trade union activity, in the end, they did not make a clear break with the corporatist system. Part of the reason stemmed from the inherent problems of the “fight from within strategy” adopted by the dominant opposition faction in the early 1980s. Like their communist predecessors, these union leaders opted for fighting for control of official unions in order to use them to mobilize workers, as opposed to organizing outside of the system. As a result, they ended up revitalizing official unions, submitting (at least to a certain extent) to the limits the system imposes on union actions, while workers demobilized.[10] With official unions and their resources under their control, radical change of the corporatist system became less urgent for many CUT leaders. Leaders of the other main faction of the movement (who would later found the Força Sindical union central in 1991) defended maintaining the system as it was.

The federal Constitution of 1988 introduced some significant changes to the corporatist labour relations system. It removed one of its most repressive elements: the Labour Ministry's extensive powers to intervene in union affairs. For instance, legal requirements for founding a trade union were relaxed and the Ministry's role in this process was reduced to that of ensuring that the paperwork was in order. The Constitution also recognized public servants' freedom of association and guaranteed all workers' right to strike. Finally, private sector unions won the right to establish representatives in the workplace in companies with 200 or more employees. However, unions have had great difficulties in overcoming employer resistance to their presence in the workplace, as we shall see shortly (Rodrigues, 2000).

Although the Constitution did recognize certain freedoms or rights, it failed to introduce radical change to the corporatist system, as it left its three pillars intact. Unicidade sindical remained untouched and the imposto sindical was merely given a more politically correct name: “contribuição sindical”, or union contribution. The maintenance of these two elements, combined with the relaxation of the requirements to form a union, produced a grave and unexpected effect: literally thousands of “paper unions”—that is, organizations existing only on paper—have been created solely for the purpose of collecting the funds raised by the contribuição sindical (Zylberstajn, 2005). As for the third pillar, the Constitution did not alter the Justiça do Trabalho's powers to intervene in labour conflicts.

In the 1990s, changes to the country's economic development model forced the trade unions to revise their positions and strategies. The adoption of neo-liberal policies and new human resource management techniques, resulting in massive layoffs, led unions to focus their efforts on defending workers' jobs. Government attempts to pass reforms of the official union system were generally unsuccessful during this period.

Methodology and Choice of Case Studies

Our review of the literature on Brazilian trade unionism has enabled us to note that few empirical studies have been conducted to examine the impact of the country's corporatist labour laws on unions' day-to-day activities. To answer our main research question, then, we carried out two in-field case studies. Our main hypothesis was: while the corporatist labour legislation continues to exert a strong influence over unions' actions in various aspects, unions have a certain room to maneuver that they can use to develop practices that go beyond what is foreseen in the legislation.

Our initial criteria for selecting unions for our case studies were: (i) they be based in Sao Paulo, the hearth of all industrial and economic activity in Brazil; and (ii) they represent workers from the private sector. To narrow our options, we used affiliation to a union central as a third criterion for dividing unions into subcategories. A union's choice of central indicates its leaders' and members' ideological positions, which often influence union practices. Here, although there are several centrals in Brazil, we chose to study unions affiliated to the CUT and Força Sindical (FS), as they are the two most important organizations in terms of number of affiliates and actors on the country's political scene.[11] Historically, CUT and FS hold very distinct political positions: while the former defends socialist or social-democrat ideas, the latter clearly embraces capitalism.

Within both the CUT and FS, one may find organizations with varying levels of activity in terms of their attempts to establish relationships with workers and defend their interests. A highly active union, for example, consults workers before initiating annual collective bargaining activities and tries to establish its presence in the workplace. An “inactive” union rarely has contact with workers and limits its activities to offering a few of the “traditional” union services (legal aid, health care or leisure facilities).

In general, only a minority of private sector unions affiliated to the CUT and an even smaller proportion of FS affiliates can be classified as “highly active”; most CUT and FS unions lack effective contact with the workers they claim to represent (Rodrigues, 2000). Even though the majority of Brazilian labour organizations can be considered “inactive,” we chose to focus on those in the “active” category, as our interest lies in the unions' daily activities. Also, it is difficult to gain access to the routine activities of inactive unions.

