Theorising pension reforms

In the past few years, measures to reduce the (future) financial commitments of public pension systems have been a high-priority item on the political agenda of many governments, even though significant difficulties have hindered the conversion of reform plans into actual legislation. Public pension systems are often cited as being notoriously resistant to change, regardless of the paradigm and methodology used to examine them. Despite having a more favourable demographic situation than Western countries, Latin America is also faced with the prospect of having a more rapid shift in its demographics. While population ageing took 50 to 60 years in Western Europe, it is likely to take 20 to 30 years in Latin America. In addition, this region is characterised by a significant and growing informal sector, in which up to 60 per cent of new jobs are created – depending on the country – and in which workers do not contribute to pension schemes. These elements accentuate Latin American governments’ difficulties in adequately financing the needs of their growing population of older people in a constrained fiscal environment monitored closely by international financial institutions (IFIs) (Bertranou and Rofman, Reference Bertranou and Rofman2002; Brooks, Reference Brooks2002; Weyland, Reference Weyland2002). Thus, even though the social magnitude of the ‘pension problem’ is currently still not as significant in Latin America as it is in Western Europe, its fiscal impact is already very considerable.

This section seeks to underline key contributions to current welfare reform literature while stressing their applicability to Latin America in general, and more specifically to Mexico. We devote particular attention to three of these approaches, respectively focusing on: the effect of democratic structures, the role of socio-political institutions, and social construction.

Strength of democratic structures

Various strands of the literature have underlined that the limited democracies and authoritarian regimes often found in Latin American countries until the 1980s facilitated the passing of unpopular socioeconomic adjustment and restructuring measures (see in particular Haggard and Kaufman, Reference Haggard and Kaufman1992). The logic behind this argument is that citizens of these countries typically feared stark reprisals from authoritarian governments if they dared to organise to oppose such policies publicly.

There is strong support for this argument in the field of pension reform literature, which often uses as its point of reference the pension restructuring that occurred in Chile in 1981, during the Pinochet dictatorship (Huber and Stephens, Reference Huber and Stephens2000; Brooks, Reference Brooks2002). Largely crafted by the notorious ‘Chicago Boys’, this pension reform abolished the existing public system in favour of a fully funded private one. Despite the scope and importance of this reform, no external consultations took place and Pinochet easily disregarded the few criticisms that arose from within his administration.

A nine-country study conducted by Mesa-Lago and Müller (Reference Mesa-Lago and Müller2002) similarly concludes that the degree of democratisation of a political regime represents the most important variable explaining the degree of privatisation of pension systems in Latin America. Indeed, this research concludes that the more democratic a regime, the less likely it is to engage in significant reform seeking to privatise the country's pension system. According to this interpretation, the 1995 Mexican pension reform was made possible by the weakness of opposition groups and the authoritarian nature of Mexican politics at the time. This allowed the PRI, still firmly in control of government despite increasing electoral competition, essentially to bypass opposition to the reform and proceed with large-scale privatisation of the country's pension system (Mesa-Lago and Müller, Reference Mesa-Lago and Müller2002: 698–9).

This theoretical framework assumes that public pension schemes are generally quite difficult to reform under democratic rule for several reasons. First, as a result of population ageing, a growing number of voters are retired or nearing retirement, particularly in industrialised countries. For a government or party, this increases the political risks of advocating pension restructuring, since workers approaching retirement or already retired tend to be apprehensive of modifications to existing pension schemes. This phenomenon is accentuated by the weak political participation of younger citizens, whose electoral weight is often surpassed by that of their elders.

Second, most public pension systems are based on the pay-as-you-go principle, in which the contributions of active workers finance the pension of retirees, with the understanding that the following generation of workers will do the same. Originally, this system was adopted as a means to alleviate poverty among older people without increasing the fiscal burden of the state. The maturity of these strongly path-dependent systems has increased the likely negative socioeconomic impact of privatisation-oriented reform options, as retired workers fear losing their retirement benefits, despite numerous years of contributing to the system originally in place (Pierson, Reference Pierson2000). Under democratic rule, this context also significantly increases the likely political costs for a government proposing or attempting to reform a public pension system.

Thus, based on this literature, we would expect that the culmination of a regime transition to democracy in a country would considerably reduce the probability of subsequent governments moving forward with pension system reforms. In the case of Mexico, this implies that the 2004 pension reform should neither be comprehensive nor favour privatisation of the system in place.

Institutional constraints

If we assume, as most specialists do, that Mexican political institutions have become democratic – at least in procedural terms – with and since the July 2000 general elections, Western-based welfare state reform theories should be applicable to the 2004 changes in Mexico's pension system. In particular, Pierson's (1996) influential research has argued that retrenching the welfare state is far more difficult than expanding it. He demonstrates that social policies have generated powerful organised interests, such as the American Association of Retired People, which defend the continuation of these measures and their benefits. As a result, Pierson argues that politicians typically seek to avoid blame when enacting retrenchment policies, rather than claim credit.

In that context, Pierson affirms that the type of democratic institutions within which politicians operate holds central importance, since it dictates the extent to which they can avoid or deflect blame. On the one hand, a ‘divided’ presidential system – that is, a system in which the President's political party or coalition does not hold an absolute majority in parliament, such as is the case in Mexico – could favour more reformist policy making, since the President can blame Congress (and vice versa) for unpopular social reform measures, thus confusing the electorate as to who is truly responsible for crafting these changes. On the other hand, a system where the democratic political authority is highly centralised – such as the UK's – decreases the likelihood that unpopular social system restructuring will be enacted, since the electorate can easily trace the origin of such reform and punish the party (or parties) deemed responsible for it in the following election. As such, we would expect the actors who supported and/or enacted the 2004 pension reform in Mexico to attempt to avoid blame rather than to claim responsibility and seek credit for this restructuring.

More recently, other scholars working on welfare state retrenchment have been engaged in a debate concerning the impact of organised labour on welfare reforms. This body of research came about in part as a response to Pierson's (1996) study, which dismissed the power resource theory by arguing that labour held negligible political influence during pension restructuring processes.Footnote ¹ In contrast, many specialists in this literature have argued that, within the European context, unions have held substantial political influence over welfare reform (Anderson and Meyer, Reference Anderson and Meyer2003; Béland, Reference Béland2001; Bonoli, Reference Bonoli2000; Natali and Rhodes, Reference Natali and Rhodes2004). This is particularly the case in social insurance (or ‘Bismarckian’) countries, where labour unions are often directly involved in trust fund management. Indeed, as the state-integrated social partners for the administration of the public pension system, unions strongly identified their interests with this scheme and framed ideological property claims over retirement benefits. Hence, this literature affirms that, in the context of welfare policy reform, wide-scale union mobilisation – such as the French strikes of 1995 – highlights the high political costs to government of ignoring organised labour's demands and interests. This background also emphasises the need for government to negotiate with some of the groups opposing reforms, in order to reduce their political impact (Béland and Marier, Reference Béland and Marier2006).

Key elements present in Bismarckian countries, such as the fragmentation of the pension system, are present within the Latin American context and play an important political role. For example, Madrid (Reference Madrid2003) argues that unions have focused their energy on vigorously protecting certain well-defined labour rights and benefits – such as pension schemes – rather than seeking to fight broader political battles that they could not pursue intensively and had little chance of winning. According to Madrid, this line of explanation accounts for the state's actions in the field of labour and pension policies, where it has typically resolved to jettison or dilute significant reform proposals. This framework would also explain why organised labour mounted weak opposition to economic adjustment and restructuring policies implemented by most Latin American governments in the past 25 years.

This research on the political influence of labour over welfare reform appears very relevant to the Mexican case examined here. The majority of unions are affiliated to a labour confederation known as Congreso del Trabajo (Labour Congress, CT). That confederation still has strong ties to the PRI, which ruled continuously over the civil authoritarian regime in place between 1929 and 2000. Under that regime, Mexico was characterised by a form of state corporatism: that is, an institutional framework of interest intermediation in which the power of the state significantly outweighed that of labour (Patroni, Reference Patroni2001; Schmitter, Reference Schmitter, Schmitter and Lehmbruch1979). The corporatist framework helped to legitimise the PRI's power, but limited the policy input and political autonomy of CT-affiliated unions. Also, this framework politically marginalised independent unions: that is, labour organisations that refused to participate in the PRI's corporatist structure (De La Garza, Reference De La Garza Toledo and Middlebrook1991; Middlebrook, Reference Middlebrook1995: 141–5). As a result of this system, CT union representatives played an important role in the administration of public pension schemes.

