Portal de Periódicos da CAPES

Basic Old‐Age Protection in Latin America: Noncontributory Pensions, Coverage Expansion Strategies, and Aging Patterns across Countries

2019; Wiley; Volume: 45; Issue: S1 Linguagem: Inglês

10.1111/padr.12267

1728-4457

Camila Arza,

Intergenerational Family Dynamics and Caregiving

Population and Development ReviewVolume 45, Issue S1 p. 23-45 SUPPLEMENTOpen Access Basic Old-Age Protection in Latin America: Noncontributory Pensions, Coverage Expansion Strategies, and Aging Patterns across Countries Camila Arza, Camila ArzaSearch for more papers by this author Camila Arza, Camila ArzaSearch for more papers by this author First published: 24 July 2019 https://doi.org/10.1111/padr.12267Citations: 6AboutSectionsPDF ToolsRequest permissionExport citationAdd to favoritesTrack citation ShareShare Give accessShare full text accessShare full-text accessPlease review our Terms and Conditions of Use and check box below to share full-text version of article.I have read and accept the Wiley Online Library Terms and Conditions of UseShareable LinkUse the link below to share a full-text version of this article with your friends and colleagues. Learn more.Copy URL Share a linkShare onFacebookTwitterLinkedInRedditWechat Historically, Latin American pension systems were designed following a social insurance model geared toward offering pension benefits to formal-sector workers. From 1981, and especially over the 1990s, several countries implemented structural pension reforms to introduce privately administered individual accounts that more closely linked individual benefits to past contributions (Mesa-Lago 2004). By design, these systems excluded the redistribution of pension resources among workers and pensioners, promoting individual savings instead. In several countries, the contribution records required to receive a minimum pension from the social security system left ineligible most informal-sector workers, as well as agricultural-sector workers and unpaid family workers. As a result, the reformed pension systems suffered from persistently low levels of coverage, and a large share of older adults had no access to a pension. In more recent years, in a context of economic growth and changing ideas and political balances, several Latin American countries have implemented policies to extend economic protection to older adults who could not obtain a pension in the contributory system. Some countries reformed pension systems that had been privatized in the previous decade (including cases of reform reversals, as in Argentina). Many countries also promoted the expansion of noncontributory pensions for older adults and persons with disabilities or made access to contributory pensions more flexible for some types of workers (Arza 2013; Arza and Knox-Vydmanov 2017; Bosch, Melguizo, and Pagés 2013; Palacios and Knox-Vydmanov 2014; Rofman, Apella, and Vezza 2013). These policies were part of a wider process of welfare expansion in other areas of social protection, which involved greater social expenditures and more social assistance benefits (Barrientos 2011). This article evaluates the patterns of old-age protection in Latin America after recent expansions of basic pension coverage. It characterizes noncontributory pensions and related programs oriented to offer benefits to older adults lacking a record of formal employment and contributions. It is based on the systematic study of the design features of these policies in 14 Latin American countries, and relies on various data sources, including quantitative data (mainly from IDB 2017 and CEPAL 2017a), comparative databases of noncontributory pensions' design features (CEPAL 2017b, HelpAge International 2015), secondary sources, and official data from countries to complement and update other sources. This article first discusses the need and opportunities for pension coverage expansion in Latin America, referring to pension coverage deficits and the window of opportunity for dealing with them in the past two decades. Second, it documents existing programs that offer cash benefits to older adults not qualifying for contributory pension benefits and identifies different strategies and policy instruments across countries. Finally, the article explores the cross-country differences in living arrangements and retirement choices of older adults, indicators of the interactions between states, markets, and families in providing economic security in old age. Inequality, informal work, and population aging: The need and opportunity for pension coverage expansion Socioeconomic inequality in Latin America is very high and involves earnings inequality as well as unequal access to formal employment, social security, and social services. In most countries, the majority of low-income workers are in the informal labor force, have no access to contributory social security benefits and limited access to social protection at all. Thus, a large share of the population of Latin America lacks what scholars have called the "double incorporation" (Martínez-Franzoni and Sánchez-Ancochea 2014)—i.e., incorporation both into the formal labor force and into the social protection system. As has been widely documented, many older adults have no right to a contributory pension. Structural pension reforms implemented first in Chile in 1981 and in several other Latin American countries over the 1990s and early 2000s have not solved this problem. These reforms established privately administered individual accounts as a key component of the mandatory pension system, to substitute for or operate jointly with the public earnings-related system. Variations of this model were implemented in Argentina, Bolivia, Chile, Colombia, Costa Rica, the Dominican Republic, El Salvador, Mexico, Panana, Peru, and Uruguay (Mesa-Lago 2004, 2006, 2012a; Madrid 2002). The reformed pension systems did not perform well. Coverage remained low, and there were concerns in some countries that benefits for many covered workers—especially those with interrupted work histories or low earnings—would be too low (see e.g., Pension Commission 2006 on the Chilean case). This encouraged governments to introduce new measures, and some countries (like Argentina, Bolivia, and Chile) implemented post-privatization pension reforms that in different ways enhanced the role of the state in old-age protection (Kritzer, Kay, and Sinha 2011; Bertranou, Calvo, and Bertranou 2009; Mesa-Lago 2012b; Mesa-Lago and Ossio Bustillos 2012). Many countries also started a process of expanding pension