On the Causes of Social Policy Expansion

Over the past few years, several Latin American countries have begun to expand social policy. This social policy shift has generated a rich literature that explains why some countries have been able to expand social protection and others have not. These accounts emphasize the relevance of several macroexplanatory factors. Some authors emphasize the impact of the “left turn,” significant because for left-wing governments, “social justice has always been a central goal of the left” (Weyland et al. Reference Weyland, Madrid and Hunter2010, 160). In this way, “left governments are ideologically committed to improving social inequalities and expanding social citizenship rights” (Cook and Bazler Reference Cook and Bazler2013, 4). When left-wing parties were relatively strong, social policy development was possible (Anria and Niedzwiecki Reference Anria and Niedzwiecki2016; Huber Reference Huber2011; Huber and Stephens Reference Huber and Stephens2012). A related argument is that the economic bonanza generated by the “commodity boom” that took place until the early 2000s provided left-wing governments with the necessary resources to finance social welfare expansion (Hagopian Reference Hagopian2016; Levitsky and Roberts Reference Levitsky and Roberts2011; Roberts Reference Roberts2008).

Furthermore, in political systems characterized by a high level of electoral competition, parties have more incentive to strive for new votes (Fairfield and Garay Reference Fairfield and Garay2017; Garay Reference Garay2016; Pribble Reference Pribble2013). Put bluntly, “electoral competition is what motivates politicians to support comprehensive reform” (Ewig Reference Ewig2015, 197). Thus, “parties facing intense electoral competition are more likely to pursue universalistic social policy reform than parties that face a weak opposition” (Pribble Reference Pribble2013, 176).

Some institutional and partisan features may translate into social policy continuity or change. Partisan powers, presidential constitutional powers, the executive’s degree of control over the legislature, and some informal institutional arrangements may facilitate or obstruct social policy reform (Castiglioni Reference Castiglioni2012). Policy architectures—that is, a mixture of eligibility, funding, entitlements, delivery, and regulation for outside options—may also shape social policy outcomes (Martínez-Franzoni and Sánchez-Ancochea Reference Martínez-Franzoni and Sánchez-Ancochea2016). Moreover, political party character, defined as the combination of ideology, organizational structure, and predominant linkage mechanism, is crucial, as parties have a fundamental role in shaping and transforming social policy (Pribble Reference Pribble2013).

The role of civil society organizations and mobilization has also been considered an important factor in expansion. Garay (Reference Garay2016) argues that expansion occurred in countries that experienced large-scale social mobilization from below. Case study research conducted in different Latin American countries has also suggested that social movements were behind the expansion of social policies in recent years (Anria and Niedzwiecki Reference Anria and Niedzwiecki2016; Silva Reference Silva2015).

Relying on these macroexplanatory factors, the comparative social policy literature has been able to offer several compelling explanations for why and how social policy expansion occurred across countries. In contrast, this article seeks to explain why expansion has not been consistent within single countries, across policy areas. Accepting macroexplanatory factors, this research seeks to contribute to existing theories by accounting for national variation across social policy sectors.

Case Selection, Methods, and Proposed Explanations

The main goal of this research is to produce some general theoretical propositions that explain those intersectoral inconsistencies and variations. Chile offers a unique opportunity to study the process of social policy expansion across sectors. The country saw the most radical market-oriented social policy reforms in Latin America in the 1970s and 1980s. This model rapidly diffused to other Latin American countries and continued after the democratic transition in 1990 for more than a decade. However, with the advent of the Socialist governments of Presidents Ricardo Lagos (2000–2006) and Michele Bachelet (2006–10), the largest social programs, healthcare and pensions, expanded. Both governments introduced policy innovations that sought to extend coverage, introduce new benefits, and increase state presence, particularly in the provision and financing of those programs. However, expansion was uneven: the government was able to accomplish many more of its goals with pensions than with healthcare.

This research carries out a hypothesis-generating type of case study. Hypothesis-generating (or heuristic) case studies examine a case with the purpose of developing general theoretical propositions (Bennett Reference Bennett2004, 22; Levy Reference Levy2008, 5–6). This type of case study contributes to the process of theory construction and hypothesis generation by offering a specification of causal mechanisms (Levy Reference Levy2008, 5–6). It seeks to identify variables and hypotheses heuristically through induction. The analysis of documents, archives, interviews, and other sources obtained during fieldwork may lead the researcher to identify relevant explanations (Bennett Reference Bennett2004, 35).

Due to the inductive nature of this research, the study also relies on inductive process tracing. As Bennett and Checkel explain, “the inductive, theory development side of process tracing uses evidence from within a case to develop hypotheses that might explain the case” (2015, 8). This logic of inquiry of process tracing proposes an “inductive approach in which the analyst derives propositions and formulates sequences from empirical observations. This mode of process tracing is often used for the purpose of theory development through the identification of key events and through the specification of hypotheses” (Falleti and Mahoney Reference Falleti and Mahoney2015, 229). In inductive process tracing, empirical material is used to detect “a plausible hypothetical causal mechanism whereby X is linked with Y” (Beach and Pedersen Reference Beach and Pedersen2013, 16). In this instance, “the analysis first traces backward from Y to uncover a plausible X, turning the study into an X-Y–centric analysis” (Beach and Pedersen Reference Beach and Pedersen2013, 16). Working with inductive process tracing, the researcher “might be best served by not deploying too-strict theoretical expectations that could act as blinders and straightjacket the interpretation of the process under study” (Falleti and Mahoney Reference Falleti and Mahoney2015, 230).