Based on this theoretical perspective on welfare reform, we would expect that the labour movement had a considerable political impact in resisting the 1995 and 2004 pension reforms in Mexico, given the unions’ presence and active role in the administration of the public pension schemes for private sector workers, which were the target of both reforms. However, we expect this influence to be weaker than European unions because Mexico recently emerged from a state corporatist arrangement. This is in stark contrast to European countries where societal corporatism has developed since World War II (see Schmitter, Reference Schmitter, Schmitter and Lehmbruch1979).

Social construction of welfare retrenchment

Another significant strand of the welfare state retrenchment literature has focused on the way in which politicians ‘frame’ the policy debate to ensure the public support of unpopular measures. Politicians do not present policy options in a neutral fashion. On the contrary, they seek to manipulate the policy agenda and voters’ preferences to muster popular support for their own policy priorities (Lau et al., Reference Lau, Smith and Fiske1991; Campbell, Reference Campbell1998: 394). As demonstrated by Cox's (Reference Cox2001) analysis of welfare retrenchment, governments in both Denmark and the Netherlands were able to carry substantial policy changes regardless of ideological inclination and despite the risks entailed by such a strategy. However, Germany – a similarly politically configured country – failed to replicate the success of the aforementioned cases, due to the use of ‘provocative language’ by actors promoting social programme reforms, who opposed original principles enshrined in the creation of the welfare state (497). According to Cox, the key to success lies in the way in which leading politicians in Denmark and the Netherlands were able to generate among citizens a sense that there existed a pressing need for swift and comprehensive reform. This was accomplished by relating reform proposals to popular national symbols, thus shaping a reform path. Hence, this approach assumes that political elites can construct and drive a policy reform agenda.

However, this literature indicates that the ability of political elites to frame the debate is constrained by pre-existing socioeconomic structures, political institutions, as well as the cultural context ‘within which meanings are embedded’ (Ross, Reference Ross2000: 173). As a result, the framing of social policy reforms is to be seen as a significant conditional variable. Hence, the acceptance or failure of policy initiatives does not depend solely on the delivery of policy proposals. Indeed, a successful policy entrepreneur is one who can identify policy windows by effectively coupling ideas which linger in the ‘policy primeval soup’ with current policy problems, while simultaneously considering the political environment of the time (Kingdon, Reference Kingdon1984).Footnote ²

Within the Mexican context, non-governmental actors – with the possible exception of IFIs – had been inefficient policy entrepreneurs until 2000. This was mainly due to the authoritarian nature of the PRI-led political regime in place until then. Democratisation is likely to change this context. However, as will be demonstrated later, members of civil society who are opposed to governmental proposals still face numerous difficulties, making it highly unlikely that they could become policy entrepreneurs – in the way conceptualised by Kingdon – in the near future. Thus, we expect governmental elites to act as the main policy entrepreneurs, and to be the ones framing social policy reform proposals.

The Mexican pension system: a contributor to inequality

In keeping with the situation found in most Latin American countries, the Mexican pension system is strongly fragmented along occupational lines, resulting in divergent rates of coverage and benefits. Governmental employees at the national and state levels as well as teachers receive benefits from the Instituto de Seguridad y Servicio Social de los Trabajadores del Estado (Institute for the Social Security and Services of State Workers, ISSSTE) or from special systems set up by regional authorities or public companies. On the other hand, until 1997 the majority of workers active in the formal private sector contributed to and received their retirement benefits from a pension system administered by the Instituto Mexicano del Seguro Social (Mexican Institute of Social Security, IMSS). Since 1997, this segment of the labour force contributes to a private pension system co-funded by workers themselves and their employers.

In this context, four key elements are worth emphasising. First, roughly 50 per cent of Mexico's workforce is involved in the ‘informal’ sector and therefore does not have access to any work-related social benefits, including pensions. This segment of the labour force is likely to grow further in the future, since job creation has occurred mainly in this sector over the past few years (INEGI, 2005). Second, even within the formal private sector, many retired workers are not being adequately covered. The vast majority of retired workers (86 per cent) are still covered by the previous public pension system and thus receive only minimal benefits, equivalent to the minimum wage (Espinosa-Vega and Sinha, Reference Espinosa-Vega and Sinha2000). In 2005, this amounted to US $4.30 per day: an insufficient sum to cover an individual's basic needs (INEGI, 2005). Third, despite a ‘forced transition’ to private pensions in 1997, only 40 to 45 per cent of affiliated workers are still contributing. The remaining retirement accounts are deemed inactive, often due to their holders having joined – willingly or not – the informal sector (de Ferranti et al., Reference De Ferranti, Leipziger and Srinivas2002: 2; Claramunt, Reference Claramunt2004). Fourth, public sector workers (affiliated with ISSSTE, IMSS and other state schemes) have their own pension system, which typically provides generous benefits representing at least 100 per cent of their pre-retirement wage – although this has changed for new IMSS workers, as a consequence of the 2004 pension reform. These retirement benefits were expanded in the 1960s and 1970s to ensure the support of governmental workers for the PRI-led regime (Stracke, Reference Stracke2003).