coverage, by developing new noncontributory pension programs or by extending existing programs to provide economic support to older adults not covered by the contributory pension system. A positive macroeconomic and fiscal context, in an international environment of high commodity prices, was favourable for the growth of public social expenditures in general, and social assistance programs in particular. On average, social expenditures in Latin America rose from 14.7 percent to 18.8 percent of the gross domestic product between 1998–1999 and 2012–2013, and the share of social to total public expenditures increased over the same period, from 53.8 percent to 64.8 percent (CEPAL 2014). A political and ideological shift away from market-based orthodoxy toward greater state intervention also provided the political context for expanding cash benefits in several countries. The number of governments situated to the left of center on the political spectrum (notwithstanding their differences) increased (Huber and Stephens 2012). The expansionary trend also went beyond left-of-center governments. In the case of Mexico, for instance, initiatives from progressive forces in local governments and the opposition were relevant in placing these policies within the political debate and eventually encouraged their implementation at the federal level under a government with a different political orientation (see Willmore 2014). Noncontributory pensions started to be promoted by various relevant actors (including international organizations, notably through the International Labour Organization's [ILO] Social Protection Floor Initiative; see ILO 2011, 2012) and, just as happened with conditional cash transfer programs, it became clear that noncontributory pensions could reach many people, involve relatively limited budgets, and enjoy popular support. Coverage expansion also happened in a context of population aging. On average, the share of the population aged 65 and older reached 7.5 percent in 2015 and is projected to rise to almost 25 percent by 2065 (CEPAL 2017c). In some countries, such as Argentina, Chile, and Uruguay, the aging process is more advanced, with the population aged 65 and older already representing more than 10 percent of the total population. As the share of the elderly population increases, pension coverage and adequacy are likely to become more central public policy issues. For instance, in countries with underdeveloped pension systems characterized by low coverage, the aging process can put greater pressures on younger generations in the family, as fewer children may have to provide economic support for parents (and sometimes grandparents) who lack a pension. On the fiscal side, the impact of population aging on pension financing (a key concern in more advanced economies) varies across Latin American countries, depending on how advanced the aging process is, as well as on the level of pension coverage and the generosity of benefits. The expansion of noncontributory pension coverage occurred in the context of these shared needs and opportunities. High levels of socioeconomic inequality and informal work and the inability of contributory systems and pension reforms to guarantee broad coverage, as well as changes in the socioeconomic and political context, were the scenario in which new basic old-age pension programs developed and expanded. However, not all Latin American countries expanded coverage similarly, and progress remains limited in some. In addition, as we will see, countries that implemented policies to expand the coverage of older adults opted for a variety of specific policy instruments, with differences in aims, reach, and generosity. Policy design and policy instruments across countries The expansion of old-age pension coverage occurred in countries with a variety of pension systems—public Pay-As-You-Go (PAYG) systems, privately administered individual accounts, and mixed or parallel systems in which individual accounts coexist with public pensions. While coverage expansion strategies varied across countries in kind and scope, they all shared a key feature: the creation or expansion of programs offering old-age benefits that did not require a record of past contributions as a condition for access (Rofman, Apella, and Vezza 2013; Arza 2013; Arza and Knox-Vydmanov 2017). Table 1 lists the most important of these policies in Latin America. Noncontributory pensions—also referred to as "social pensions"—are "regular government provided cash transfers to older people" (Palacios and Knox-Vydmanov 2014, 251) that are not linked to past contributions. For that reason, they are labeled "noncontributory," and that common usage is followed here, although some key basic pension programs analyzed in this article are partly financed through contributions and administered as part of the general pension system (such as the rural pension in Brazil or the pension moratorium in Argentina, described below).1 In most cases, noncontributory pensions target the elderly poor. In others, these programs have wider reach and aim to cover, on their own or in combination with contributory benefits, the entire population of older adults. Some benefits are integrated with the contributory pension system, but others are separate programs that work more as poor relief for the impoverished aged than as retirement benefits. Table 1. Background information on noncontributory benefits and other coverage expansion programs for older adults in Latin America Program Targeting Argentina Noncontributory pension program Means-tested Pension Moratorium Not means-tested Universal pension for older adults (PUAM) Pension-tested Bolivia Renta Universal de Vejez "Renta Dignidad" (previously Bonosol) Universal Brazil Benefício de Prestação Continuada (BPC) (Continuous benefit program) Means-tested Rural Pension (Previdência Rural) Not means-tested (small farmers and subsistence fishermen) Chile Basic Solidarity Pension for Old Age (previously PASIS) Means-tested Colombia Colombia Mayor (previously PPSAM) Means-tested Costa Rica Noncontributory pension scheme by basic amount (Régimen no contributivo de pensiones por monto básico) Means-tested Ecuador Pension for elderly and pension for persons with disability (human development grant program) Means-tested El Salvador Our Elderly Rights (Nuestros Mayores Derechos) Means-tested and geographic (shanty towns) Guatemala Economic support for the older adult (Aporte Economico del Adulto Mayor) Means-tested Mexico Pension for the elderly (previously "70 and over" program) