.

The Transformation of the Mexican Labour Regulation Model and its link to North American Economic Integration

Graciela Bensusán

# Abstract

This working paper analyses the transformations of the Mexican labour regulation model in its different phases. The first phase is selective inclusion of waged workers among the beneficiaries of the development driven by the internal market. The second phase includes the deactivation of labour protections in the wake of successive economic crises and the move towards a growth model based on exports in the framework of the North American Free Trade Agreement (NAFTA). The third phase begins with the deep institutional reforms linked to renewed trade negotiations with the United States. Finally, the paper presents the genesis and characteristics of the new labour regulation model, which is based on the 2017 constitutional reforms and the 2019 Federal Labour Law, and examines the impact that these developments have had on workers. The paper has at its core the linkages between the Mexican labour regulation model and trade liberalization.

# Introduction

The Mexican labour regulation model has passed through various stages since its institutionalization between 1917 and 1931, having later stabilized under the post-revolutionary political system of the 1940s.1 During the first stage, from the 1930s to the mid-1970s, salaried workers were part of a select group that benefited from an economic model driven by the internal market. The second stage was characterized by a weakening in the protections offered by labour institutions, in the wake of successive economic crises and as a result of the move towards a growth model based on exports and on economic integration with other North American countries. The last and most recent stage began with the reform of article 123 of the Political Constitution of the United States of Mexico on 24 February 2017 and its transposition into the Federal Labour Law in 2019.2 The origin of these reforms lies in the negotiations on the Trans-Pacific Strategic Economic Partnership Agreement (TPP)3 and on the new trade deal to replace the North American Free Trade Agreement (NAFTA), known as the Agreement between the United States of America, the United Mexican States and Canada. These reforms were also driven by long-standing calls for the democratization of the Mexican working world, going back at least to the trade union revolt in the 1970s and continuing to the political shake-up in Mexico in July 2018.

The institutionalization of the previous labour model was the result of a dual legislative and political process. In 1931, the Federal Labour Law regulating the implementation of article 123 of the 1917 Constitution was adopted and, in 1938, trade unions became part of the structure of the single-party State, ruled by the Party of the Mexican Revolution (Partido de la Revolución Mexicana), which held uninterrupted power between 1929 and 2000 (first as the National Revolutionary Party (Partido Nacional Revolucionario), then the Party of the Mexican Revolution, then the Institutional Revolutionary Party (Partido Revolucionario Institucional) (Bensusán 2000).

As Lorenzo Meyer (1977) warns, one of the central features of the former Mexican political system was its authoritarianism, as defined by Juan Linz (1975) in his classic study of political regimes. The system was characterized by limited pluralism, a poorly defined ideology and an absence of large-scale social mobilization, except during the initial phase. Although Mexico claims to be a democracy, and was formally established as such by the 1917 Constitution, a single political coalition held power in the country from the time when the new post-revolutionary order first emerged until 2000. From 1929, this coalition eventually formed a single-party State that held executive power at the federal and local levels, as well as a majority in the legislature and the judiciary, thereby consolidating their “effective and continuous monopolization of power”. In short, the formation of the Mexican State was characterized primarily by the centralization of power around the figure of the President, the non-division of power in practice and the corporatism of social organizations (for both workers and peasants), represented in a single-party State (Meyer 1977).

As often occurs in this type of political system, the corporate–statist arrangement required the recognition of monopolies to represent the interests of workers, which acted as privileged agents in their dealings with the Government (Schmitter 1992). Mexican corporatism, which took shape in the late 1930s and was consolidated during the following decade, expanded the social base of the post-revolutionary regime. Unionized workers were given certain benefits in exchange for their loyalty, and institutional channels were created to allow the State to take control of social and political conflicts (Meyer 1977; De La Garza 1988 and 1991). The process that led to widespread corporatism among trade unions received significant institutional support through the promulgation of a labour law that, despite its apparent pluralism, did not recognize any other form of representation for workers except for that provided by organizations that were supported, and controlled politically, by the State. Coercive forms of affiliation were used in collective agreements (such as exclusion clauses that make trade union membership obligatory), and practices were popularized that entirely undermined the internal democracy and transparency of trade unions, which led to the creation of artificial majorities (Bensusán 2000, 218).

Consequently, despite the political transition to democracy that began in 1988 — or even earlier, according to other studies4 — and the political change seen in the Government following the victory of the opposition candidate in the 2000 presidential elections, state corporatism remained unshakeable and continued to serve various political parties and economic models. The transition to a new labour model and the democratization of the trade unions was therefore postponed until a better opportunity arose.

Conversely, each labour reform implemented over the course of the past nine decades maintained the three pillars of the Mexican labour model: state control over worker organization and vindication, intervention of the executive branch in dispute resolution through a tripartite justice system, and the concentration of power in the hands of trade union executive boards. Notably, the executive branch maintained control over the administration of justice on labour matters through the use of conciliation and arbitration boards. These boards, which comprised government representatives, workers and employers, provided a mechanism through which the Government could control the system by acting both as the arbitrator and as an interested party.5 The weak rule of law paved the way for the introduction of this model, which was characterized by its ambiguous design and high levels of discretion and which operated in different ways under the two development strategies used in Mexico over the years.

At first, when the approach to economic growth was focused on strengthening the internal market through the substitution of imports, the labour model allowed formally employed workers to benefit from development (De La Garza 1988 and 1991; Palma 2011). Later, when the critical failures of the old economic model became apparent and the State was forced to refocus the economy away from commercial liberalization towards exports, primarily towards the United States (US), the labour market became more flexible and jobs became more precarious, a trend that has continued to this day and has resulted in a loss of purchasing power and decreased access to benefits for workers.6

As will be discussed in this article, a combination of different mechanisms and practices were used to undermine the protections supposedly provided by labour institutions. From the point of view of successive Mexican governments and various multinational companies, this strategy fulfilled its objectives, having managed, surprisingly, to push down wages without upsetting the industrial peace (Bensusán, Carrillo and Ahumada Lobo 2011). Consequently, between the mid-1980s and 2018, successive governments from across the political spectrum (both the Institutional Revolutionary Party and the National Action Party) maintained this strategy and pursued the same political policy. As a result, efforts to democratize the labour model, as called for during the trade union uprising in the 1970s, were postponed. Furthermore, many researchers (De Buen 2013; Blanke 2007; Bensusán and Middlebrook 2013; Vega 2019) agree that the system of trade union representation in Mexico was almost entirely distorted, serving more to protect the interests of employers and maintain industrial peace than it did to defend the rights of workers, in exchange for which trade union leaders received unfair economic advantages.

The third stage, which began with the constitutional reform of 24 February 2017 and its transposition into labour legislation on 30 April 2019, involved a radical reworking of the old model with the aim of giving workers control over the exercise of their collective rights. This was accompanied by a new labour policy to improve wages, adopted by the recently elected Government in December 2018 (Bensusán 2019a).

Although this transformation was driven in part by international pressure during negotiations on trade agreements (the TPP and the Agreement between the United States of America, the United Mexican States and Canada), there were other factors. As mentioned earlier, the history of these changes dates back to the demands made by various sectors for the recognition of their rights, such as during the “Democratic Trend” protests held by the national electricians’ union in the 1970s. Various attempts to establish independent trade unions have also been made, albeit with little success (Trejo Delabre 1978). These demands were recognized much later in the various reform initiatives adopted by opposition parties, independent trade unions and labour experts (Bensusán and Middlebrook 2013). During the most recent presidential elections in June 2018, they also formed part of the winning candidate’s policies on freedom of association and wages.7

To further develop these arguments, we will examine the changes made to the Mexican labour model during the first two stages. We will then compare the agreements and labour standards linked to NAFTA and to the Agreement between the United States of America, the United Mexican States and Canada, and we will identify the differences between the two instruments. Lastly, we will discuss the genesis and characteristics of the new labour model, which is based on the 2017 constitutional reforms and the 2019 Federal Labour Law, and we will examine the impact that these developments have had on workers. Throughout this analysis, we will focus on the relationship between the Mexican labour regulation model and trade liberalization.

# From selective inclusion to deactivation of workers’ protections

Between the 1950s and the 1980s, the Mexican labour regulation model favoured workers in the most dynamic economic sectors, such as oil, mining and the automotive industry. During this period, economic growth was dependent on the expansion of domestic consumption. This model perfectly suited large private and state businesses, in which official trade unionism reigned supreme, owing to its links to the State. These institutional agreements were most effective in the run up to the mid-1970s, a period of “stabilizing development” characterized by economic and real wage growth and by price stability. Rather than ensuring industrial peace, this model led to a substantive exchange of benefits with regard to wages, social security, a rise in purchasing power and social benefits.8 Although trade unions steered clear of the productive sector, a political negotiation (“controlled negotiation”) over salaries took place that allowed salaries to be linked to the productivity of each sector, with the exception of the most developed sectors such as the automotive industry (Reyna 1978; Palma 2011).9

During the second stage of the development of the Mexican labour model, which was marked by successive economic crises (in 1976, 1983, 1987 and 1994) and, later, by the adoption of the export model (with the signing of NAFTA), various aspects of the labour model (number of employees, functions, working hours, and so on) became more flexible of their own accord, without the need for legal reform. Employment conditions declined as a result of the losses endured during the negotiation of collective agreements and contratos-ley [industry-wide legal agreements] in important economic sectors. Another factor was the exponential expansion in subcontracting in sectors such as the automotive industry, which encouraged enterprises to outsource jobs to companies that manufactured car parts or provided other specialist services such as logistics (Arteaga and Carrillo 1988).

This rise in outsourcing, together with the sudden drop in, and subsequent stagnation of, wages, was the result of the deactivation of labour institutions that had been responsible for the important progress achieved during the previous stage, such as the introduction of collective bargaining with large businesses, which had previously been carried out mainly by the State. Another important achievement was the adoption of the active wage policy, which the National Commission on Minimum Wages had been responsible for implementing since the 1960s; however, following the start of a wave of privatizations in 1982, the Commission began operating in direct contradiction to its mandate (De La Garza 1992; Bensusán 2019a).