Based on these criteria, the following unions, which we consider archetypical, were selected, with the intention of verifying if they differ and if so, in what dimensions:

  • Seamstress Union for the Sao Paulo and Osasco Region (Costureiras): This union is one of the most active unions affiliated to FS and the large majority of the workers it represents are women.

  • ABC Metalworkers Union (Metalworkers): This union is linked to the CUT and is one of the most advanced unions in terms of organizational capacity and political influence. In fact, the ABC Metalworkers union plays a strong leadership role in the movement, as other unions often attempt to implant the Metalworkers' practices in their own bases.

We used two types of research methods for the study of these unions: participant observation and semi-guided interviews with union leaders and members. We observed a variety of union activities ranging from monthly barbecues for affiliates to protests to negotiations with company representatives. We conducted nine interviews with leaders and members of the Metalworkers union and eight with leaders and members from the Costureiras union. In both cases, the number and choice of interviewees was restricted by our limited access to workers.[12]

We chose these research methods, as we felt they complement one another well. Semi-guided interviews are ideal for obtaining more detailed information on union practices as well as union leaders' and workers' perceptions of labour relations. Observing union activities helps to put interviewees' responses into context and allows us to compare our results from the interviews to our observations, so as to detect differences between discourse and what is actually done.

Results of Our Case Studies

The Seamstress Union of the Sao Paulo and Osasco Region

The Sindicato das Costureiras,[13] as it is commonly known today, was officially created in 1942. Since its foundation, the union has held the monopoly of representation over all workers employed in the clothing industry in the municipalities of Osasco and Sao Paulo. There are approximately 7 500 legalized confection companies operating in the union's territory today, the majority of which are small and medium-size enterprises.[14] Union leaders estimate the number of registered workers in the union's territorial base to be currently at 80 000 in the formal sector and 100 000 in the informal sector.[15] The union's current president is also the vice-president of the Força Sindical union central.

The union's top-down organizational structure is divided into two levels, the first being the union's Board of Directors, which is made up of an Executive Committee, a Fiscal Council and a “Representative Council linked to the Federation.” All together, there is a total of 24 members on the Board (the maximum number allowed by the CLT) and they make all the decisions regarding union activities.

Factory commissions constitute the second level of the union structure. The union claims to have established approximately 1 100 commissions[16] in the workplace in the past ten years by using the law of the “acordos de compensação de horas” or compensation of hours agreements. According to one union leader, the CLT stipulates that any company operating in a “compensação de horas” system—that is, from Monday to Friday with compensation for work on Saturday—is obliged to sign an agreement with the union regarding the work schedule. The union succeeded in including in these agreements the right to set up a factory commission. This is one example of how the Costureiras have been able to use a mechanism of the CLT to gain access to the workplace in a way that was not foreseen by the legislation.

Another legal mechanism used in a similar (albeit less successful) way is the Internal Commission for the Prevention of Accidents, or CIPA in Portuguese.[17] A CIPA is an in-company bipartite commission (employer, worker), foreseen by the CLT, whose mandate is to identify measures necessary to prevent accidents in the workplace. The Seamstresses' Union attempts to use the CIPAs to establish contact with workers by offering a course for CIPA members (cipeiros) and encouraging companies to send their employees for CIPA training at the union's office. The union also “accompanies” the elections for the commission and in some firms, one cipeiro is a member of the union. In many companies, however, the CIPAs still only exist on paper.

The Costureiras' limited presence in the workplace makes it difficult for the union to resolve conflicts directly with employers.[18] Our observations indicate that the factory commissions are not very effective in forcing employers to respect agreements or the law. This is due, at least in part, to the fact that commission members receive little, if any, training for their position. Union leaders do not encourage them to act independently to resolve grievances directly with employers, instructing them to consult union officials before acting. Without a union representative with effective negotiating powers in the workplace, workers often keep their complaints to themselves for fear of employer reprisal, reporting their problems to the union only after they have been fired. Another factor that encourages this type of behaviour is the fact that the majority of seamstresses are females working in small enterprises (i.e. close proximity to often male bosses) and harassment is still a common practice in Brazil.