These four elements are mostly regressive and accentuate income inequalities between public and private sector employees, as well as between formal and informal sector workers (Scott, Reference Scott and Wodon2003). For instance, the benefits received by public sector retirees are, on average, ten times what average pensioners in the formal private sector get. Nevertheless, as indicated below, these public pension systems were actuarially bankrupt by 2004, and relied on continuous governmental budgetary inputs for their maintenance. Not surprisingly, since the early 1990s IFIs have repeatedly advised the Mexican government to reform the public sector pension systems. The federal government of Mexico has since then proceeded with two pension restructurings.

The 1995 pension reformFootnote ³

Several factors prompted the PRI federal government of President Zedillo to enact a significant reform of the IMSS pension system in 1995. First, the IMSS had always used the money accumulated from the tripartite worker–employer–state contributions to compensate for recurring deficits in health and maternity programmes. Internal studies estimated that, as a result, the IMSS would start running a deficit by 2004 (IMSS, 2001). Second, Mexico's high rates of inflation during the 1980s and 1990s impacted negatively on the real value of pension payments (Cerda, Reference Cerda1996). Third, many employers and workers in the formal sector adopted an ‘exit strategy’. Indeed, the dissociation between high levels of contributions and low levels of benefits led many workers to participate in the programme only until they reached the mandatory minimum number of contribution weeks (set at 500, under the old system) in order to qualify for benefits. Once this threshold was reached, the benefits accrued as a result of continued contributions over a long period of time increased only marginally, leading many workers to avoid paying IMSS salary deductions either by continuing to work unofficially – that is, without appearing on their employers’ payroll – or by joining the informal sector. Similarly, business owners often failed to declare the real number of workers they employed, in order to reduce their contribution to the IMSS (Cerda, Reference Cerda1996: 2–3; Espinosa-Vega and Sinha, Reference Espinosa-Vega and Sinha2000: 4).

In response to these significant structural problems, the 1995 reform sought to transform the IMSS pension system from the previous pay-as-you-go arrangement to a model characterised by individual retirement accounts with a minimum pension guaranteed by the government, and administered by the private sector (Carstens, Reference Carstens1997; Arvizu Treviño, Reference Arvizu Treviño2002: 5). Investment management companies (known as AFORES) were authorised by the government to charge ‘reasonable’ management fees to their clients, and were supervised and regulated by a state agency: the National Commission for the Saving System for Retirement (CONSAR).

Under the new mandatory system, the contributions of 8.5 per cent of workers’ salaries to the previous IMSS pension plan were separated into two sections: 4.5 per cent would be solely devoted to the new private pension scheme, while the remaining 4 per cent would still be collected by the IMSS in order to fund the Institute's survivorship, health care and disability programmes. Since 1995, employers have also had to supply an extra 2 per cent to the new individual pension accounts. In the case of workers earning minimum wage, the state contributes an additional 5.5 per cent of the official minimum salary to these individuals’ pension funds (Carstens, Reference Carstens1997: 154). Previous thresholds on the minimum number of weeks necessary to qualify for a pension were abolished, augmenting the incentives for their sustained participation in the new scheme.

Following the announcement of the reform project in the spring of 1995, many unions – both official and independent – stated their opposition to the proposed changes. The resistance movement included the Sindicato Nacional de los Trabajadores del Seguro Social (the National Union of Social Security Workers, SNTSS). This was a fairly surprising move, since the SNTSS was a member of the CT and was therefore – to a great extent – under the control of the PRI. But the SNTSS was also directly involved in the administration of the IMSS pension scheme then in place. As a result, the union stood to lose this position in the event of the implementation of the reform. The SNTSS hence felt that the proposed restructuring represented a direct attack on its interests and openly resisted the PRI's project. For its part, the CT originally rejected the project as well, but it eventually obtained selective and limited compromises from the Zedillo government – such as the continuing administration of both health care and housing (INFONAVIT) funds via the IMSS. It should, however, be noted that the essence of the reform remained untouched despite those modifications. This new version of the reform proposal nevertheless satisfied most CT-affiliated union leaders, whereas it infuriated the SNTSS, which was in fact removed by the restructuring from the administration of private workers’ pension funds. In protest, the SNTSS left the CT soon thereafter, and in 1997 joined a newly created independent labour confederation (the UNT), in which it obtained key leadership positions (Mayer, Reference Mayer2003).