As a way of combating the successive crises of the 1980s, the Government agreed a number of economic deals with the main allied trade unions with the aim of reducing inflation through wage increases; however, these increases were ultimately lower than the overall rise in prices. This strategy led to a considerable loss of purchasing power in real terms for workers on the minimum wage, the real value of which had, by 2016, dropped to 25 per cent of its value in 1982, according to the National Commission on Minimum Wages (Moreno-Brid and Garry 2015).10 Contractual wages were also affected, first by the introduction of limits on wage increases resulting from economic deals and later by the imposition of implicit minimum wage limits during negotiations, in line with the model percentage increases for the minimum annual wage set by the Commission. In fact, as it was the labour authorities themselves that controlled collective bargaining in the most important economic sectors (which were under federal jurisdiction), institutional channels were used to impose wage caps despite the inflation crises (De la Garza 1993). These wage adjustment policies included in the various economic deals were accompanied by other fiscal and currency exchange measures, as well as a rise in privatization and trade openness, which enabled Mexico to accede to the General Agreement on Tariffs and Trade (GATT) in 1986 (De la Garza 1993).11

Although official trade unionism, in particular between 1983 and 1987, threatened to lead to general strikes in protest against the introduction of an adjustment process that would lead to lower working conditions, the leaders of the largest central trade unions ultimately embraced government policy on the matter (Bensusán 1987, 262). Since that time, few collective agreements with public and private businesses have survived without suffering dramatic changes to working conditions, in particular in the automotive industry. Although the flexibility of such agreements varied, they tended to lead to unilateral decision-making on issues such as technological change, the employment of casual workers and trabajadores de confianza [employees in positions of trust, who are not permitted to join the same trade union as other employees], and worker mobility (De La Garza 2000, 204–205).

Although employer protection agreements were commonly used in Mexico and there is evidence that were already in use in the 1930s, they remained limited to the construction and fashion sectors until at least the late 1980s (Bensusán 2007a). Such agreements became a valuable tool for expanding the new economy linked to productive restructuring and the export model, however. Although through this institutional arrangement trade unions were formally granted a considerable range of coercive powers during bargaining, in practice the agreements made it possible for employers to make many unilateral decisions regarding working conditions. This was the case in the service sector (for example, air carriers, banks and commercial companies in the retail and wholesale sectors), the maquila [offshore processing or assembly plant, often in a free trade zone] export industry in the north of the country and the car parts sector of the automotive industry, which was the most successful branch of Mexico’s export market (Bouzas 2007; Bensusán and Reygadas 2000; Carrillo and Gomis 2013).

# Opening up to trade and the export model: consequences for workers, a literature review

This section will examine the correlation between the institutional deactivation of protections for workers and the process of developing trade openness and economic integration with North America. Productive restructuring in various sectors, such as the automotive industry, and the expansion of the maquila export industry allowed Mexico to begin to establish a new position for itself within the world economy, which was accompanied by a loss of job security and a drop in wages, a trend that accelerated from the late 1980s onward.

According to Bizberg (2015b, 44), Mexico developed towards a model of “capitalism based on international subcontracting”, with a productive apparatus that was “unorganized, as the configuration of the productive structure was decided externally”. Bizberg understands international subcontracting to mean the process through which, on the basis of contracts with multinational companies, Mexico imports goods from Asian countries for assembly and subsequent export, primarily to the United States. This process is characterized by a low level of state intervention in the economy and by minimal coordination between unions and the business sector, primarily owing to the fragmentation and weakness of workers’ organization. Furthermore, according to Bizberg, this type of capitalism coincided with “a system of industrial relations dominated by flexibility and a lingering system of well-being oriented towards assistencialism” (Bizberg 2015b, 44).

The maquila export industry serves as an example of the correlation between this form of integration into the world economy, the loss of sense of protection provided by labour institutions and the quality of jobs. This is supported by the studies carried out by the Northern Border College (COLEF) based on surveys conducted in 1990 and 2000 and, in particular, the studies carried out by Carrillo and Gomis in the late 2000s (2013, 35 and 51–52). These studies demonstrate that the factors leading to the development of the maquila export industry in Mexico included low wage levels, the lack or low quality of trade union representation and the country’s geographical proximity to the United States, as the industry was initially based in the border area.

The researchers also demonstrated that one characteristic of multinational maquila companies (compared with other types of companies) was the high proportion of female employees and the precarious nature of many jobs. More women were employed in US companies, in companies based in the north of Mexico and in smaller companies. Working conditions for employees of maquila companies were worse with regard to wages and to the lack or low level of trade union representation.13 The situation was even worse in companies with a large proportion of female employees. The situation varied from city to city; for example, in Tijuana there were more female employees than in Ciudad Juárez (70 per cent compared with 55 per cent). There were also sector-based differences; in the car parts sector, for example, almost 70 per cent of workers were men, while in the electronics industry more than 75 per cent were women (Carrillo and Gomis 2013, 35). The researchers also observed that, in the mid-2000s, maquila companies began to employ a higher proportion of men, although female employees remained dominant (Carrillo and Gomis 2013, 36).

These findings were corroborated by a research project in which the researchers examined five cases of economic and productive success in various maquila companies located in the state of Chihuahua. The study showed that job and productivity growth under the export model did not lead to higher wages, as had been promised by NAFTA. The main factors behind this were macroeconomic pressures, the restrictive minimum wage policy in place since the 1980s, corporate inertia and the ability of maquila associations to artificially push down workers’ contractual wages. Owing to the lack of truly representative unions capable of negotiating agreements and providing grassroots support, wages were always set unilaterally or through employer protection agreements and were systematically set lower than productivity increases or even inflation. Another factor was the absence of mechanisms to incentivize workers’ participation in production management. As a result of all these factors, even in model businesses that were succeeding in their respective sectors, workers did not see any improvement in their wages or working conditions and did not benefit from the success of the export model. The same study also showed that, between 1981 and 1988, the salaries of manual labourers and technicians dropped while benefits increased slightly, including for employees who, unlike other workers, had received a wage increase (Bensusán and Reygadas 2000, 39 and 53–54).

According to another study on the fashion maquila industry conducted in the mid-2000s, working conditions in companies that manufactured various brands of clothing were characterized by high workforce turnover and geographical mobility, the extensive use of labour, a low level of qualifications among workers, low wages, little-to-no unionization and heightened vulnerability for workers. Jobs in the weakest areas of the supply chain – clandestine workshops and domestic work – were even more precarious (Bensusán 2008).

The restructuring of the automotive industry led to greater integration into the US economy and, in turn, a new technical composition of the labour force, driven by relocation to the north. This process went hand in hand with the expansion of export maquila businesses (such as in the car parts sector) that had not previously existed in Mexico, which encouraged the subcontracting of work and services in the sector. In addition, more women joined the workforce, old labour agreements were made more flexible, and new factories were opened in the north of the country that operated based on collective agreements that met only the minimum requirements established under the Federal Labour Law. On top of all that, wages dropped to levels last seen in the 1980s (Arteaga and Carrillo 1988).

Years later, under NAFTA, the Mexican automotive industry significantly increased the number of jobs provided and became far more attractive to automotive manufacturers. Between 2006 and 2014, Mexico moved from sixth to fourth place in the global list of exporters in this sector. This had a negative impact on wages in the three NAFTA signatory countries, however (Bensusán and Florez 2019).14 Not only did NAFTA fail to narrow the wage gap between the three countries, as had been promised, but in some cases the gap widened. Nevertheless, Mexico remained competitive owing not only to its low wages, but also to its increasing labour productivity; between 2007 and 2018, the employee productivity rate rose from 86.7 to 95.6, while unit labour costs dropped from 106.8 to 99.8.15

The automotive industry was at the very heart of the controversy over NAFTA that arose in the United States and Canada, both of which accused Mexico of social dumping and of causing an artificial drop in wages (Levin 2015). For this reason, and in order to put an end to institutional factors such as employer protection agreements that had prevented wages from increasing in Mexico despite the success of its export model (at that time, only 20 per cent of production was destined for the internal market), specific requirements were introduced during the NAFTA negotiations. In the automotive sector, the Confederation of Mexican Workers held control over almost all trade unions in assembly plants, which had previously conducted negotiations in isolation. Only workers at the Volkswagen and Nissan Cuernavaca plants maintained independent unions and, as expected, managed to negotiate better contractual wages. Furthermore, wages for workers in the car parts sector, which accounts for around 85 per cent of jobs in the automotive sector, are far lower than wages for assembly plant workers (Covarrubias 2014; Bensusán and Gómez 2017; Levin 2015).

# From the North American Agreement on Labor Cooperation (NAALC) to Annex 23-A of the Agreement between the United States of America, the United Mexican States and Canada16

Initial worries regarding the negative impact of labour asymmetries between Mexico, the United States and Canada were assuaged by the signing of the North American Agreement on Labor Cooperation (NAALC) between the three countries in 1993, a parallel agreement to NAFTA, and its entry into force in 1994. While NAALC included 11 important principles, notably freedom of association and the freedom of collective bargaining, it did not set out any plans to put an end to unfair competition from low wages in Mexico. The commitments contained in NAALC focused primarily on cooperation, as suggested by its name. NAALC also remained outside the scope of the dispute resolution mechanisms provided for in NAFTA, which only set out sanctions for the violation of certain technical standards: minimum wages, safety and hygiene in the workplace and child labour. These standards could only be applied through an extensive complaints procedure, for which a panel of experts was never even selected. Beyond preventing social dumping, the real aim of NAALC was to encourage the US Congress to ratify NAFTA, as Democrats were calling for pressure to be placed on Mexico to prevent job losses. However, it soon became evident that, owing to weaknesses in its design, NAALC – like most other global governance trade agreements and institutions – was toothless when it came to guaranteeing its own implementation and preventing social dumping. It therefore did little to prevent wage decreases or improve working conditions in the region (Bensusán 1994; Kay 2011).