According to leaders interviewed, when they do receive complaints from a worker, they attempt to initiate direct negotiations with the company. If the firm refuses to respond, union leaders organize protests na porta da fábrica (at the factory door). If the union is not able to obtain satisfactory results or negotiations reach an impasse, the union requests a meeting with company representatives at the Labour Ministry's “round tables”, in which government agents act as mediators. Union recourse to this mechanism is quite frequent.[19] Should the problem remain unresolved, the union can appeal or file individual cases with the labour tribunals. All union leaders confirmed that the union's legal department is currently defending several cases in the labour tribunals. It is clear, then, that the Costureiras union has not developed much autonomy in this area, remaining highly dependant on the State to intervene to resolve disputes.

State intervention during the collective bargaining process, on the other hand, is not common. Union leaders claim that neither the Costureiras union nor the employers' organizations normally request court intervention. Recourse to the labour tribunals, as well as strike action, are to be used as last resort mechanisms:

Judicial actions and strikes are only used as a last resort. This has always been our line of thinking. When you end up with one of these kinds of actions, you have already lost . . . You have to negotiate, negotiate, negotiate. If there is no other way, then you have to stop production . . . We make it clear that we don't want to get to this point.

Wanda, Union President[20]

In the union president's eyes, requesting court intervention is a sign of defeat, mainly because the labour tribunals tend to side with the employers. Strikes are to be averted so as to avoid causing economic losses for the companies, which could have a negative impact on workers.[21]

While the union's collective bargaining activities are relatively free from State intervention, the process of negotiating the collective agreement that stipulates the working conditions and salaries of all workers in the Costureiras' base (convenção coletiva) continues to follow the routine and rules defined by the corporatist labour legislation. Negotiations between representatives from the Costureiras union and the employers' organizations begin a few months prior to their database—the official date for annual negotiations—in July. When an agreement has been reached, the union presents the employers' offer to workers in an assembly for their approval. Once approved, the union begins negotiating “acordos coletivos”—that is, company-level agreements negotiated to try to obtain conditions that are better than those contained in the collective agreement.

In addition to imposing a framework on the collective bargaining process, the CLT also contains a detailed list of benefits and rights that are (in theory) guaranteed for all registered workers. One may suppose, then, that the CLT influences the content of negotiations. For Costureira officials, the CLT is an important reference used during the collective bargaining process: “we take the CLT, identify a few elements and try to gain something better” (Wanderley, Executive Committee Director). Thus, the CLT shapes union demands. An examination of the union's collective agreements confirms this: one finds numerous clauses that merely repeat what is stipulated by the labour laws.

The Costureiras union leaders defend the monopoly of representation awarded to local unions by the State. They argue that if unicidade sindical were eliminated, thousands of unions would be created in the bases of existing unions, thereby further contributing to the fragmentation and weakening of unions' bargaining power. They are willing to accept pluralism only at the higher levels of the structure (federations, confederations and centrals).

With regards to the third pillar of the corporatist system, the union's main sources of funding are the contribuição sindical and voluntary membership fees. Interviewees claimed that the elimination of the imposto would not affect their union. That said, the union president is in favour of some kind of obligatory tax for all workers, as she believes that no one would pay if the law did not force them to do so.

ABC Metalworkers Union

The Sindicato dos Metalúrgicos do ABC,[22] or the ABC Metalworkers Union, was founded as an official union in 1932. It has often played a central role in the labour movement, especially during the resurgence of union activity at the end of the 1970s. The union also contributed greatly to the founding of the CUT, in which it still has much influence today. President Lula's political career began as a leader of this union.

The ABC Metalworkers Union (herein referred to as the Metalworkers) currently represents all metalworkers in the metallurgical, mechanical and electric material plants in the ABC region—with the exception of those from Sao Caetano do Sul.[23] There are over 100 000 metalworkers employed in more than 1000 companies in the region. The employees of the automotive plants account for 60% of all metalworkers in the union's jurisdiction, of which approximately 75% are union members.

The union's organizational structure differs greatly from that of the Costureiras and has changed significantly since the 1970s. It was originally composed of 24 directors, plus a few workers' committees operating clandestinely in the factories in the 1950s, the 1960s and again in the late 1970s. In 1981, the union succeeded in establishing its first “official” factory committee in the Ford plant—official in the sense that the committee's activities were regulated by a company-level agreement.[24] Following the victory in the Ford factory, the union set up committees in several other companies, the majority of them being in the larger automotive plants. Their mandate is typical of any factory committee and all workers—union members and non-members—may participate in elections.

In the 1980s, in addition to establishing factory committees, the union mobilized to get its members elected to the CIPAs:

Before winning the right to establish a factory committee, the mechanism union activists used here in the factory was the CIPA, because their stability is protected by law. They succeeded in acting to a certain extent in the factory, talking to workers and all. Thus, the CIPAs were an important instrument for the union movement at the end of the 1970s, early 1980s.

Ronaldo, factory committee member at the Mercedes plant in Sao Bernando do Campo

As this quote suggests, the union would get their members elected to the commission, so as to benefit from the legal protections bestowed on CIPA members during their mandate. Cipeiros would use their position to communicate with workers and set up factory committees. In this way, CIPAs played a central role in the union's efforts to expand its presence in the workplace.

In 1999, the union began to create Comitês Sindicais de Empresa (CSE) or Company Union Committees in order to further its presence in the factories. The CSEs were to have the same function as factory committees and their activities are regulated by company-level agreements. However, only union members may participate in CSE elections. Once again, the union used the CIPAs to implant CSEs in factories throughout its territory:

And then something very interesting happened. As there was no legal provision on this matter, there was no legal impediment. . . . So we started to put in the [CSEs] people who had stability as elected members of CIPAs. . . . There wasn't much the company could do because these people already had their stability guaranteed for another year—the time they were still in the CIPA. We gave our CSE members a 3-year mandate.

Danilo, Executive Council member

When a cipeiro's mandate ended, the union would convince the company to let workers finish their CSE mandates. By doing so, the company accepted de facto that a CSE be established in their factory.

The CSEs did not, however, replace the existing factory commissions. In some factories, one may find a CSE and a factory commission, as well as union activists occupying positions in the CIPA (CIPAs are not part of the union structure).[25] In plants where factory commissions are already in place, the union may also set up a CSE in order to establish the highest number of workers' representatives on the shop floor as possible.

The CSEs are at the heart of the union's organizational structure today. Elections for CSEs are held in the factories and the number of CSE members is proportionate to the number of union members in a given factory. As of March 2006, the union had CSEs in 88 enterprises and one CSE for retired workers, with a total of 240 CSE members. All 240 of them make up the “Diretoria pleina do sindicato,” or the Full Directorate, which constitutes the first level of the union's organizational structure.

The next level is that of the Conselho da diretoria or Executive Council. The council is made up of 27 members elected from the 240 CSE members. Council members choose 9 persons amongst themselves to sit on the union Board of Directors and decide which position each board member will occupy. The only member of the board who is elected directly by workers is the union president.

As for collective bargaining activities, like the Costureiras union, the Metalworkers follow the schema stipulated in the CLT. The Metalworkers' union's database is set for November of every year and negotiations begin a few months earlier so as to reach an agreement before the database. The CUT Metalworkers' Federation for the State of Sao Paulo and the São Caetano do Sul Metalworkers' union (FS) represent workers at the bargaining table and must negotiate with several groups of employers' organizations.[26] When an agreement has been reached between parties, the union holds an assembly to present the employers' offer to all workers for their approval or disapproval. Once approved, the ABC Metalworkers Union leaders then proceed to negotiate acordos coletivos with individual companies to try to secure conditions that are more advantageous than those included in the collective agreement.