The SNTSS and some allied unions carried out demonstrations during the 1995 negotiation process, but did not resort to strikes. This is attributable to several factors. In particular, the pro-government position of the CT's leadership impeded the building of a strong and unified union coalition in opposition to the reform project. Also, the PRI government of Zedillo put tremendous pressure on rebellious unions and granted them very little leeway for expressing their dissent (Mesa-Lago and Müller, Reference Mesa-Lago and Müller2002: 698–9). These factors contributed to the effective political marginalisation of the SNTSS and its allied unions’ opposition movement.

Among the three major political parties, the governing party (the PRI) and a minority of Congress persons from the main rightist opposition party (the PAN) supported the pension restructuring, while the Partido de la Revolución Democratica (Party of the Democratic Revolution, PRD), the main leftist opposition party, as well as a majority of PAN Congresspersons, opposed it. Since the PRI controlled the majority of seats in Congress as well as the Presidency at the time, the reform was easily promulgated. The authoritarian nature of the political regime ensured that the PRI was able to control the discourse promoting the reform project, while muzzling the opposition.

The 2004 pension reform

Despite the 1995 pension system restructuring, the financial health of the IMSS remained precarious. Both governmental reports and IMSS internal documents indicated that this was largely due to the generous pension scheme enjoyed by the Institute's own employees (IMSS, 2003), who were required to complete fewer years of service than, and received benefits well above those received by, retiring workers in other sectors of the economy. Before the 2004 reform, female IMSS workers could retire after completing 27 years of service (28 years for men), which set the average retirement age of IMSS workers at 53. By comparison, Mexico's official retirement age is 65. The pension benefits of retired IMSS employees represented 130 per cent of their final salary, and pensions were indexed when the salaries of active workers were increased (IMSS, 2004). As a result, retired IMSS workers received an average monthly pension of US $1,500 in 2004. In comparison, retiring private sector workers who earned the same salary as IMSS workers at the time of retirement would only receive US $210 per month in pension.

A significant problem with the IMSS worker pension scheme resided in the Institute's practice of using the health care contributions of the 12.5 million private sector workers – and of their 800,000 employers – affiliated to the Institute to cover the expenses resulting from the payment of pensions to retired IMSS workers. According to an IMSS internal analysis, the Institute paid US $1.83 billion in pension and health care expenses for IMSS retirees in 2003. Of this amount, the contributions of IMSS workers represented only US $138.6 million, and the money supplied by the federal government as employer represented US $380 million (IMSS, 2004). To bridge this gap, the Institute used its reserves and channelled resources away from investments in infrastructure and services to affiliated members (Cevallos, Reference Cevallos2004). The IMSS had also accumulated a debt estimated at US $35 billion by 2004.

It is in this context that a reform project was introduced to Congress in July 2004 by the PRI, now an opposition party to the PAN government of Vicente Fox. The reform sought to make it mandatory for new IMSS workers to contribute to a new pension fund, managed by the Institute, to be used only for the purpose of providing retirement benefits for these workers. Indeed, the reform intended to make it illegal for the IMSS to use its reserves or the contributions of affiliated private sector workers and their employers, in order to pay for the specific pensions and benefits of these new IMSS employees (Beltrones, Reference Beltrones Rivera2004). The reform therefore created two categories of IMSS workers. Those employees already in place would remain under the previous retirement and social advantages plan, while the new workers’ benefits would be much less considerable. Under the new scheme, workers contribute at least 10 per cent of their salary (instead of the 3 per cent they previously paid) to their pension fund, retire after 35 years of service (instead of 27 or 28), and receive 100 per cent (instead of 130 per cent) of the value of their final monthly salary in pension payments (MLNA, 2004).