In pursuing inductive process tracing, evidence is taken from diverse sources. One source is interviews with qualified informants. Semistructured interviews with open-ended questions were conducted in late 2016 and early 2017. Also quoted are some other interviews carried out between 1999 and 2002 that illuminate the perceptions of some key actors. I work with nonprobability sampling not only because it is the most recommended approach to combine with process tracing but also because of several advantages: control over the selection process, inclusion of key political actors, and suitability for reconstructing events or sets of events (Tansey Reference Tansey2007).

The research also collected and analyzed information from newspaper coverage of the reforms; reports prepared by advisory commissions, government offices, and stakeholders; and transcripts from technical, government, and congressional meetings and sessions. I analyzed information obtained from the History of the Law system of the Chilean Library of Congress, which offers background information extracted from the Bill-Making System (Sistema de Tramitación de Proyectos de Ley) of the National Congress and from the transcripts of the Chamber of Deputies and Senate sessions.Footnote ¹

On the basis of the evidence collected through fieldwork, I generated three main hypotheses to account for governments’ ability to promote expansionary social policy. The independent variables proposed here are not entirely new, since the comparative literature has stressed their relevance for explaining cross-national variations in social policy. Yet they have not been incorporated into an analysis of uneven social policy expansion across social policy areas. I argue that uneven expansion may be explained by policymakers’ perceptions of budgetary constraints and fiscal costs of producing (or failing to produce) a reform; the composition, cohesion, and ideas of technical teams; and the relative power of nongovernmental, prowelfare actors in relation to market stakeholders. These independent variables interact through conjunctural causation. In other words, it is the combination of these variables (instead of their individual, separate effect) that explains uneven social policy outcomes.

A key variable in explaining social policy outcomes is policymakers’ perceptions of reform costs. Research on welfare state reform suggests that retrenchment is more likely to occur under conditions of budgetary crisis (Coleman et al. Reference Coleman, Atkinson and Montpetit1997, 459; Pierson Reference Pierson1996, 156). Moreover, Mesa-Lago (Reference Mesa-Lago1994, 115) claims that before democratic breakdown, Chilean social security “constituted a heavy economic burden … suffered financial and actuarial disequilibrium, and required substantial state subsidies.” The subsequent reform constituted, at least in part, an effort to deal with these problems. Similarly, in her analysis of the process of pension privatization, Brooks (Reference Brooks2009) shows that globalization generates incentives and constraints on governments that seek to restructure their pension systems, creating a double-bind effect. Capital-scarce nations might have incentives to avoid reforms that create short-term costs associated with capital flight, transition costs, and financing leeway; however, such reforms may also be perceived as a way to achieve long-term macroeconomic goals (Brooks Reference Brooks2009).

Informed by this literature and the findings through fieldwork, the first hypothesis of this work is that social policy expansion depends on the decisionmakers’ perceptions of actual or potential financial costs of producing or failing to produce a reform. In market-oriented social policy models, whenever government officials and policymakers consider that the actual or potential fiscal costs of perpetuating existing social policies will rise, they will be likely to pursue a reform strategy. Governments and policymakers will be more prone to promote the expansion of social policies that are considered fiscally viable and have the necessary resources allocated for implementation. This was precisely the case of pension reform in Chile. Conversely, when the resources to finance reforms are not available or policymakers fear that the adoption of such reforms might eventually push the system into fiscal deficit, reforms will either be unlikely or remain moderate. This was the case in healthcare reform.

In addition, several authors have argued that in the past few decades, Chile has experienced a marked technocratization of policymaking, which, in turn, has had an impact on public policy. After the transition to democracy, public policy was produced by the so-called transversal party, a vast network of professionals “whose personal identities were subordinated to the preservation of policy cohesion and stability” (Montecinos Reference Montecinos1998, 135). These professionals were highly cohesive and relatively insulated from party pressure, championing programs and measures within the existing market-oriented system.

One of the crucial functions of these technocrats was to ensure the rational administration of resources (Silva Reference Silva2009). The Concertación’s technocratic teams had a great deal of freedom to negotiate policy proposals informally with key actors, but their activities were coordinated and decisions highly controlled by the presidency, particularly when presidential leadership was strong (Fuentes Reference Fuentes1999, 204–8). Informed by this literature and the empirical evidence, the second hypothesis contends that disparities in the characteristics, levels of cohesion, and composition of technical teams in charge of different social policy areas produce uneven levels of expansion. In pensions, the presidential advisory council leading the “technical” discussion was homogeneous in composition and exhibited a very limited level of ideological dispersion, in a context in which its members shared a common diagnosis of the problems affecting the pension system and possible solutions. Conversely, in healthcare, two different, ideologically diffuse advisory commissions operated, which held three different visions of healthcare reform and did not share a common diagnosis.