Over the course of 25 years, interactions between trade unions and social organizations in the three countries, and even between academic institutions, often outside the scope of NAALC, helped promote the idea that the agreement and the dominant trade union system in Mexico were responsible for the negative trends in working conditions that were been seen in North America.17 Naturally, this was not the only factor behind the drop in wages. Another contributing factor was, undoubtedly, the impact that the Fourth Industrial Revolution had on manufacturing jobs in the United States, where, up until that point, workers had enjoyed the highest salaries. The use of robots and other innovations allowed businesses to increase production while reducing the number of workers. This was compounded by the increasing weakness of trade unions in the United States as a result of both structural and institutional factors (Tourliere 2018; Zepeda 2016). In either case, both the fact that new automotive plants were opening in Mexico whilst others were closing in the United States and the fact that there was a huge wage gap between the two countries undeniably fed interest among trade unions in the United States in making changes to the Mexican labour regulation model, which they would eventually achieve. Conversely, however, attempts to push through legislation to improve protections for individual and collective labour rights for workers in the United States failed on numerous occasions, owing to resistance from Republican senators (Tourliere 2018; Zepeda 2016).

Between 2015 and 2016, during negotiations on the TPP under the Obama Administration, the United States moved away from the position that it had taken in 1994, when it had argued that Mexico should not be required to amend its legislation as it labour laws were more demanding than those in the United States (Bensusán 1994). On the contrary, in order to make headway in the negotiations, conditions were introduced to force Mexico – which was accused of social dumping – to amend the institutional framework of its labour laws in order to protect freedom of association and collective bargaining (Gascón 2015; Puyana 2015). This pressure led to the unexpected reform of article 123 of the Mexican Constitution on 24 February 2017 and the subsequent legislative reform of the Federal Labour Law, the amended version of which was promulgated on 1 May 2019. These reforms will be discussed further in the last section of this working paper. First, however, we shall examine the factors that led negotiators to include a chapter on working conditions within the new trade agreement (the Agreement between the United States of America, the United Mexican States and Canada). This Agreement is unique as it is one of the most highly regulated and extensive commercial agreements on labour issues, thanks to the provisions set out in both chapter 23, which defines the commitments of all three signatory countries, and Annex 23-A on worker representation in collective bargaining in Mexico, which is applicable only to Mexico.18

## Greater demands on Mexico

There are various factors that can explain the change – or radicalization – in demands regarding Mexico’s labour agenda, which were driven initially by the Obama Administration during negotiations on the TPP and later by the Trump Administration during the renegotiation of NAFTA (Levin 2015; Levin and Shaiken 2019). First, there was evidence of low wages and of a drop in labour unit costs in the automotive industry, as discussed above, which were supported by the rise in the use of employer protection agreements. International trade union federations submitted complaints to the International Labour Organization (ILO) regarding the alleged violation of workers’ collective rights in Mexico, as a result of which Mexico was brought before the ILO Committee on Freedom of Association for undermining the right to freedom of association and collective bargaining.19 The United States and Canada argued that employer protection agreements prevented true association and collective bargaining, which the Mexican Government and the trade unions allied with employers were also taking pains to prevent. In 2009, the International Metalworkers’ Federation and other unions, brought a complaint Case No. 2694 before the Committee on Freedom of Association on the grounds of violation of freedom of association and the proliferation of employer protection collective agreements. 20 Furthermore, during the 104th International Labour Conference in June 2015, the Committee on the Application of Standards considered serious complaints made by two major trade union confederations against Mexico on the charges of undermining the principle of freedom of association and of allowing false collective bargaining, as a result of which Mexico was examined by the ILO (Bensusán and Covarrubias 2016). 21 Even international textile industry brands with contractors in Mexico urged the Mexican Government to change the labour laws to prevent trade unions from favouring employers and to prohibit employer protection agreements on the grounds that they violated the ILO Freedom of Association and Protection of the Right to Organise Convention, 1948 (No. 87) (Bensusán and Covarrubias 2016).

As discussed in the previous section, the situation in the automotive industry in Mexico became a cause for alarm, as it was one of the industries that benefited most from NAFTA, with Mexico becoming the fourth largest automotive export market in the world. This appearance of success was relative, however, as a limited number of components were actually produced in Mexico. Moreover, good manufacturing jobs in the United States became less desirable when transferred to Mexico, meaning that Mexican workers did not benefit from free trade. The US Secretary of Commerce was highly critical of this, despite having previously opposed raising the minimum wage in the United States based on the belief that it would accelerate automation. At his confirmation hearing, however, the Secretary stated that not only had there been no gradual convergence between living standards in the United States and Mexico, as had been expected, but that the average Mexican worker was, in terms of purchasing power, “far worse off than he or she was five or 10 years ago”, which “was not the original intent of NAFTA” (Isodore 2017).

The controversy that arose at the Honda plant in Alto Jalisco in 2015, in which an independent trade union disputed a collective agreement during a trial that had started seven years previously, sparked an important solidarity movement among US trade unions. The case also provided undeniable evidence of the fact that the car plant managers were blocking real worker representation during collective bargaining in the sector, which was controlled almost entirely by trade unions affiliated with the Mexican Worker’s Confederation, thereby undermining the principle of bilateralism in negotiations on working conditions. Paradoxically, rather than allowing workers to place pressure on car companies to engage in fair negotiations, the fact that the workers all belong to the same plant was used as an opportunity to push down wages and ensure industrial peace (Bensusán and Covarrubias 2016).

Through the then Secretary of Economy, the Mexican Government attempted – albeit unsuccessfully – to challenge the demands for change in Mexico’s internal labour policy, arguing that the loss of manufacturing jobs in neighbouring countries, and in the most developed countries in general, was driven primarily by technological change and automation. The Secretary argued that trade unions in the United States and Canada were only interested in promoting unionism internationally and were “fighting for something that, at the end of the day, is going to disappear” (Tourliere 2018).

Studies such as those carried out by Acemoglu and Restrepo (2018) and Brynjolfsson and McAfee (2014) provide evidence of the threats posed to manufacturing jobs in Mexico as a result of the Fourth Industrial Revolution, which, to a certain extent, supports the arguments made by the Secretary of Economy. Nonetheless, the closure of plants and the relocation of jobs to Mexico not only provoked concern among trade unions in the United States and Canada, but they also paved the way for the adoption of forceful measures. For example, during the General Motors strike organized in September 2017 by Unifor – the largest private sector trade union in Canada with more than 300,000 members – 3,000 workers opposed plans to relocate part of the production line to Mexico and to lay off 600 employees (Expansión 2017). Unifor was also one of the most active trade unions in Canada in the efforts to ensure that working conditions were discussed during the NAFTA negotiations, a proposal that received the support of the Canadian Prime Minister (United Steelworkers 2017). In the United States, the most recent General Motors strike lasted 40 days and was triggered by, among other things, the closure of one of the company’s US plants and the opening of another plant in Mexico (Juárez 2019).

Another factor that served to further harden each party’s stance on preventing social dumping during the renegotiation of NAFTA was the inability of NAALC to induce any change in workers’ rights in Mexico.22 Although 22 complaints of labour law violations in Mexico had been presented by 2015, out of a total of 40 complaints received under the framework of NAALC, almost none had a notable impact on collective rights and freedom of association in Mexico.23 Nonetheless, one positive result was the removal of requirements stipulating that prospective employees must undergo a pregnancy test before they could be granted a job in a maquila company; this requirement had been the subject of one of the complaints and was later prohibited under the Federal Labour Law (Bensusán and Middlebrook 2013).24 As has been argued by researchers, however, it was not only NAALC that was incapable of properly defending labour rights, but rather all clauses designed to protect such rights in trade agreements had thus far been ineffective owing to, among other things, the lack of compliance mechanisms and the shortage of social actors, such as trade unions and civil society organizations, that were capable of ensuring that such mechanisms were used (Dombois 2006; Kay 2011).

Despite its poor design, NAALC led to greater cooperation between trade unions in North America. Through meetings, they improved visibility and understanding of the regulations and dynamics of the Mexican labour market, which varied greatly from those in Canada and the United States. 25 In doing so, US and Canadian unions, and their governmental and parliamentary allies, were able to identify the mechanisms that were undermining labour rights in Mexico. It became clear that, in order to avoid unfair competition, Mexico needed to do more than simply have stronger legislation to protect individual and collective rights than its neighbour, sign over 70 ILO conventions – compared with the mere ten signed by the United States – or provide, at least in theory, legal facilities for establishing trade unions, negotiating collective agreements or exercising the right to indefinite strike with total suspension of work. The massive gap between law and practice in Mexico was partly the result of structural features left over from the old labour model. During negotiations on the TPP and NAFTA, some of these features were called into question: the tripartite structure of the National Commission on Minimum Wages and the conciliation and arbitration boards, both of which were dependent on the executive branch; the proliferation of false trade unions and false collective bargaining under employer protection agreements; and the weak rule of law regarding labour issues. These features were the target of various criticisms, which ultimately led to important institutional reforms.

In January 2018, for example, more than 180 US Congress members sent a letter to the US Trade Representative, in which they argued that “any new NAFTA must have strong, clear and binding provisions that address Mexico’s labor conditions” (Muñoz 2018). Two important members of the Subcommittee on Trade of the Ways and Means Committee, Congress members Bill Pascrell and Sander Levin, led the response, which had been prompted by statements issued by the Mexican Secretariat of Labour and Social Security questioning the importance of discussions on labour issues during the renegotiation of NAFTA. To avoid any doubt as to which aspects needed to be modified, the letter states that: “Low wages, a lack of independent unions and an inability for workers to collectively bargain in Mexico have hurt American workers and led to the outsourcing of jobs to Mexico” (ReformaLaboralMX 2018).26

## Political change in the United States and Mexico

In addition to the issues already discussed, it is worth noting that the political context in North America at the time of the NAFTA renegotiations was radically different to that in 1994. In line with his election promises, President Trump was seeking to halt investment in Mexico to prevent jobs migrating there, even if the act of keeping jobs in the United States would encourage industrial automation as a way of avoiding paying high wages. Consequently, his priority for both the domestic agenda and the NAFTA renegotiations was, in addition to measures such as fiscal reform, to limit Mexico’s ability to attract investment by offering lower wages than its commercial partners and by allowing false trade unions. Paradoxically, this objective aligned with the demands and interests of US trade unions, which felt that such efforts would respond to their previous calls for an end to social dumping caused by Mexican exports owing to Mexico’s comparative wage advantage.28

## Chapter 23 and Annex 23-A of the Agreement between the United States of America, the United Mexican States and Canada

During the renegotiation of NAFTA between 2018 and 2019, it quickly became apparent not only that the loss of jobs to Mexico was not merely limited to low quality jobs, but also that the only way of stopping the phenomenon was to make regional trade conditional on profound socio-occupational change in Mexico. While chapter 23 of the Agreement between the United States of America, the United Mexican States and Canada sets out the labour commitments of the three participating countries, Annex 23-A refers unilaterally to “worker representation in collective bargaining in Mexico” and stipulates expressly that Mexico must adopt labour laws to implement its constitutional reforms (article 123 of the Constitution) before the Agreement can be brought into force.