Union leaders admitted that like the Costureiras Union, they still use the CLT as a reference prior to negotiations to verify what is already guaranteed by law and which aspects they can improve upon. In this way, the CLT shapes this union's demands to a certain extent. Contrary to the Seamstresses' Union, however, the Metalworkers Union is able to introduce new items at the negotiating table and in collective agreements.[27] The factory committees are a case in point: the union began mobilizing workers clandestinely (thus outside of law). Through strike action, they were able to have the committees officially recognized in an acordo coletivo.

Justiça intervention during the Metalworkers' collective bargaining process is also rare. The union does not rely on the tribunals to intervene during negotiations, yet for different reasons than the Costureiras. Negotiating agreements and resolving conflicts by themselves is clearly seen by Metalworker leaders as a sign of union strength; for them, only pelegos or weak unions resort to the Justiça. Companies, on the other hand, occasionally request intervention, usually in reaction to the union's decision to initiate strike action. During annual negotiations in November 2005, for instance, Volkswagen asked the tribunals to prevent workers from going on strike. The judge responded positively to the company's request, threatening to impose a heavy fine on the union if it declared a strike. Workers ignored the threat and stayed off the job for 30 days. Thus, even though the union still has to deal with possible intervention from the Justiça, this does not stop it from engaging in autonomous forms of action (strikes) to defend its demands during negotiations.

As for conflicts arising in the workplace, union leaders attempt to negotiate a solution directly with the employer. In companies where a CSE, a CIPA or a factory committee exists, committee members are responsible for this task; in smaller firms where the union is not present, higher-level union leaders are called upon to intervene directly.[28] Union intervention is more effective in companies where union leaders are present.

According to union leaders, the large majority of conflicts are resolved through direct negotiations. If no agreement is reached, they may organize a short-term strike to pressure the company to give in to their demands. When strike activity fails to produce the desired results, the union may explore the possibility of presenting a case before the labour tribunals. Judicial actions initiated by the Metalworkers Union are quite rare and are employed only as last-resort mechanisms, as they are, in the leaders' opinion, signs of weakness. This negative view of the Justiça—common among Novo sindicalistas in the 1980s—has been incorporated into the values of this faction of the movement, which indicates that there has been a change in union culture.

The Metalworkers Union would also like to have more autonomy in deciding which group of workers they represent. Past attempts to negotiate benefits for the administrative or cleaning staff of a company, for example, have been contested and won by smaller unions whose monopoly of representation of these workers is guaranteed by unicidade sindical. Thus, the Metalworkers are in favour of its abolition, as this quote clearly suggests: “If [unicidade sindical] were to be eliminated tomorrow, half of the unions would disappear. And what's worse, workers wouldn't even notice.” (Danilo, Executive Council member).

As for the obligatory union tax, union leaders are proud to say that their union must go to court every year to have workers in its jurisdiction exempted from the contribuição sindical. The Metalworkers Union was, in fact, one of the first unions to return these funds to workers. To ensure financing for its activities, the union has created a new tax, the “negotiating fee”, that all workers in its base must pay. It justifies its actions by stating that since all workers benefit from the union's collective bargaining activities, everyone should have to contribute. Although union leaders express some pride about going against the system, in the end, they reproduce its logic by imposing fees on all workers in their base.

Discussion: Our Case Studies and the Resilience of the Industrial Relations System to Reform

Our case studies clearly show that the corporatist labour legislation in Brazil continues to have a significant impact on the practices of private sector unions. The collective bargaining process still runs according to the rules and procedures established by the CLT, and the CLT shapes union demands. That said, it is clear that despite the rigidity of the laws, some unions attain the capacity of going beyond them, imposing new rules on employers and adopting better codes of conduct for internal democracy. As stated earlier, literature on Brazilian trade unions suggests that our Costureiras case is much more the norm than the Metalworkers. Indeed, the ABC Metalworkers Union[29] is one of the few Brazilian unions with sufficient organizational and mobilizing capacity to introduce new subjects on the bargaining table; the majority of unions, like the Costureiras, settle for slight improvements on what the CLT already guarantees. While union recourse to the Justiça during annual negotiations has become less frequent, the same cannot be said for the resolution of workers' complaints in the workplace. The majority of unions in Brazil are similar to the Costureiras in that they still rely on the intervention of the Ministry of Labour or the Justiça to resolve such disputes. Only a few unions are able to negotiate solutions directly with employers like the Metalworkers do. The Metalworkers union leaders' disdain for requesting State intervention indicates that there has been a change in union culture among members of this faction of the labour movement.