The comparative social policy literature has shown that reforms depend on the capacity of nongovernmental actors and stakeholders to influence the policy process, promoting or obstructing policy change. I define nongovernmental, prowelfare actors as individual or collective actors (such as unions, social movements, associations of beneficiaries) that mobilize, organize, or seek to influence decisionmakers to promote the expansion of social policies. If they are divided or weak, such actors are less likely to influence the policy process. Conversely, if they are cohesive or strong, they will be more likely to influence social policy (Castiglioni Reference Castiglioni2012). Additionally, it is necessary to assess the balance of power between those in support of policy change and those who favor maintaining the status quo (Dávila Reference Dávila2005).

It is particularly important to evaluate the relative power of nongovernmental, prowelfare actors in relation to market stakeholders; that is, market service providers and insurers, whose interests might be threatened or damaged by expansionary social policy. Scholars studying pension reforms and rereforms have stressed the relevance of nongovernmental actors, particularly those able to influence the agenda or block initiatives to change the system (Baba Reference Baba2015; Kay Reference Kay1998; Madrid Reference Madrid2002). Moreover, Pierson (Reference Pierson1994) finds that whenever traditional networks of organized support for welfare programs break down, these networks often end up losing their ability to promote expansionary policy. This research hypothesizes that when the balance of power tilts toward cohesive nongovernmental, prowelfare actors, expansionary social policy is more likely to occur than when the balance of power tilts toward strong market stakeholders, unless the stakeholders’ interests are not threatened.

The degree to which individual or collective nongovernmental actors or stakeholders influence policies depends, in turn, largely on their levels of cohesion, their resources, and their organizational capabilities. Highly organized, cohesive actors are much more likely to influence policies effectively than loosely organized, divided actors. Strength also depends both on the economic resources available to exert their influence and on their ability to access policymakers. As Immergut asserts, “the ability of interest groups to influence such legislative outcomes depends upon their ability to threaten the passage of the law” (1992, 63).

Empirical evidence shows that in Chile, expansion in pensions was possible because unions did not oppose the reform proposal (even if they favored deeper reform) and because the proposal did not threaten market stakeholders. In healthcare, by contrast, some aspects of the reform proposal met resistance from the medical association and unions, and a key component of the reform threatened the interests of the Instituciones de Salud Previsional (ISAPREs) the private insurers, the most powerful market stakeholder in healthcare. All these factors contributed to the process of social policy expansion that Chile experienced.

Uneven Social Policy Expansion in Pensions and Healthcare

Chile experienced a marked process of social policy retrenchment in the 1970s and 1980s, when the military government of Augusto Pinochet began adopting radical, market-oriented reforms in all social policy areas. After redemocratization in 1990, under the center-left Concertación governments, market-oriented policies persisted. However, from 2000 on, social policy entered an expansionary phase.

In effect, Chile has seen some attempts to soften the most conspicuous aspects of the inherited model. Presidents Lagos and Bachelet modified the existing healthcare and pensions systems, respectively. These reforms did not seek to structurally modify social policies but to deal with their more flagrant inequalities. Additionally, the breadth of these reforms was different: the Bachelet administration gained the necessary support to get approval for most of its pension policy proposals, but the Lagos administration failed to gather support for one of the most important components of the healthcare reform, designed to introduce solidarity.

For the Lagos government, healthcare was a priority from the outset. Policy change in healthcare constituted an “attempt by the political left to move closer to the goals of solidarity and equity” (Dávila Reference Dávila2005, 7). This policy innovation represented a significant move toward universal coverage (Ewig Reference Ewig2015; Huber Reference Huber2011; Pribble Reference Pribble2013). However, as we will see, the reform that was finally approved was a far cry from what the Lagos administration had proposed.

President Lagos introduced six reform bills, five of which were finally approved. Two of these bills were related to key initiatives that the government eventually implemented. The most important of these initiatives was Law 19.966, corresponding to the AUGE-GES Plan (Plan de Acceso Universal y Garantías Explícitas de Salud). AUGE-GES seeks primarily to grant access to medical attention, on a clear timetable, to all patients who meet inclusion criteria for established pathologies (González F. Reference González F.2003, 546). The Ministry of Health maintains a list of pathologies and associated benefits for individuals insured under both the public and the private system. The list was prepared using several different criteria; namely, health impact, effectiveness of the treatment, and costs of treating each pathology (MINSAL 2004, 27). A pilot plan, which covered three pathologies, was launched in August 2002. The number of pathologies included under the GES plan was gradually expanded in 2006 to 56, in 2010 to 69, and today includes 80 pathologies.

However, the most emblematic aspect of the first bill, the FCS (Solidarity Compensation Fund or Fondo de Compensación Solidario), was not approved. This compensation fund was intended to reduce inequalities and promote solidarity by distributing the costs of the AUGE-GES Plan among the members of the public and private healthcare systems. As Ewig (Reference Ewig2015, 200) claims, “the health fund, in its original formulation, would have held a percentage of each insured person’s annual contribution, to cover the costs of high-risk patients and the poor, with the costs and benefits pooled between the private and public sectors.”