The pension restructuring was promulgated by the Chamber of Deputies only two weeks after its introduction in Congress. It benefited from the support of the parliamentary factions of the PRI (the instigator of the reform, which also held a simple majority of seats in both houses of the Mexican legislature), the PAN (the second most important party in Congress), and the minority parties PVEM (the Green Party of Mexico) and Convergencia (Convergence, a centrist party). Members of the third force in Congress, the PRD, opposed the reform project, and were joined by the small leftist Partido del Trabajo (Labour Party, PT) and a few dissident PRI deputies (Guerrero, Reference Guerrero2004). The Mexican Senate approved the reform a week later, as each party adopted the same position in the Upper House of Congress as they had in the Lower House. The reform project also received the backing of President Fox, who argued that his administration had been concerned with this issue since his election (Aranda, Reference Aranda2004).

Proponents of the reform, which included the CT as well as IMSS administrators, claimed that the principal objective of these changes was to stop the haemorrhage of IMSS financial resources and to ensure that the Institute would properly utilise contributions from workers, employers and the state (Beltrones, Reference Beltrones Rivera2004; Pensamiento and Guerrero, Reference Pensamiento and Guerrero2004). Among other things, reform advocates argued that these monies were currently used to cover the pension requirements of only 120,000 retired IMSS workers, whereas they should rightfully be invested in infrastructure and in the provision of enhanced care to the 45 million beneficiaries of the IMSS services.

As such, supporters framed the reform as an issue of social justice. They insisted that the reform sought to reduce the profound economic gap between a minority of ‘privileged’ public sector workers on the one side, and a majority of disadvantaged private sector workers on the other. Proponents added that pensions and benefits of existing workers and retirees of the IMSS would not be touched by the reform. Nevertheless, new workers would be affected by the changes, which would allow the Institute eventually to return to a position of solvency (Muñoz Rios, Reference Muñoz Rios2004).

Opponents to the 2004 reform included the SNTSS and its allies from the two main independent union confederations – the UNT and the FSM – as well as social movements such as El Barzón and El Campo No Aguanta Más (debtor and peasant organisations). These actors adopted four main lines of argument. First, they insisted that the proposed pension reform was illegal, as the working and retirement conditions of the IMSS employees were detailed both in Mexico's Constitution and in the collective contract tying them to the Institute (Alcalde Justiniani, Reference Alcalde Justiniani2004). Opponents affirmed that the reform would violate the collective contract between the IMSS and the SNTSS, and that this would represent an infringement on constitutionally guaranteed labour rights of IMSS employees, and the Institute's financial obligations towards them (Mural, 2004). Second, the SNTSS and independent unions stated that the proposed reform constituted an attack on the independent labour movement in general, and on the SNTSS specifically, so as to break its cohesion and power as well as significantly reduce the benefits of the Institute's workers (MLNA, 2004).

Third, reform opponents recognised the financial difficulties faced by the IMSS. Nevertheless, they were adamant that those problems were due to the mismanagement of funds by IMSS's administrators, to lack of proper government planning, and to many employers failing to pay their contributions to the IMSS. Opponents claimed that the government and the political parties supportive of the reform were opting for an ‘easy solution’, hence unfairly penalising IMSS workers for the Institute's financial ills (UNT, 2004). Finally, the SNTSS and its independent labour allies claimed that the proposed pension changes were particularly insidious, as they represented an initial step in the process of privatising the remaining public pension schemes: particularly the one administered by the ISSSTE (UNT, 2004).

As such, the SNTSS claimed that it was effectively defending the pension benefits of all public sector workers, and staged important work stoppages on 31 August and 1 September 2004 to give added political weight to their line of argument. Although the SNTSS and its allies could count on the PRD's support in Congress, that party did not hold sufficient political weight to stop the reform process. Also, since most unions of public sector workers were affiliated to the CT, which in turn supported the reform, the claim of opponents that they were speaking for and defending the interests of all public sector workers in resisting changes to the IMSS pension scheme appeared to lack credibility. In fact, polls taken during the last weeks of the debate of the IMSS reform indicated that a majority of citizens accepted the point of view presented by proponents of the reform (see, for instance, Reforma, 2004). As such, the SNTSS and its allied independent labour organisations were unsuccessful in convincing citizens and at extracting any significant compromises from either the government or the majority in Congress. The PRI and the PAN, as well as the Fox administration, refused to accede to opponents’ demands, and Congress rapidly adopted the proposed reform in its entirety.