Enrique Peña Nieto was President of Mexico at the moment when the Agreement was ratified, and it therefore fell to him to ensure that those conditions were met, even as his presidential term drew to a close. Nonetheless, the main factor responsible for reviving the negotiations that would ultimately lead to the signing of the Agreement was the presidential election victory of opposition candidate Andrés Manuel López Obrador in June 2018. Following his election, Mexico adopted a labour policy that provided for instruments capable of ensuring that labour laws were upheld and that the principles of the constitutional reform enshrined in the Federal Labour Law (promulgated on 1 May 2019) were respected.30 Other measures were also adopted around the same time, including a change in the rules of origin in the automotive industry and the adoption of a minimum wage policy tied to regional production costs, which had never before been covered by a trade agreement. As a result of this policy and other measures, passenger vehicles imported from one country in the region to another could only be eligible for tax relief if 40 per cent of the total manufacturing cost of the vehicle (rising to 45 per cent for vans) were produced by workers earning at least US16 per hour (Morales 2019). The demands made by the United States and Canada aligned with the vision set out by President López Obrador during his electoral campaign regarding freedom of association, collective bargaining and wage reactivation in Mexico. The eminently political introduction to Annex 23-A makes this clear, perhaps as a way of dispelling any lingering doubts about the unilateral nature of the provisions included in the Agreement.31 Chapter 23 on the labour commitments of the three signatory States refers to the 1998 ILO Declaration on Fundamental Principles and Rights at Work and its Follow-up.32 After signing the Agreement on 30 November 2018, the three signatory countries returned to negotiations in order to establish new guarantees to ensure that the Mexican labour model was reformed. To resolve the last remaining points of disagreement and secure the ratification of the Agreement by the United States (January 2020) and Canada (March 2020), on 22 December 2019 a protocol was adopted which set out a new mechanism for the creation of two panels – one panel comprising representatives of Mexico and the United States, and the other representatives of Mexico and Canada – to examine potential violations of trade union freedoms and democracy. These panels are able to conduct on-site inspections and, if three or more incidents of such violations are confirmed to have occurred, a trade embargo may be imposed on exports of products involved in the violations.33 The definition of labour rights violations set out in chapter 23 of the Agreement applies in such cases, which states that the failure to comply with such obligations “must be in a manner affecting trade or investment between the Parties.” A failure is defined as such if it involves “(i) a person or industry that produces a good or supplies a service traded between the Parties or has an investment in the territory of the Party that has failed to comply with this obligation; or (ii) a person or industry that produces a good or supplies a service that competes in the territory of a Party with a good or a service of another Party.” Furthermore, to give weight to these provisions and to the authority of the arbitration panels, the Agreement reverses the burden of proof, stipulating that: “For purposes of dispute settlement, a panel shall presume that a failure is in a manner affecting trade or investment between the Parties, unless the responding Party demonstrates otherwise.”34 In summary, chapter 23 and Annex 23-A of the Agreement establish a strong link between regional trade, workers’ rights in the region and the implementation of Mexican labour reforms. Although they stop short of being fully adequate, these provisions are intended to create the conditions required to induce a transformation in the competitive strategy used by Mexico, which has thus far been based on a combination of low salaries and increasing productivity in export sectors. # The invention and genesis of the new labour model: implications and perspectives The basic principles of the 2017 constitutional reform in Mexico drew on important precedents that favoured the adoption of a new impartial system of labour justice independent of the executive branch and the creation of an autonomous authority for the registration of trade unions and collective agreements. Although various political parties and social and academic organizations had presented proposals for such reforms since the 1990s,35 it was only as a result of pressure from the United States in the framework of the TPP that the constitutional reform of article 123 was eventually approved on 24 February 2017. This significant reform was achieved within a mere 10 months, reflecting the influence that President Peña Nieto had at that time over the legislature, which barely discussed the proposal. It is clear that, without such external pressure, Mexico would never have developed the political conditions required to allow the reform to take place, even if, as we will see, these conditions did not last long. The primary objective of the new model was to reduce state corporatism and to use outside pressure to ensure that the labour reform was implemented effectively. To that end, the model guaranteed the autonomy of trade unions from the State and employers and devolved trade union power to the grassroots level. This would allow for greater distributive negotiation, which would help to link wages to productivity increases. Another key deviation from the previous model was the elimination of the tripartite conciliation and arbitration boards, which, as mentioned earlier, were dependent on the executive branch. The objective of this aspect of the reform was to place labour justice under the oversight of the judiciary in order to strengthen the rule of law in the workplace. In addition to ratifying the ILO Right to Organise and Collective Bargaining Convention, 1949 (No. 98) in September 2018, the Government introduced highly important reforms to protect the rights of the 2.3 million domestic workers in Mexico. This reform was one of a series of actions taken by the legislature to regulate workers’ rights in complement to the reform of the Federal Labour Law in May 2019, with the aim of translating the new constitutional provisions into legislation. With regard to gender, the only change included in the reform of the Federal Labour Law was the requirement that the gender balance on the executive board of a trade union should reflect the balance among the union’s membership. Trade union statutes therefore had to be amended to take account of the new provisions on gender, on universal, free, direct and secret ballots and on transparency and accountability before April 2020 (240 working days from the entry into force of the Federal Labour Law reform, in accordance with the transitory article 23rd of the labour reform).36 The 2017 and 2019 reforms were crucial during the renegotiation of NAFTA, as the submission of the draft agreement to the US and Canadian congresses was conditional on the effective implementation of those reforms. Before the Agreement was ratified, doubts had been raised regarding Mexico’s commitment to implementing its new labour rules; however, the ostensible increase in the 2020 budgets of the Secretariat of Labour and Social Security and the judiciary, which are responsible for ensuring the application of the labour justice system, was seen as an indicator of the Mexican Government’s commitment in that regard (Martínez 2019).37 Moreover, as mentioned above, Mexico – together with its trade partners – also adopted a protocol amending the draft agreement in order to strengthen its commitments under Annex 23-A38 and its domestic legislation, as a result of which the US and Canadian legislatures approved the agreement.39 We will now examine the key aspects of the reform and their potential impact on the transformation of the trade union system that has dominated Mexico for almost a century, a transformation that is hoped will strengthen the bargaining position of workers during negotiations with employers. I. The principles of freedom of association and collective bargaining: Articles 357, 357 bis and 358 of the amended Federal Labour Law embody the principles set out in ILO Conventions No. 87 and 9840 on freedom of association and the right to collective bargaining, which is essential if these articles are to be implemented effectively. Autonomy from employers is a central element of Mexico’s new labour model, which includes the right of worker organizations to protection in the event that an employer attempts to interfere in trade union activities or manipulate collective rights. As mentioned above, the 2012 reform of the Federal Labour Law had already made it illegal to exclude workers from a collective agreement on the grounds that they had been expelled from or had withdrawn from the trade union. II. Protection of the democratic rights of union members (art. 358 of the Federal Labour Law): This includes: freedom of affiliation and dissociation; participation in trade unions and procedures for the election of executive board members via a personal, free, direct and secret ballot; gender equality norms; limits on the length of executive board terms (prohibiting indefinite election); and the obligation to provide members with visibility over the union’s accounts. Most importantly, the workers must approve the re-election of executive board members through a secret, personal and direct ballot, and executive board members cannot remain in the position indefinitely, as each trade union must specify in its statute the number of times that a member may be re-elected (art. 371, paragraph X, read in conjunction with art. 358, paragraph II). III. Free choice of radius of action : Trade unions are expository in nature, which means that workers must decide how to organize themselves as best suits their interests. Consequently, structures that no longer suit the characteristics of a globalized economy based on supply chains may be transformed. Another important aspect is that, by removing the requirement that workers must be active, the new legislation allows unregistered workers or non-typical registered workers (such as those involved in the digital economy, or even informal workers) to form a union. How well these opportunities are used will, undoubtedly, depend on the ability of trade unions to adopt strategies to help them gain credibility and increase their presence among the most vulnerable groups of workers and among groups working in emerging digital markets. IV. Trade union autonomy from the Government: Together with trade union democracy, this is one of the pillars of the transformation of the Mexican labour model. In article 364 bis, the Federal Labour Law establishes a new procedure for the registration of trade unions based on the principles of autonomy, equality, democracy, legality, transparency, certainty, immediacy, impartiality, respect for freedom of association and the guarantees of association, and non-imposition of fees. It also sets a ten-day deadline for labour authorities to respond to registration applications, and it grants precedence to the will of the workers and the interests of the group with regard to all formalities. In addition, the Law provides for the creation of the Federal Centre for Labour Conciliation and Registration, a decentralized agency of the Secretariat of Labour and Social Security, with national competency for registering trade unions. The Law also sets out guarantees to strengthen trade union autonomy from the Government by limiting the use of discretion during registration. In that connection, two types of procedure for verifying executive board elections have been established: the first involves voluntary verification at the request of the executive board members or of 30 per cent of members; and the second involves compulsory verification at the instruction of the relevant authority or through ex officio intervention where there is reasonable doubt regarding the documentation provided (Alcalde, Villarreal and Narcia 2019). These two methods of electoral monitoring will help prevent violations of the political rights of trade union members, thereby ensuring the effectiveness of trade union democracy. V. Guarantees of trade union democracy: The amended Federal Labour Law contains various guarantees to protect the exercise of individual freedoms. It sets out rules and procedures for the expulsion of trade union members, and it specifies that assemblies must be convened and that executive board members and section representatives must be elected via a direct, free, personal and secret ballot. The Law also incorporates criteria on gender-proportional representation. In addition, it sets limits on the maximum term that may be served by an executive board member or section representative and on the number of times that the governing bodies may be re-elected, where required by the workers’ assembly, which shall retain full voting rights. Furthermore, the Law lays down rules and procedures regarding accountability and the consultation of workers during the approval of collective agreements, which may succeed in bringing an end to employer protection agreements. Moreover, article 378 of the Law prohibits trade unions from simulating employers in order to manipulate trade union activities or labour relations or to allow employers to avoid certain responsibilities. VI. Accountability and transparency: The rules on accountability are designed to promote transparency within trade unions and provide workers with a proper understanding of how their union is run. In addition to introducing procedures to ensure that the principle of transparency is upheld, workers may now negotiate with their employer to suspend the payment of their union dues if they believe there to be irregularities in the management of the union’s funds. As this could serve as a double-edged sword by cutting off funding for the most combative trade unions, guarantees should be put in place to prevent this mechanism from being abused, for example by stipulating that all beneficiaries of a collective agreement must pay dues to cover the administration of the agreement. VII. Evidence of right to represent and legitimation of collective agreements: One of the most importance aspects of the reform is that trade unions must obtain “evidence of the right to represent” (proving that the union is supported by at least 30 per cent of workers who would be covered by the collective agreement) that would entitle them to negotiate collective agreements. The Law also specifies that a collective agreement must be approved by a majority of the workers that it covers. As the provision of evidence of the right to represent (which is regulated by article 390 bis) is the most effective way of ensuring true bilateralism in negotiations on working conditions and, especially, on wages, the Law acknowledges that the expression of worker support for the union through a free, personal and secret ballot is “a matter of public order and social interest”. This process is therefore an essential tool in efforts to put an end to employer protection agreements. Notably, the reform also stipulates that this right cannot be denied merely on the grounds that a place of work operates informally. Where two or more trade unions present evidence of having obtained the support of 30 per cent of the workers who would be covered by the collective agreement, a vote will be conducted among the workers to identify which union has the most support and is therefore entitled to sign the collective agreement. While the requirement to obtain this evidence could hinder collective bargaining, without this or similar mechanisms employers would remain able to select their own negotiating partners. In addition to the above, the transitory article 11th of the 2019 Federal Labour Law reform drecree, which refers to article 390 ter of the Law, include for the first time in Mexican history, an authentic process to verify the will of the workers themselves as a means of legitimizing existing collective bargaining arrangements. This process must be introduced within four years of the promulgation of the reform.41 If a collective agreement under negotiation does not receive the support of a majority of workers and cannot, therefore, be validated, it will be annulled, and the benefits and working conditions provided for in the agreement will remain in place until a new agreement is signed. In short, through these principles and rules, the labour reform established a new balance between collective rights, individual freedoms and worker autonomy that may lead to the thorough democratization of relations between the State, trade unions, workers and employers. Moreover, if these procedures and rules are followed in practice, the bargaining power of workers will increase. To ensure that these changes were implemented, deadlines were included as transitory articles in the Law to ensure that trade unions embraced democratic principles and brought their statutes into line with the rules on executive board elections. Trade unions had 240 days from the entry into force of the reform (a period that has already passed) to comply with the new rules and to provide workers with a free, direct and secret ballot during the election of executive board members. Furthermore, trade unions have one year from 2 May 2020 to amend their procedures for soliciting the approval of workers regarding collective agreements (Nos. 22nd and 23rd of the transitory articles of the 2019 Federal Labour Law reform). This should protect workers from any attempts to delay the introduction of democratic measures within trade unions. Another labour-related issue raised by the United States and Canada during the negotiations that was not addressed by the 2017 or 2019 reforms was the need to prevent companies in Mexico from increasing their subcontracting practices, as such practices fragment working communities and make jobs more precarious. In Mexico, this phenomenon had already been discussed during the 2012 reform of the Federal Labour Law, and it appeared again on the legislative agenda in 2019.42 Important restrictions were placed on companies that subcontract work, with the primary aim of limiting the form of outsourcing known as “contract labour”, a legal remedy used by some companies to avoid paying tax and to bypass their responsibilities as employers. The ILO (1997) defines contract labour as the provision of a workforce, in which the contractor, acting as an employer, manages and supervises the contracted workers for the benefit of the contracting company, which has no obligations as an employer. 43 The Federal Labour Law stipulates that companies cannot outsource all of the jobs that they provide and that only specialist work may be outsourced. It also states that contract labour cannot be used to undermine the labour rights of individuals working for the contracting company and that the responsibilities of both the contracting company and the contractor must be fixed (arts. 15-A to D). Nonetheless, as these new regulations were never implemented under the previous government (2012–2018), subcontracting practices have increased in some sectors (such as the service sector and manufacturing, in particular the car parts industry), which has not only jeopardized job security but has also encouraged tax evasion and non-payment of social security obligations (Castillo 2018). To stop the abuse of subcontracting arrangements, the new labour authorities, under the coordination of the Secretariat of Labour and Social Security, the Tax Administration Service and the Mexican Social Security Institute, have introduced an oversight mechanism for large enterprises that will cover more than 30,000 workplaces (Secretariat of Labour and Social Security 2019). Two initiatives have also been submitted to the Mexican Congress for approval, one supported by the House of Representatives’ Labour and Social Security Commission and the other by the Senate’s Labour and Social Security Commission and Legislative Studies Commission, with the aim of expanding the restrictions on labour subcontracting, to the point of prohibiting enterprises from subcontracting work in their principal or main area of activity. These proposals will protect workers’ labour rights (seniority, profit sharing, etc.), while permitting enterprises to outsource (as contract labour) only specific projects or specialized or one-off tasks. The proposals also class all prohibited forms of false trade unionism as organized crime (Palacios 2019). # Conclusion In this paper, we have examined how, as a result of its basic design and the fragility of the rule of law, the same institutional framework was able to function in Mexico under various different economic models and political systems without undergoing any formal modifications. We discussed the advantages and disadvantages of this model for workers throughout its lifespan. This model, which allowed for significant job flexibility to the detriment of workers, was particularly beneficial to Mexico as the economy began to open up to trade from 1985, as well as during the introduction of anti-inflation policies and during the country’s integration into the North American economy from 1994, especially as these policies were accompanied by the adoption of wage restrictions that gave Mexico one of its main competitive advantages. Both NAALC and NAFTA came into force in 1994, but wages for workers in the region were inevitably pushed down. Moreover, Mexican workers did not benefit from the boom in sectors such as the automotive industry, which experienced significant productivity increases that were not reflected in wages. Conversely, the main beneficiaries of this situation were multinational enterprises. Workers had to wait 25 years for this to change. As a result of external pressure to change its labour regulations, Mexico began a radical transformation of its labour model that brought it closer to achieving full trade union freedom and democracy. The recently ratified Agreement between the United States of America, the United Mexican States and Canada contains more provisions on working conditions that almost any other known trade agreement. The demands placed on Mexico were not only designed to force it to reform its domestic legislation, but also to address the primary problem made apparent by that the labour clauses contained in such agreements: the lack of mechanisms capable of ensuring each agreement’s efficacy. This latest trade agreement is the first of its kind to include provisions specifically to regulate wages in the automotive industry, which even large-scale regional integration projects such as the European Union had never before considering introducing. Following the 2017 and 2019 reforms, for the first time in a century the Mexican labour model now provides a favourable playing ground for the free exercise of collective rights, which workers are finally enjoying after so long under the control of trade union leaders. Given the pressure on Mexico to guarantee the application of its new legislation, in addition to the introduction of short-term panels authorized to examine potential violations of collective rights, both in Mexico and in the other two signatory countries, and to impose sanctions in the event of recurrent violations, it seems likely that greater success will be had in implementing chapter 23 and Annex 23-A of the new trade agreement than was ever had with implementing NAALC. Although this may be true, Mexico is still faced with the enormous challenge of implementing the labour reform in the near future, the success of which will depend on various factors. First and foremost, Mexico will have to overcome resistance from parties that benefit under the current status quo (including traditional trade unions and companies used to the advantage of unilaterality). It also must not be forgotten that trade unions have lost considerable credibility and prestige over the decades as a result of repeatedly betraying the interests of the workers that they were supposed to represent. For this reason and others, it is questionable whether the changes to the Mexican labour model will be enough, however essential, inspired and well-designed they may be. Or, in other words: Has this transformation come in time not only to improve conditions for Mexican workers, but also to prevent a general decline in working conditions in North America? Is it now time for trade unions in the United States and Canada to turn their attention to the domestic institutional frameworks and public policies that have allowed working conditions to deteriorate in their own countries? # References Acemoglu, Daron, and Pascual Restrepo. 2018. “The Race between Man and Machine: Implications of Technology for Growth, Factor Shares, and Employment.” American Economic Review 108 (6): 1488–1542. https://doi.org/10.1257/aer.20160696. Aguirre Botello, Manuel. 2019. “Evolución del salario mínimo en Mexico de 1935 a 2018: Comparado en función del crecimiento de la Inflación”, available at: http://www.mexicomaxico.org/Voto/SalMinInf.htm. Alcalde, Arturo, Alma Ruby Villarreal, and Eugenio Narcia. 2019. Ley Federal del Trabajo: Reforma 2019 comentada. Mexico: Editorial Porrúa. Arteaga, Arnulfo, and Jorge Carrillo. 1988. “Automóvil: Hacia la flexibilidad productiva.” El Cotidiano 21: 79–88. Bensusán, Graciela. 2019a. “Determinantes institucionales y organizacionales del salario en la industria automotriz de México.” In ¿La Industria automotriz en México ofrece salarios dignos?, edited by Humberto García, Jorge Carrillo and Graciela Bensusán. Mexico: El Colegio de la Frontera Norte, Red Itiam. Bensusán, Graciela. 2019b. “Empleos bajo presión: con o sin TLCAN.” In La renegociación del Tratado de Libre Comercio en Norteamérica: Desafíos para México y sus Regiones, edited by Oscar Fernando Contreras, Gustavo Vega, Eduardo Mendoza, Clemente Ruiz Durán and Jorge Carrillo. Baja California: El Colegio de la Frontera Norte. Bensusán, Graciela. 2019c. La subcontratación laboral en México. Mexico: El Colegio de México (MIMEO). Bensusán, Graciela. 2013. Reforma laboral, desarrollo incluyente e igualdad en México. ECLAC (Economic Commission for Latin America). Bensusán, Graciela, Jorge Carrillo, and Ívico Ahumada Lobo. 2011. “¿Es el sistema nacional de relaciones laborales en México un obstáculo o una ventaja para la competitividad de las CMNs?” Revista Latinoamericana de Estudos do Trabalho 16 (25): 121–154. Bensusán, Graciela. 2008. “Entre lo público y lo privado: Las vías de la protección laboral en la industria maquiladora de prendas de vestir en México.” Veredas: Revista del pensamiento sociológico 16 (first semester): 7–62. Bensusán, Graciela. 2007a. “Los determinantes institucionales de los contratos de protección.” In Contratación colectiva de protección en México: Informe a la Organización Regional Interamericana de Trabajadores (ORIT), edited by Alfonso Bouzas. 11–48. Mexico: Universidad Nacional Autónoma de México-Instituto de Investigaciones Económicas. Bensusán, Graciela. 2007b. “La subcontratación laboral y sus consecuencias: ¿problemas de diseño institucional o de implementación?” In Propuestas para una reforma laboral democrática, edited by Alfonso Bouzas. 129–159. Mexico: Universidad Nacional Autónoma de México. Bensusán, Graciela. 2000. El modelo mexicano de regulación laboral. Mexico: FES/Plaza and Valdés/FLACSO/UAM-X. Bensusán, Graciela. 1987. “Trabajo y Sindicatos: Crisis e instituciones laborales.” In 17 Ángulos de un sexenio, edited by Germán Pérez and Samuel León. 247–284. Mexico: Plaza y Janés/Plaza y Valdés. Bensusán, Graciela, and Nelson Florez. 2019. “Los empleos en la industria manufacturera de exportación y el TLCAN.” In Trabajo y Desigualdades, edited by Carlos Alba. Mexico: El Colegio de México (MIMEO). Bensusán, Graciela, and Willebaldo Gómez. 2017. “Volkswagen de México: Un caso de articulación positiva y fuerte pero inestable entre innovación productiva y social.” In ¿Es posible innovar y mejorar laboralmente? Estudio de trayectorias de empresas multinacionales en México, edited by Jorge Carrillo, Graciela Bensusán and Jordy Micheli. 183–238. Mexico: Universidad Autónoma Metropolitana-Azcapotzalco. Bensusán, Graciela, and Alejandro Covarrubias. 2016. “Relaciones laborales y salariales en la IAM: ¿Vendrá el cambio de fuera?” In La Industria Automotriz en México: Relaciones de empleo, culturas organizacionales y factores psicosociales, edited by Alejandro Covarrubias, Arnulfo Arteaga, Graciela Bensusán and Sergio Sandoval. Mexico: Red Innovación y Trabajo en la Industria Automotriz Mexicana (ITIAM), AM Editores-Clave Editorial, CIAD (Food and Development Research Centre), El Colegio de Sonora. Bensusán, Graciela, and Arturo Alcalde. 2013. “El sistema de justicia laboral en México: Situación actual y perspectivas.” Análisis 1. Mexico: Fundación Friedrich Ebert Stiftung. Bensusán, Graciela, and Kevin Middlebrook. 2013. Sindicatos y política en México: cambios, continuidades y contradicciones. Translated by Lucrecia Orensanz. Mexico: Flacso-México/UAM-Xochimilco/Clacso. Bensusán, Graciela, and Luis Reygadas. 2000. “Relaciones laborales en Chihuahua: un caso de abatimiento artificial de los salarios.” Revista Mexicana de Sociología 62 (2): 29–57. Bizberg, Ilán. 2015a. “México: una transición fallida”. Desacatos 48: 122–139. Bizberg, Ilán. (ed.). 2015b. Variedades del capitalismo en América Latina: los casos de México, Brasil, Argentina y Chile. Mexico: El Colegio de México. Blanke, Svenja. 2007. “El sindicalismo mexicano: entre la marginación y la recomposición”. Nueva Sociedad 211. Blecker, Robert A., Juan Carlos Moreno-Brid, and Isabel Salat. 2018. “La renegociación del TLCAN: un enfoque alternativo para la convergencia y la prosperidad compartida”. Economía Informa 408. Bolívar, Rosendo. 2013. “Alternancia política y transición a la democracia en México”. Revista Legislativa de Estudios Sociales y de Opinión Pública 6 (12): 33–53. Bouzas, Alfonso. (ed.). 2007. Contratación colectiva de protección en México: Informe a la Organización Regional Interamericana de Trabajadores (ORIT). Mexico: Universidad Nacional Autónoma de México-Instituto de Investigaciones Económicas. Brynjolfsson, Erik, and Andrew McAfee. 2014. The Second Machine Age: Work, Progress, and Prosperity in a Time of Brilliant Technologies. New York: W. W. Norton & Company. Carrillo, Jorge, and Redi Gomis. 2013. “El empleo femenino en multinacionales maquiladoras y no maquiladoras de México”. In Género y trabajo en las maquiladoras de México: Nuevos actores en nuevos contextos, edited by María Eugenia De la O. México. 31–54. Mexico City: Publicaciones de la Casa Chata. Castillo, Teresa. 2018. “Los obstáculos de la implementación de las normas de subcontratación en México”. Master’s thesis, Latin American Faculty of Social Sciences, Mexico. Celis, Fernando. 2018. “Las promesas de campaña que Trump sí ha podido cumplir”. Forbes, 19 January. CIDE (Economic Research and Teaching Centre). 2015. “Mesa 2. Justicia Laboral”. In Diálogos por la Justicia Cotidiana: Diagnósticos conjuntos y soluciones. 35–57. Compa, Lance, and Tequila Brooks. 2014. The North American Free Trade Agreement (NAFTA), North American Agreement on Labor Cooperation (NAALC) and Labor Provisions in other North American Free Trade Agreements. United States: Walters Kluwer. CONEVAL (National Council for the Evaluation of Social Development Policy). 2018. Estudio diagnóstico del derecho al trabajo. Covarrubias, Alejandro. 2014. “Explosión de la Industria Automotriz en México: De sus encadenamientos actuales a su potencial transformación.” Análisis 1. Mexico: Fundación Friedrich Ebert Stiftung. Covarrubias, Alex. 2018. “La ventaja competitiva de México en el TLCAN: un caso de dumping social visto desde la automotriz”. Norteamérica 14 (1): 89–118. http://dx.doi.org/10.22201/cisan.24487228e.2019.1.340. Cruz, Mónica. 2011. “Canadá se lleva el oro de México”. Reuters. De Buen, Néstor. 2013. “Corrupción sindical: un lugar común.” La Jornada, 29 September. De la Garza, Enrique. 2000. “La contratación colectiva”. In Trabajo y trabajadores en el México contemporáneo, edited by Graciela Bensusán and Teresa Rendón. 193–201. Mexico City: Miguel Ángel Porrúa. De la Garza, Enrique. 1992. “Reestructuración productiva, estatal y de los sujetos obreros en México.” In Crisis y sujetos sociales en México: Volumen 1, edited by Enrique de la Garza. 53–106. Mexico: CIIH-UNAM/Miguel Ángel Porrúa. De la Garza, Enrique. 1991. “Cultura y crisis del corporativismo en México”. In Transición a la democracia y reforma del Estado en México, edited by José Luis Barros, Javier Hurtado and Germán Pérez. 235–268. Mexico: Universidad de Guadalajara/Miguel Ángel Porrúa and FLACSO de México. De la Garza, Enrique. 1988. Ascenso y crisis del estado social autoritario: Estado y acumulación de capital en México (1940-1976). Mexico: Centro de Estudios Sociológicos-El Colegio de México. De la Garza, Mónica. 1993. “Políticas de Ajuste en México 1982-1991”. Revista Indicadores Económicos FEE 21 (2): 171–181. DOF (Official Gazette of Mexico). 2017. “Decreto por el que se declaran las reformas y adicionadas diversas disposiciones de los artículos 107 y 123 de la Constitución Política de los estados Unidos Mexicanos, en materia de Justicia laboral”, available at: http://dof.gob.mx/nota_detalle.php?codigo=5472965&fecha=24/02/2017. DOF (Official Gazette of Mexico). 2019. “Decreto por el que se reforman, adicionan y derogan diversas disposiciones de la Ley Federal del Trabajo, de la Ley Orgánica del Poder Judicial de la Federación, de la Ley Federal de la Defensoría Pública, de la Ley del Instituto del Fondo Nacional de la Vivienda para los Trabajadores y de la Ley del Seguro Social, en materia de Justicia Laboral, Libertad Sindical y Negociación Colectiva”, available at: https://www.dof.gob.mx/nota_detalle.php?codigo=5559130&fecha=01/05/2019. Dombois, Rainer. 2006, “La regulación laboral en los tratados de libre comercio: El caso del Acuerdo de Cooperación Laboral de América del Norte entre México, Canadá y los Estados Unidos.” Foro Internacional XLVI (4): 741–761. El Financiero. 2020. “Siempre sí: Senado de Canadá aprueba ley de implementación del T-MEC.” 13 March. Ermida Uriarte, Oscar, and Colotuzzo, Natalia. 2009. “Descentralización, tercerización, subcontratación”. ILO. Expansión. 2017. “Trabajadores de GM en Canadá se van a huelga por traslado de producción a México.” Expansión, 18 September. Flores, Zenyazen. 2019. “Secretaría del Trabajo entrega la primera constancia de legitimación a sindicato de Cemex.” El Financiero, 7 October. Forbes Staff. 2018. “El TLCAN ha sido un horrible desastre para este país: Trump.” Forbes, 11 May. Gamarro, Urias. 2012. “Se suspende panel arbitral.” Prensa Libre, 15 November. Gándara, Natiana, and Urias Gamarro. 2017. “Guatemala y EE.UU. ponen fin a demanda.” Prensa Libre, 22 June. García Pureco, and Diana Vianey. 2018. “El TLCAN y los bajos salarios en México.” Nexos, 4 September. Gascón, Verónica. 2015. “Obliga TPP a México a adecuar ley laboral.” Reforma, 12 October. Graubart, Jonathan. 2008. Legalizing Transnational Activism: The Struggle to Gain Social Change from NAFTA’s Citizen Petitions. Pennsylvania: Penn State University Press. Guimón Pablo. 2020. “El Senado de EEUU aprueba el acuerdo Comercial con México y Canadá.” El País, 16 January. ILO. 1997. Trabajo en régimen de subcontratación. ILC.95/VI. IndustriALL Global Union. 2010. “Presentada en la OIT una segunda queja sobre los contratos de protección mexicanos”, available at: http://www.industriall-union.org/es/archive/imf/presentada-en-la-oit-una-segunda-queja-sobre-los-contratos-de-proteccion-mexicanos. Instituto Belisario Domínguez. 