In April 2005, Congress approved Law 20.015. This law sought to address some of the issues the original AUGE-GES bill could not tackle, such as reducing age and sex discrimination in the private health system and strengthening the rights of the insured. This law limits the circumstances under which private health insurers (ISAPREs) can unilaterally terminate a contract, grants coverage to the family of deceased contributors for at least one year, forces ISAPREs to offer new treatments, limits healthcare plan price increases to no more than 30 percent of the mean adjustment of all ISAPREs, allows prices to be modified but not services, and creates an inter-ISAPRE fund to compensate individuals with a higher level of risk, such as the elderly and women (BCN 2004).

Pension reform was a key priority for the first Bachelet administration, and under it, the system experienced the most important transformation of individual capitalization since its inauguration in 1981. In March 2008, the government introduced a basic solidarity pillar (pilar básico solidario) through which 40 percent of the poorest individuals, who had never contributed to the system, would be entitled to receive an old age pension (for individuals 65 and older) or a disability pension (for disabled individuals older than 18 and younger than 65) of 60,000 Chilean pesos (in March 2009 approximately US$101) per month (BCN 2008). This amount was gradually raised, reaching CLP$102,897 (about US$155) in January 2017 (BCN 2008). Coverage was increased gradually to 60 percent of the poorest population by 2012. Individuals who receive an old age or disability pension below 70,000 pesos (increasing gradually to 309,231 pesos after July 2017, around US$470) are entitled to a solidarity pension contribution (aporte previsional solidario), except those enrolled in the military or police pension systems (BCN 2008). Both the basic solidarity pillar and the solidarity pension contribution were designed to reach the lowest income quintiles, around 60 percent of the poorest population segment (Arza and Chahbenderian Reference Arza and Chahbenderian2014, 22).

In addition, under the new system, all women older than 65 and retiring after July 1, 2009 are entitled to receive, for each child born alive, a state bond representing 10 percent of 18 monthly minimum wages. Other modifications include ability to enroll in the AFPs (Administradoras de Fondos de Pensiones) for people who are not currently working but would like to keep making contributions (such as women leaving the labor force for maternity), the possibility of dividing savings in case of divorce, new incentives for voluntary individual and collective savings, and other minor changes. The only measure that was not approved would have permitted banks to enter the pension market and, by extension, the emergence of a state-owned AFP. Yet all the components of the reform designed to promote equity and solidarity went through.

As this brief discussion suggests, Presidents Lagos and Bachelet were committed to expanding the existing healthcare and pension systems. However, in one area, pensions, the government managed to get the approval needed to pass most of its policy proposals, but in another, healthcare, it had to significantly modify the original proposal. Pension reform meant a 40 percent to 50 percent improvement in the monetary value of noncontributory pensions and the incorporation of new beneficiaries, as the basic solidarity pension covered 60 percent of the poorest population (Huepe and Larrañaga Reference Huepe and Larrañaga2010, 161). In healthcare, reforms were certainly a significant departure, but benefits reached only those individuals affected by one of the predefined pathologies. Moreover, the key policy innovation to promote equality and solidarity, the solidarity compensation fund, was rejected. Consequently, the government could get approval only for a watered-down version of its initial healthcare reform project. Table 1 summarizes the most important measures introduced in pensions and healthcare.

Table 1 Expansion in Pensions and Healthcare

Source: Author, with information from BCN, Plan GES o AUGE and Reforma previsional.

This uneven outcome arguably can be explained by the impact of the three interrelated factors: policymakers’ perceptions of the fiscal costs of producing (or failing to produce) a given reform, the technical teams in charge of social policies, and the relative power of nongovernmental, prowelfare actors in regard to market stakeholders. Although these explanations were identified inductively through fieldwork, they have been framed in light of the relevant comparative social policy literature. As a result, each explanation is placed in the context of the relevant literature.

Financial Sustainability and the Costs of Reforms

Policymakers’ perceptions about the financial sustainability of reforms and the costs of maintaining or reforming the existing system may change from one social sector to the next. In the case of pensions in Chile, the 1981 pension reform established that the state would grant a minimum pension to all insured workers with 20 years of contributions, so that if a worker reaching retirement age did not achieve the minimum pension level, the state would make up the difference (Arellano Reference Arellano1985, 63; Borzutzky Reference Borzutzky1990, 143; Huneeus Reference Huneeus2000, 460). The introduction of the basic solidarity pillar and the solidarity pension contribution reflected the realization (by both the government and the opposition) that in the near future, a significant segment of the labor force would not reach the amount of savings required to retire, and that the state would perforce transfer those resources. In fact, “low density of contributions and a large informal sector meant that nearly half of the workforce would not earn enough to receive a minimum pension” (Ewig and Kay Reference Ewig and Kay2011, 77).

According to Alberto Arenas de Mesa (Reference Arenas de Mesa2010), one of the key figures at the Ministry of Finance behind the Bachelet pension reform, a conference held in September 1999 was explicitly designed to spark a discussion of the system’s adverse fiscal effects. Arenas de Mesa presented new data from the Superintendent of AFPs that suggested that the transition costs had been much higher than anticipated (reaching annually an average 5.7 percent of GDP between 1981 and 1998) and that they would continue to increase over the next 40 years. Information about minimum pensions and adverse fiscal effects was widely disseminated by the local media and provoked concern and debate among policymakers, prompting the issue’s incorporation into the campaign that led Michelle Bachelet to the presidency (Arenas de Mesa Reference Arenas de Mesa2010, 37–38).