Yet even within the rigid framework of the corporatist laws, some margin to maneuver does exist and is used by unions to strengthen their organizational capacity and improve their relationship with workers. A prime example of this is the Metalworkers' use of CIPAs to set up factory committees with the capacity to effectively defend workers' rights. In this aspect, this union has gone far beyond what is foreseen by the law. Unfortunately, the Metalworkers Union is unique in this area. Other CUT unions have attempted to implement the Metalworkers' model in their bases, but they have had only mitigated success. Though many have been able to get union members elected to CIPAs and some have delegates present in the workplace, very few have in-factory committees. As for FS affiliates, the Costureiras union is one of the rare organizations that have attempted to establish any presence in the workplace, let alone contact with workers. As we observed, this union's factory committees are not very successful in defending workers' rights.

As for union leaders' views on unicidade and the contribuição sindical, the ideas they defended during the interviews are in sync with the CUT's and FS' official positions. While the CUT's national executive supports putting an end to unicidade, the FS wants it upheld so as to guarantee the survival of its members, allowing pluralism only at the upper levels of the structure, which, in fact, already exists. Both the CUT and the FS have their own National Metalworkers Confederation, for example. With regards to the obligatory union tax, leaders from both unions hold somewhat ambivalent positions towards what should be adopted in its place.

Our research demonstrates, then, some heterogeneity in union practices and views. A small minority of CUT unions have developed more autonomous actions, whereas the vast majority of Brazilian unions find themselves stuck in the CLT straight jacket, relying on it for their survival. This is why the newly elected President Lula, who had experienced as a trade unionist the difficulties of building strong unions within the framework of the CLT, included reforming the corporatist system in his list of priorities during his first presidential mandate.

Shortly after assuming office in 2003, President Lula announced the creation of a tripartite committee—the National Labour Forum (Forum Nacional do Trabalho)—whose mandate was to draft a proposal for reforming the corporatist labour relations system.[30] After 18 months of heated negotiations and despite the lack of consensus among the Forum's members, the Minister of Labour submitted a reform project with over 200 articles to Congress. Although the proposal is too complex to be analyzed in detail here, we will briefly discuss the main points and their possible impacts on unions' actions and autonomy.

One of the most controversial issues was that of union representation in the workplace. If the reform had been adopted, unions would have been granted the power to establish at least one representative in every business with 35 or more employees. The representative's mandate would have been to negotiate solutions to workers' problems directly with employers. Thus, the reform would have introduced a new first step in the conflict resolution process. This alteration, defended by progressive CUT unions, should have helped active unions to increase their bargaining power with employers and encouraged them to develop more autonomous actions. On the other hand, as the reform would have maintained and even facilitated union recourse to the labour courts, it would have allowed unions that currently rely on State intervention to continue to do so. It is likely, then, that the overall impact of this change would have been limited.

Proposed changes to financing unions included replacing the contribuição sindical and two other existing union taxes with a single “negotiating fee.” Union leaders claim that this tax would be “more democratic” as workers would define the amount during an assembly. Even so, all workers—union members or not—would still have been obliged to pay the fee. Thus, although labour determines the amount, by making it obligatory, the State would have continued to guarantee union funding to a certain extent. Such a functioning could not be qualified as heretical by North American trade unionists who find normal that all employees pay the union dues because they benefit from collective bargaining activities. In Quebec, it is even compulsory. Another example is the French trade unions that are financially supported by the State without question. But the situation in Brazil raises other problems. Employers do not always recognize trade unions, collective bargaining is often a futile exercise, and cases of union corruption persist.