2018. “Actualización de acuerdos comerciales (TLCAN y TPP11) y asimetrías salariales en América del Norte.” In Temas Estratégicos 58. Isodore, Chris. 2017. “Wilbur Ross wants a higher minimum wage -- in Mexico.” CNN, 3 March. Juárez, Pilar. 2019. “Termina la huelga más larga de General Motors.” Milenio, 26 October. Levin, Sander. 2015. “TPP in Focus: Why Mexico´s Labor Standards Matter for TPP.” Blog of the Ways & Means Committee, US House of Representatives, 10 July. Levin, Sander, and Harley Shaiken. 2019. “How Can Americans Compete With Mexicans Making a Tenth of What They Do?”. The New York Times, 19 November. Linz, Juan. 1975. “Totalitarian and Authoritarian Regimes.” In Handbook of Political Science, Volume 3: Macropolitical Theory, edited by Green Stein and Nelson Polsby. Reading, Massachusetts: Addison-Wesley. Lobosco, Katie, Brian Fung, and Tami Luhby. 2019. “Cinco diferencias entre el TLCAN y el T-MEC.” CNN, 11 December. López Obrador, Andrés Manuel. 2018. 2018 La Salida: Decadencia y renacimiento en México. Mexico: Planeta. Martínez, María Del Pilar. 2020. “T-MEC intensificará supervisión laboral: abogados.” El Economista, 21 February. Martínez, Martha. 2019. “Dan más a la Ley laboral que a educación.” Reforma, 22 November. Merino, Mauricio. 2003. La transición votada: Crítica a la interpretación del cambio político en México. Mexico: Fondo de Cultura Económica. Meyer, Lorenzo. 1977. “La etapa formativa del Estado Mexicano Contemporáneo 1928-1940.” Foro Internacional XVII (4): 7–30. Morales, Roberto. 2019. “Regla salarial en autos pone piso parejo en el T-MEC: EU.” El Economista, 25 April. MORENA (National Regeneration Movement). 2017. National Plan of the National Regeneration Movement 2018-2024. https://contralacorrupcion.mx/trenmaya/assets/plan-nacion.pdf Moreno-Brid, Juan, and Jaime Ros. 2009. Desarrollo y crecimiento en la economía mexicana: una perspectiva histórica. Mexico: Fondo de Cultura Económica. Moreno-Brid, Juan, and Stefanie Garry. 2015. “El salario mínimo en México: en falta con la Constitución mexicana y una aberración en América latina”. In Del salario mínimo al salario digno, edited by Miguel Ángel Mancera. 105–121. Mexico: Mexico City Economic and Social Council. Muñoz, Patricia. 2018. “Derechos de empleados, ‘tema central en el TLCAN’”. La Jornada, 27 February. Nolan, Kimberly. 2014. “Internalización de las normas laborales internacionales en el marco del TLCAN: el caso mexicano.” Norteamérica 9 (1): 111–141. Office of the United States Trade Representative. 2019a. “Agreement between the United States of America, the United Mexican States, and Canada 12/13/19 Text”, available at: https://ustr.gov/trade-agreements/free-trade-agreements/united-states-mexico-canada-agreement/agreement-between. Office of the United States Trade Representative. 2019b. “Protocol of Amendment to the United States – Mexico – Canada Agreement”, available at: https://ustr.gov/trade-agreements/free-trade-agreements/united-states-mexico-canada-agreement/protocol-amendments. Ortega, Ariadana. 2019. “El PEF para 2020 da 1,401.9 mdp para implementar la nueva reforma laboral.” El Economista, 23 November. Ortiz, Oscar, Francisco Forastieri, and Rocío Mejía. 2020. “Apartado Laboral - Protocolo modificatorio al Tratado de Libre Comercio México, Estados Unidos y Canadá.” EY, 13 March. Palacios, Óscar. 2019. “Gómez Urrutia presenta iniciativa para regular la subcontratación.” MVS Noticias, 29 October. Palma, José Gabriel. 2011. Homogeneous Middles vs. Heterogeneous Tails, and the End of the ‘Inverted-U’: The Share of the Rich is What It’s All About. Cambridge: Cambridge Working Papers in Economics. Porter, Eduardo. 2017. “Los sindicatos estadounidenses quieren que el TLCAN sea su aliado… pero hay un problema.” The New York Times, 24 August. Puyana, Alicia. 2014. “El Acuerdo de Asociación Transpacífico (TPP) y la Asociación Transatlántica Sobre Comercio e Inversiones (TTIP) ¿La OTAN para el Siglo XXI?” Presentation at Coloquio sobre el Acuerdo de Asociación Transpacífico (TPP): Sus posibles repercusiones para México, El Colegio de México, 19 May 2014. ReformaLaboralMX. 2018. “DeLauro, Levin y Pascrell exigen al gobierno de Trump terminar el outsourcing.” Reforma Laboral para Todos, 23 January. Regeneración. 2018. “Trudeau insiste a México subir salarios para tener un TLCAN justo”. 8 February. Rendón Corona, Armando. 2005. Sindicalismo corporativo: la crisis terminal. México: Grupo Editorial Miguel Ángel Porrúa, Universidad Autónoma Metropolitana, Cámara de Diputados. Reyna, José Luis. 1978. “La negociación controlada con el movimiento obrero.” In Historia de la Revolución Mexicana. 73–106. Mexico: El Colegio de México. Ruiz Nápoles, Pablo. 2017. “Donald Trump, NAFTA, and Mexico.” In Voices of Mexico 103: 19–23. Schmitter, Philippe. 1992. “¿Continúa el siglo del corporativismo?” In Neocorporativismo I, más allá del Estado y el Mercado, edited by Philippe Schmitter and Gerhard Lehmbruch. 15–67. Mexico City: Alianza Editorial. Secretariat of Economy, Government of Mexico. 2020. “T-MEC entrará en vigor el 1 de julio 2020”, 24 April. Secretariat of Labour and Social Security, Government of Mexico. 2019. “Estrategia de inspección laboral permite combatir subcontratación ilegal.” Bulletin 196/2019, 12 November. Tijerina, Walid. 2017. “Dumping social: México, Estados Unidos, el TLCAN y el discurso del mal vecino.” Nexos, 21 November. Tourliere, Mathieu. 2018. “En renegociación del TLCAN, sindicatos de EE.UU. y Canadá ‘luchan por algo que dejará de existir’: Guajardo.” Proceso, 9 January. United Steelworkers. 2017. “Trudeau Government Must Intervene in Labor Rights Crackdown at Canadian-Owned Mine”, available at: https://www.usw.ca/news/media-centre/articles/2017/trudeau-government-must-intervene-in-labour-rights-crackdown-at-canadian-owned-mine. Vega, Salvador. 2019. “Inmunidad sindical.” In Reporte Índigo, 9 July. Zepeda, Roberto. 2016. “Determinantes del declive sindical en Estados Unidos.” Norteamérica 11 (2): 139–164. http://dx.doi.org/10.20999/nam.2016.b006. # Acknowledgements The author wishes to thank Elizabeth Echeverría Manrique for her valuable contributions and the anonymous reviewers for their suggestions. 1 A labour regulation model can be defined by a series of distinct characteristics: the level of state intervention (the greater the level of state intervention, the less liberal the system), the preferred means or instrument for improving labour conditions (confrontation vs cooperation), the level of inclusion/protection of workers (generalized or selective), the recognized level of protection (higher or lower) and the type of relationship established between the State and trade unions (pluralist or corporate). This concept can be used for both analytical and descriptive purposes and to distinguish labour regulation models from one another. In Mexico, the labour regulation model is statist, non-cooperative, corporate–statist, and inclusive yet selective, and it offers a high level of protection for the period in which it was designed. This model was developed through a legal–political process of institutionalization that culminated in the late 1940s. For further discussion about this concept, the various options for labour regulation models and the models used in countries such as Germany, Canada, the United States, Brazil and Argentina, see Bensusán 2000, 35, paras. 312–366. From here on, we will use either “labour regulation model” or “labour model” to refer to this concept. 2 See DOF (2017). 3 This later became the Comprehensive and Progressive Agreement for Trans-Pacific Partnership (CPTPP). 4 For some authors, the origin of the transition to democracy lies in the result of the 2000 elections, when the electorate chose political change by electing a new political party (National Action Party (Partido de Acción Nacional)) to the presidency for the first time the Institutional Revolutionary Party – originally called the National Revolutionary Party, but renamed the Party of the Mexican Revolution in 1938 and the Institutional Revolutionary Party in 1946 – seized control in 1949. For other authors, this transition began with the end of the Institutional Revolutionary Party’s hegemony following the hard-won 1988 election, when an opposition candidate (albeit one that had emerged from the ranks of the Party) won a large portion of the vote and cast doubt over the certainty of the outcome. Other authors go further back still to the process of political liberalization that began with the electoral reforms of 1977, or even further back to the student movement of 1968. See Bizberg (2015a), Bolívar (2013), Merino (2003) and Méndez de Hoyos (2007). 5 For more information about the tripartite nature of the conciliation and arbitration boards, their dependency on the executive branch and their operational challenges, see Bensusán and Alcalde (2013). 6 For an analysis of the primary labour indicators in the late the 2010s and of the inequality among various groups of workers regarding access to decent work, see CONEVAL (2018). 7 See the 2018–2024 National Plan of the National Regeneration Movement (MORENA) in MORENA (2017) and López Obrador (2018). 8 For information about the process of Mexican development, the various stages and the impact on the labour market, see Moreno-Brid and Ros (2009). 9 During this stage, real wage increases were achieved through political negotiation between the State, trade unions and employers, under the control of the Government, which improved workers’ purchasing power within the limits set by government policy (Reyna 1978). Following successive crises and various deals, however, the opposite trend then emerged: wage control to reduce inflation and, later, to keep wages competitive for exports (Palma 2011). 10 For an estimate of the change in minimum wages and the loss of purchasing power between 1935 and 2018, based on data from the National Commission on Minimum Wages, see Aguirre Botello (2019). For information about changes in minimum wages between 1992 and 2020, see the website of the National Commission on Minimum Wages, https://www.gob.mx/conasami/documentos/tabla-de-salarios-minimos-generales-y-profesionales-por-areas-geograficas. 11 For more information on the industrialization model between the 1940s and the 1970s, the loss of macroeconomic stability, the adjustment process, the structural reforms and the move towards the export model in Mexico, see Moreno-Brid and Ros (2009, 132–270). 12 There are two types of clauses on union-building in the Federal Labour Law: a clause on exclusion based on starting to work for an employer, which is still in force and which makes it compulsory for workers to become affiliated to the union that holds the collective agreement under which they are working; and a clause on exclusion based on separation, which obliges employers to dismiss employees who leave or are expelled from the union. This clause was declared unconstitutional and was removed during the 2012 reform of the Law (Rendón Corona 2005; Bensusán and Middlebrook 2013). 13 Maquila companies benefited from a special fiscal system that allowed them to import, assemble, process and export goods without paying duties on them. The maquila programme was wound down in 2006 in response to the tariff relief offered under NAFTA, whereupon it was replaced by the Manufacturing and Maquila Export Industry (IMMEX) Programme (Carrillo and Gomis 2013, 39). 14 In the terminal automotive industry, the average hourly wage dropped in all three countries, although this trend was most pronounced in Canada, where the wage dropped from US34.09 to US$26.34 per hour between 1994 and 2016. In the same period, wages dropped from US$35.91 to US$28.60 per hour in the United States, and from US$6.65 to US\$3.14 per hour in Mexico. Data collected by the Mexican National Institute of Statistics and Geography (INEGI), the US Bureau of Labor Statistics and Statistics Canada (see Bensusán (2019b)).