Another crucial consideration had to do with “recognition bonds” and the broader economic context. The 1981 reform established that all workers joining the new individual capitalization system who had previously contributed to any retirement fund for at least five consecutive years would get a recognition bond, monetary compensation acknowledging their previous contributions. More than three decades after the approval of the 1981 reform, most bonds have been paid, leaving additional funds available to finance the new solidarity pillar. Moreover, the reform proposal was developed amid good economic conditions. As a member of the presidential advisory council explained, “resources were available; at the time, the price of copper was booming … the state had accumulated plenty of financial assets abroad, we had a 7 percent fiscal surplus … so there was room to move. Besides, they were stopping paying the recognition bonds from the transition to the previous reform, so those fiscal resources were freed” (interview 14).

In December 2006, the government put together a Pension Reserve Fund, which was based on an initial contribution of more than six hundred million U.S. dollars. According to the Ministry of Finance’s website, this fund acknowledged “greater future retirement expenditures and the need to guarantee basic solidarity pensions to those who were not able to save enough for their retirement.” For both Congress and the presidential council, concerns about how the reform was going to be financed were central (interviews 1, 11, 12, 14, 15). However, by the time the bill reached Congress, there was widespread agreement that the reform was fiscally viable, due to the Pension Reserve Fund (interviews 1 and 14).

In July 2006, the advisory council delivered a report with 70 policy recommendations. Based on some of these recommendations, the executive developed and submitted a reform proposal to Congress in May 2007 (Mardones Reference Mardones2007, 79–80). The final version of this reform was approved in January 2008. Throughout this period, in the Congressional Commission on Finance, Labor, and Social Security, as well as in the two chambers, the projections of the adverse fiscal effects of not reforming the pension system were a key element of the parliamentary debate (BCN 2008). These concerns facilitated the approval of pension reform and gave policymakers strong incentives to endorse policy expansion.

In the case of healthcare, Lagos emphasized the need for reform during his presidential campaign. Before the Lagos government, policymakers concurred that healthcare was very costly for individuals and that it generated sharp inequalities, as well as a clear gap in access and quality between beneficiaries of the public and the private subsystems. However, policymakers did not achieve a consensus on a possible solution. As Muñoz asserts, until the late 1990s, there was “an absence of substantive reflection on what we wanted. … We were all critics, we had a more or less common diagnosis, but the paths to a solution were not there yet” (Muñoz 1999). Nevertheless, Lagos’s presidential campaign platform of October 1999 explicitly expressed the need for a reform to grant access to healthcare regardless of income level.

In the executive office, one of the most widely debated issues was how to finance the healthcare reform. This concern was fueled partly by the perception that the economic context was adverse in the aftermath of the Asian crisis of 1998, which Chile felt acutely during the Lagos administration (interview 3). According to ex–minister of health Álvaro Erazo, “the key debate between healthcare and financial authorities was how to limit the potential pressures in terms of costs that the delivery of explicit rights could exert over fiscal resources” (Erazo Reference Erazo2011, 20).

In this context, perhaps the most controversial component of the proposal was the Solidarity Compensation Fund (FCS). The FCS was the main feature of the reform, and it was designed to instill solidarity among beneficiaries of the public and private systems and to reduce inequalities of access and quality. This compensation fund sought to distribute the costs of the AUGE-GES Plan among the members of the public and private healthcare systems. The fund was to be financed through a fixed amount paid by the members of both systems and from the 7 percent payroll tax levied on employees. The state would provide additional funds to finance the costs of AUGE for the poor (Dávila Reference Dávila2005, 26–27; Ewig and Kay Reference Ewig and Kay2011, 80–81).

Yet the FCS became the most controversial component of the reform (interviews 2, 4, 5, 6, 7, 8). Instead, the new law created an inter-ISAPRE fund to compensate individuals with a higher level of risk in the private system, such as the elderly and women. It is important to emphasize that the inter-ISAPRE fund did not require additional fiscal resources from the state, in contrast to the FCS, which required fiscal contributions to finance the poor (Infante and Paraje Reference Infante and Paraje2010, 98).

Concerns about adopting the FCS and the fiscal viability of the healthcare reform as a whole led the government to make two main decisions. On the one hand, it sought to incorporate a “mechanism of rational spending” (Erazo Reference Erazo2011, 22). On the other, it decided to eliminate the FCS from the project, fearing that otherwise the entire bill would be rejected (interviews 2, 4, 5, and 7). When asked why the government decided to moderate the reform and why the FCS met so much resistance, one of the key government players went so far as to argue that the FCS was controversial “because it had to do with money. The only thing they [the opponents of the reform] care about is money. I mean, they do not care about solidarity, they do not care about the healthcare of the people, they do not care about reducing inequalities, the only thing they do care is how money is spent” (interview 5). Another said that “the right … saw this [the FCS] as a sort of tax, an expropriation of money that belonged to them. … If you ask me what I think, I would say that my interest was to get it [the FCS] approved. However, much more important was to get the AUGE approved” (interview 7).

In the presentation of the final bill, with the FCS out of the picture, to gain the support of the opposition and some reluctant Christian Democratic legislators, the government reframed its narrative, making an explicit effort to stress that equity and solidarity rested on the efficient use of resources (BCN 2004, 11). During congressional debates, right-wing senator Evelyn Matthei clearly explained why the final version of AUGE constituted an opportunity for a rational and efficient use of resources. The key aspect of AUGE was “to concentrate resources on those sicknesses that, treated in a quick and proper manner, can extend a healthy life or significantly improve the quality of living of those suffering [them]. … In the end, the aspiration of the AUGE Plan is to invest resources where they will produce a return in healthcare” (BCN 2004, 926).

Despite a clear emphasis on the need to use scarce resources efficiently, rationalization was not enough, and the government managed to get congressional approval only for increasing the value added tax from 18 percent to 19 percent until 2007. However, this ostensibly temporary measure eventually became permanent (Infante and Paraje Reference Infante and Paraje2010, 99). In addition, the Budget Law of 2005 also considered allocating resources for up to 1.02 UF per beneficiary (or CLP$17,655; in January 2005, approximately US$30.75). As a result, according to the report of the Senate Finance Commission, the healthcare reform would “neither produce budgetary imbalances nor negatively affect the economy of the country” (BCN 2004, 881). Because of these budgetary concerns, policy expansion was more limited in healthcare than in pensions.

Technical Teams

The military government brought with it a marked process of technocratization and depoliticization of policymaking. From the transition to democracy in 1990 until very recently, the Concertación “deliberately depoliticized class and redistributive issues. … Compared to the transitions to democracy in Argentina and Brazil, the Chilean transition was marked by a discourse in which themes of social justice, class, and redistributive issues were muted” (Mainwaring and Torcal Reference Mainwaring and Torcal2000, 30–31). This nonconfrontational approach to policymaking shaped social policy. Concertación governments also sought to mitigate business concerns that a center-left government could jeopardize economic stability. This forced the Concertación not only to construct a close system of collaboration with the business sector but also to emphasize that the previous government’s development strategy would be maintained (Silva Reference Silva1996, 231).

The Lagos and Bachelet administrations maintained this technocratic emphasis. Social policy during this period was designed in accordance with the existing economic development model and subordinated to macroeconomic performance (Silva Reference Silva2009). However, the specific form that technocratization adopted changed from one policy area to the next. In pensions, right after taking office, Bachelet put together a presidential advisory council headed by former director of budget and finance minister Mario Marcel, with the goals of diagnosing the pension system and crafting a reform proposal. As a member of this council asserted, the presence of Marcel, given his background, “made it impossible to overlook the financial aspects of the reform” (interview 14).

In general terms, the Concertación governments endorsed the existing individual capitalization model, believing that “the state is a bad administrator, because the funds managed by the state end up being captured by interest groups” (Walker 2002). Thus, “within the [Concertación] coalition nobody … suggested replacing the existing model with a pay as you go system. … This does not mean that the idea [was] not to introduce some adjustments to the system, but within the individual capitalization system” (Cortázar 2001). On the council, debates were bounded by the endorsement of individual capitalization, which went unchallenged. The council also developed an agreement on the introduction of a solidarity pillar, assuming, as discussed above, that the resources necessary to sustain it were available.

Economists had an important presence on this council, which excluded representatives from the political parties. It is quite telling that it was headed by a former director of budget and finance minister and that the liaison between the council and the executive was the director of budget of the Finance Ministry. It was nonetheless open to feedback from different individuals and institutions. The composition and function of the council were explicitly intended to avoid “a surge in populist politics” (Pribble Reference Pribble2013, 79). The council was markedly homogeneous.

Although the general political setting in which social policy reform occurred was similar in pensions and healthcare, the approved reform project in pensions was closer to the one originally presented by the executive than in the case of health. This is partly the result of the characteristics and level of cohesion of the technical team behind this project. In pensions, the council comprised 15 members: 11 economists or business administrators with similar formative backgrounds and visions, and 4 members trained in law or sociology. Some of the members belonged to conservative think tanks; others were public servants—mostly from Concertación administrations—who specialized in social security. One was a former minister of the military government, one belonged to an independent organization working on gender, and others were scholars (Arenas de Mesa Reference Arenas de Mesa2010, 60). The homogeneous council, with limited ideological dispersion, facilitated the emergence of an agreement on pension reform (interviews 9 and 15).

In the case of healthcare, immediately after the election, President Lagos asked one of his closest collaborators and a longtime friend, Dr. Hernán Sandoval, to lead a presidential advisory commission on healthcare reform to provide technical advice to the president (Dávila Reference Dávila2005, 23). Sandoval was a well-known physician and member of the Partido por la Democracia (PPD). The “technical” commission, in charge of crafting policy recommendations for the reform, worked in tandem with an interministerial commission responsible for defining the content of the reform proposal and pushing for the necessary legislative, legal, and financial changes (Muller Silva Reference Muller Silva2004, 61–62). The technical commission had five members (two physicians, one of them also trained in law; two lawyers; and one economist). The interministerial commission consisted of the ministers of finance (an economist), health (a physician), and labor (an economist); the general secretary of the presidency (an economist); and the undersecretary of health (a physician), along with the director of budget in the Finance Ministry (an economist), and Dr. Sandoval. The interministerial healthcare commission was rather heterogeneous and reflected a constant tension between public health and fiscal concerns.