The boundaries of union organizing at the local level would also have continued to be defined by the system, and not by unions. First of all, the reform project proposed that workers be regrouped by ramos or “branches” of economic activity, as opposed to socio-professional categories. Although this would have allowed unions to broaden their base of representation, they would still not have had full autonomy to define the limits of their membership, as a tri-partite commission would have been responsible for drafting the definitions of the branches. Secondly, existing unions would have been allowed to retain their monopoly of representation provided that they prove to the Labour Ministry they are “representative”—that is, 20% of all workers in their jurisdiction are union members. The other option would have been to “borrow representativeness” from higher-level organizations (federations, confederations or centrals). In addition to reintroducing the Labour Ministry's control over unions' internal activities, this measure could have increased unions' reliance on the centrals.

One of the few proposed measures obtaining consensus in the labour movement was the recognition of the centrals as part of the official union system. The extensive powers they would have been awarded, however, gave rise to much debate. According to the reform project, the centrals could initiate negotiations at any level and the results of these negotiations could not be altered by bargaining activities at lower levels. Furthermore, the centrals would have been able to found lower-level organizations to complete their “organic” structure. Many actors criticized this attempt to concentrate power in the hands of the centrals at the expense of local unions.

The reforma sindical submitted to Congress in 2005 was, in fact, the result of a series of compromises between a minority of union leaders wishing to introduce some change to the system and the large majority of actors who defend maintaining the status quo. Had it been adopted, it would have contributed little to providing unions with a greater leeway to increase their autonomy and strengthen their relations with workers. On the contrary, the reform as proposed might have helped to perpetuate the top-down, bureaucratic model of union organization and to maintain a significant level of State intervention. We can hypothesize that the authors of the reform tried at the same time to favour the progressive Cutista unions, control the corrupt elements and permit unions to build a better balance of power vis-à-vis employers. Such contradictory ambitions had the inevitable consequence of raising criticisms from all sides. Employers were unanimously against it, mainly because it would have forced them to accept union presence in the workplace and because it did not include proposals to “flexibilize” workers' rights. Trade unions were divided, some declaring in public their opposition to it. In the end, the government was forced to abandon the reform project, opting more pragmatically for presenting parts of the reform as separate laws to the Congress. Since the beginning of Lula's second presidential mandate, for instance, a law recognizing union centrals as part of the official union system—with more restricted powers that what was originally proposed—has been adopted. Other laws proposing change to certain aspects of the labour relations system are awaiting judgment in the Congress.

Conclusion

In this article, we attempted to contest a determinist view of the influence of Brazilian labour laws on industrial relations and trade unions' ways of functioning. We demonstrated that, while the CLT is still a straightjacket for many unions (whose leaders do not necessarily try to free themselves from it), it was possible for other unions to innovate and introduce more autonomous practices. It would be too simplistic, however, to conclude that if some were able to innovate, everyone else should be able to do the same. Labour laws determine the ability of trade unions to do so only to a certain extent; other elements of analysis must be considered. Bargaining strength, for instance, differs greatly according to a series of objective factors. The configuration of the Brazilian labour market—with its millions of unqualified workers seeking stable employment (the majority of them being unemployed or employed in informal sector)—presents a major challenge for trade unions in all sectors. Another part of the equation is the will to change. In this sense, establishing union democracy in Brazil still requires profound reform of the country's current labour laws.

One of the main challenges to reforming the system is the difficulty in making a clear distinction between rights that have been conceded to unions as part of a corporatist framework and rights that should have developed gradually in a democratic society. To put it another way, President Lula's explicit intention to transform an authoritarian corporatist framework into a pluralist neocorporatist one has proven to be too ambitious. Another challenge is convincing actors involved that increased autonomy from the State could contribute to a much-needed revitalization of the movement. Finally, for real change to occur, it will be necessary to remove from positions of power those whose interests lay more so in their organization's survival than in the defence of workers' rights and interests.