15

See Bensusán and Florez (2019), based on the Monthly Manufacturing Industry Survey conducted by INEGI over various years, with base=100 in 2008.

16

Part of this section is taken from Bensusán (2019b).

17

For more information on the unexpected transnationalism emergency experienced by some unions in the United States and on the immigration policy change in the context of NAFTA, in addition to the efforts by trade unions and NGOs in the three countries to work together, see Kay (2011).

18

See the official text of the Agreement published by the Mexican Government in 2019. English versions are available on the websites of the US and Canadian governments. The Agreement will come into force on 1 July 2020 (Secretariat of Economy 2020).

19

See Case No. 2694 (Mexico) of the Committee on Freedom of Association. The complaint was presented in February 2009 by the International Metalworkers’ Federation, supported by the International Trade Union Confederation, the Independent Union of Workers of the Metropolitan Autonomous University, the National Steel and Allied Workers Union, the National Union of Mine, Metal, Steel and Allied Workers of the Mexican Republic and the Union of Telephone Operators of the Mexican Republic. The Committee is still examining the case, according to its most recent report published in October 2018.

20

For information on the complaints presented to the ILO, see Bensusán and Middlebrook (2013, 162–166) and IndustriALL Global Union (2010).

21

In 2015, 2016 and 2018, the Committee on the Application of Standards examined Mexico’s implementation of the provisions on freedom of association and the protection of the right to organize set out in ILO Convention No. 87. At its most recent examination of the case in 2018, the Committee encouraged the Mexican Government to “continue to pursue further legislative action envisaged in the context of the Constitutional reform in continued consultation with the social partners at national level; ensure, in consultation with the social partners, that the secondary legislation required to enact the reforms to the Constitution and federal labour law are in conformity with the Convention; continue to fulfil its existing legal obligation to publish the registration and statutes of trade unions, as well as existing collective agreements; and ensure that trade unions are able to exercise their right to freedom of association in law and practice.”

22

For a detailed analysis of the design flaws and weaknesses of NAALC, see Bensusán (1994).

23

The complaints were examined by the US Department of Labor: see https://www.dol.gov/agencies/ilab/submissions-under-north-american-agreement-labor-cooperation-naalc.

24

For a review of the impact of the complaints submitted under NAALC, see Compa and Brooks (2014).

25

See Graubart (2008) and Nolan (2014).

26

See Covarrubias (2018), Tijerina (2017) and Levin and Shaiken (2019).

27

Letter dated 9 August 2011 from the US Trade Representative Ron Kirk to the Guatemalan Minister of the Economy Luis Velásquez, requesting the establishment of the arbitration panel. See Gamarro (2012). For more on the end of the dispute, see Gándara and Gamarro (2017).

28

For more on President Trump’s electoral promises regarding the renegotiation of NAFTA and his government programme, see Celis (2018). In line with his promises, once President Trump came to power he urged leading car manufacturers to produce more cars in the United States, arguing that “NAFTA has been a horrible, horrible disaster for this country, and we’ll see if we can make it reasonable” (Forbes Staff 2018). For more on President Trump’s agenda during the NAFTA renegotiation, see Ruiz Nápoles (2017) and Blecker, Moreno Brid and Salat (2018). For more on the link between the demands made by President Trump and by US trade unions regarding the need to recover jobs and increase wages in Mexico in order to correct market asymmetries, see Porter (2017), García Pureco (2018) and Instituto Belisario Domínguez (2018).

29

30

DOF (2019).

31

Paragraph 1 of Annex 23-A states: “Mexico shall adopt and maintain the measures set out in paragraph 2, which are necessary for the effective recognition of the right to collective bargaining, given that the Mexican government incoming in December 2018 has confirmed that each of these provisions is within the scope of the mandate provided to the government by the people of Mexico in the elections.”

32

These fundamental principles and rights include the freedom of association and the effective recognition of the right to collective bargaining, the elimination of forced or compulsory labour and the abolition of child labour, among others.

33

The three possible sanctions are the suspension of preferential tariffs, the imposition of fines on the product or service and, in repeat cases, the suspension of exports (Ortiz, Forastieri and Mejía 2020). See also Lobosco, Fung and Luhby (2019) and Martínez (2020).

34

See Chapter 23 of the Agreement in Office of the United States Trade Representative (2019a).

35

One of the immediate precedents of the 2017 constitutional reform was the document drawn up by the Working Group on Everyday Justice, under the coordination of the Economic Research and Teaching Centre (CIDE 2015). Some years earlier, during discussions on the 2012 reform of the Federal Labour Law, the political dimension of the reform (as regards trade unions and collective bargaining) proposed by President Calderón had sparked heated internal debate. Previous initiatives developed by the Institutional Revolutionary Party, as well as the National Action Party’s 2010 initiative, had all skirted around such changes. Conversely, although Calderón’s 2012 initiative reflected the demands of various sectors included in earlier initiatives – such as the initiatives developed by the National Action Party in 1995 and the Party of the Democratic Revolution (Partido de la Revolución Democrática) in 1997, as well as those developed by the latter in conjunction with the National Union of Workers in 2002, 2010 and 2012 – it also included proposals regarding democracy, transparency and accountability in trade unions. These included proposed provisions to ensure the introduction of free, universal and secret ballots in trade union elections, the approval of collective agreements by a majority of beneficiaries, and the right of workers to refuse to pay their membership dues if the union failed to publish its accounts. The reform that was ultimately approved was very limited in scope, however, and failed to make any progress towards the democratization of the trade unions. See Bensusán (2013).

36

As a result of measures adopted in response to the COVID-19 pandemic, this deadline was suspended to provide more time for the amendments to be made.

37

In the 2020 budget, a total of 1.409 billion Mexican pesos was allocated to the Secretariat of Labour and Social Security for the purpose of implementing the labour reform (Ortega 2019).

38

For the English text of Annex 23-A, see Office of the United States Trade Representative (2019a, 2019b).

39

The Agreement was approved by the US Senate on 16 January 2020, with a total of 89 votes to 10, and by the Chamber of Representatives in December 2019 (Guimón 2020). The Senate of Canada approved a law to implement the Agreement on 13 March 2020 (El Financiero 2020).

40

Mexico ratified ILO Convention No. 98 in September 2018 and deposited its instrument of ratification with the ILO on 23 November 2018. Twelve months later, in November 2019, the Convention entered into force in Mexico.

41

Following the publication of the protocol on the legitimation of existing collective labour agreements in the Official Gazette on 31 July 2019, the first certificate of legitimacy for a collective agreement was awarded to the Union of Workers in the Industry of Cement, Lime, Asbestos, Plaster, Containers and Similar and Related Products of the Mexican Republic (CEMEX) in September of the same year. According to data from the Secretariat of Labour and Social Security, two months after the protocol came into force, 50 collective agreements had been submitted for legitimation via its digital platform. CEMEX confirmed that it was in the processes of applying for certificates in eight other states (Flores 2019).

43

There are various interpretations of the concept and forms of subcontracting: see Bensusán (2007b; 2019c) and Ermida and Colotuzzo (2009). For these authors, there are three forms of decentralization and outsourcing: mediation, subcontracting and provision of a workforce. According to the authors, this approach differs from the definition used by the ILO, for which labour subcontracting is equivalent to the provision of a workforce (ILO 1997).