Within the Lagos administration, there was a high level of ideological diversity, as three different visions of healthcare reform prevailed. According to Olavarría Gambi (Reference Olavarría Gambi2011), there was a “statist” view that rejected profiting in the realm of social provision, questioned whether regulation could correct market inequalities, and favored limiting the role of ISAPREs. Additionally, he claims that there was an “integrated health” approach that sought to increase public sector efficiency, wanted to improve management mechanisms, recognized the role of ISAPREs but wanted to reduce discrimination in the private healthcare subsystem, and advocated the creation of a superintendency with monitoring responsibilities in both the private and public systems. A third, “proprivate” view, endorsed by right-wing sectors outside the executive office, favored improving the ISAPRE system, increasing efficiency in the management of public healthcare, boosting the competitiveness of the public and private sectors, introducing new subsidies that beneficiaries could freely use in the private or public subsystems, and avoiding increasing taxes and regulatory pressures on ISAPREs (Olavarría Gambi Reference Olavarría Gambi2011, 363–64).

Partly because of the existence of these contrasting views and partly because of tension over the roles of both commissions (Lenz Reference Lenz2007, 12; Olavarría Gambi Reference Olavarría Gambi2011, 364), healthcare reform was comparatively much more complex than pension reform.

One of the most fiercely debated issues on the interministerial commission was how to finance the healthcare reform. Discussions about budgetary issues and the FCS dominated debates in Congress, where ideological differences were, similarly, quite prominent. The FCS met the opposition of both right-wing and Christian Democratic legislators and confronted the powerful private healthcare provider lobby (Dávila Reference Dávila2005; Ewig Reference Ewig2015). ISAPREs lobbied through right-wing legislators, who, in turn, used their position to block the FCS (Ewig and Kay Reference Ewig and Kay2011). Opposition in Congress to the FCS was so fierce that in the end, the government backed down and the reform project was approved without its most controversial and groundbreaking component.

Nonstate Actors and Stakeholders

A long period of market-oriented social policy and privatization in Chile brought with it not only the weakening of organized civil society but also the emergence of new market actors in charge of the provision of social services, such as AFPs and ISAPREs. These market stakeholders became crucial players whenever policy change was likely to affect their interests. Since the transition to democracy in 1990, Chile has exhibited an imbalanced relationship between nongovernmental actors and the business sector.

Nonstate actors organized around social sectors have displayed different strategies and levels of success in achieving their goals. Until recently, Chile lacked a major national organization of pensioners. Organizations like “No + AFP” (a social movement that opposes the individual capitalization system, as well as the presence of AFPs, and favors the reinstallation of a pay as you go system) have existed for some time, but it was not until 2015 that they became relevant (interview 14). The Central Unitaria de Trabajadores (CUT), the largest federation of workers in Chile, was critical of the individual capitalization system, decrying insufficient levels of coverage, low density of contributions, and high administration costs (Maldonado Valera and Palma Roco Reference Maldonado Valera and Palma Roco2013, 44). It favored a reform that included the development of a solidarity pillar and the direct participation of the state in the administration of workers’ savings. To that end, CUT maintained informal contacts with individual legislators from the left, hoping that its ideas would echo in the reform proposal (Maldonado Valera and Palma Roco Reference Maldonado Valera and Palma Roco2013, 44).

ANEF (Agrupación Nacional de Empleados Fiscales) sought unsuccessfully to push for a reinstallation of the pay as you go model. Both CUT and ANEF did not oppose the adoption of the solidarity pillar, even if they wanted a deeper reform. As a union leader explained, the solidarity pillar “allowed those who had no access to a pension to get a basic one. … But we did not achieve structural progress, because … this still is an individual capitalization system administered by for-profit private entities and this is not a social security system” (interview 10). In short, the weakness of nongovernmental actors and interest groups, such as pensioners’ associations, contributed to the adoption of reforms that did not challenge the individual capitalization model.

These relatively weak civil society organizations confronted powerful promarket actors. AFPs are organized and regularly coordinate their actions through an association (Asociación de AFPs). AFPs and ISAPREs possess substantial economic and technical resources and have direct access to legislators, particularly those from the center-right. The private AFP system has used its power to “immobilize initiatives related to the social dimension of the pension system. Even gathering information on coverage, density, and pension history of citizens meant, for governmental authorities, a conflict during the first years of the early 2000s” (Maldonado Valera and Palma Roco Reference Maldonado Valera and Palma Roco2013, 43). In this context, it is not surprising that most of the pension reform measures under study did not threaten the interests of AFPs. This alone is a sign of how powerful and influential AFPs are.

Nevertheless, one specific measure was directly detrimental to AFPs’ interests. In 2006, the presidential council discussed the possibility of allowing private and public banks to operate in the AFP system. This measure could have permitted the emergence of an AFP belonging to the State Bank (Banco Estado). Some members of the presidential council argued that a state-owned AFP constituted a controversial but minor issue (interview 12), and others claimed that even in the executive office, some key players opposed the idea (interview 14). The main concern with this measure was that beneficiaries would prefer a State Bank AFP and that this would force existing AFPs to face unfair competition.

In an interview conducted in July 2006, Guillermo Arthur, the president of the AFP Association, emphasized that allowing a state-owned AFP would be like “introducing a monster” into the system and would “discourage the appearance of new players” (quoted in García Reference García2006). Subsequently, AFPs used their power to block this initiative successfully, particularly through right-wing legislators. As a key figure from the executive office explained,

Thus the plan to have a state-owned AFP was quickly dismissed.

The pension milieu contrasts sharply with that of healthcare, where interest groups, particularly the Colegio Médico (CM), remain relatively active. During military rule, the CM lost most of the regulatory, policymaking, and wage-setting prerogatives it had gained before the democratic breakdown (Castiglioni Reference Castiglioni2005). Despite its institutional weakness, it was an important voice in the realm of healthcare policy after redemocratization. The CM has remained critical of the market-based healthcare system, which it blames for the disintegration of the public system, unequal healthcare access and treatment, the weakening of primary services, the lack of resources for public hospitals, a low response to chronic pathologies, and the excessive costs and lack of transparency of ISAPREs (CM 1998).

During the discussions of the healthcare reform, CM appeared as one of the most critical actors. Interviews and a review of the press reveal two main reasons behind CM opposition. First, physicians thought their working conditions would be harmed by AUGE (interviews 3, 5, 7). One of the leading figures behind the reform explained this point bluntly: “this [reform] imposes a rhythm, a pace, and obligation to the system and to the doctors who must do things that, before, they performed whenever they felt like it. So [doctors] interpreted this [reform] as a loss of freedom, as an imposition” (interview 5). Second, the CM claimed that it opposed the market-oriented rationale of the reform, as AUGE did not offer universal coverage to all patients (only those affected by AUGE pathologies) and the reform would allow the transference of public resources to private providers (interviews 4, 13). This rejection of the alleged neoliberal rationale of AUGE glued together different civil society organizations that joined CM in its opposition. As a union leader explained,

Despite their institutional weaknesses, civil society organizations in healthcare were able to coordinate their efforts in order to express their discontent with the reform.

Inspired by different reasons from those of the civil society organizations, ISAPREs managed to influence the policy process. In this way, “ISAPREs averted a reform that would have struck a major blow to ISAPRE interests. The Solidarity Fund was a clear threat to profits, market share, and long-term viability” (Ewig and Kay Reference Ewig and Kay2011, 84). ISAPREs lobbied to block policy change through the Asociación de ISAPREs, an industry organization. Since their creation, ISAPREs have shown remarkable organizational capacity; it is also clear that their opinions have penetrated center-right political parties (Dávila Reference Dávila2005, 41–42). During the debates over healthcare reform, ISAPREs publicly held moderate positions regarding AUGE. However, President Lagos himself directly accused ISAPREs of hiding their true positions while lobbying against the reform through right-wing legislators, whose campaigns ISAPREs financed (Ewig and Kay 2001, 82).

AUGE suffered attacks on two different fronts. Civil society organizations opposed AUGE’s alleged neoliberal nature in a context in which the CM saw the reform as a threat to the working conditions of physicians and the functioning of the healthcare system. Nevertheless, since these actors remained institutionally weak, their capacity to influence the policymaking process remained low. Furthermore, ISAPREs used their power and resources to block the FCS. Thus, the government decided to remove the component of solidarity from the project in order to avoid putting the entire reform process at risk.

In pensions, the weakness of civil society actors was coupled with the strength of AFPs, but the reform project did not threaten their interests. Thus the reform was approved without delay. The only measure that was not approved was the one that would have permitted a state-owned AFP. Since this measure also had detractors in the executive office and Congress, it was promptly removed from the reform project. Table 2 summarizes the main arguments and findings of this article.

Table 2 Main Argument and Findings

Final Remarks

Chile underwent a process of uneven social policy expansion in healthcare and pensions. Although the main features of the individual capitalization pension model and the healthcare systems were maintained, under the presidency of Lagos and the first administration of Bachelet there was a commitment to mitigating the most conspicuous inequalities of decades of radical, market-oriented social policy change. In pensions, the government was able to advance most of its proposed policy changes, but in healthcare, some of the most important components of the reform were not approved. The aim of this article has been, thus, to explain uneven social policy expansion in Chile.

There is an abundant literature on recent social policy expansion. This literature argues that a series of macroexplanatory factors are relevant to explaining social policy outcomes. Pribble (2014), for example, shows that legacies, electoral competition, and party character account for different social policy results. However, when explaining differences across policy sectors, these systemic explanatory factors lose relevance, as they are not sector-specific. In other words, explanatory factors such as the “left turn,” the commodity boom, legacies, electoral competition and party character (to name a few) are systemic factors that do not vary from one sector to the next. This study, in contrast, leaves constant most macroexplanatory factors—which the literature proves are crucial for cross-country analysis—to focus on accounting for differences in the scope of expansion across sectors.

Accounting for that variation requires generating hypotheses that may be useful for theory building in the realm of intersectoral dynamics of social policy expansion. Thus, through a hypothesis-generating type of case study and inductive process tracing, this article identifies three explanatory factors that together explain uneven social policy expansion. These factors are the policymakers’ perceptions of the potential fiscal costs of producing (or failing to produce) a reform; the composition, level of cohesion, and ideas of the members of the technical teams in charge of social policies; and the relative power of nongovernmental, prowelfare actors in regard to market stakeholders.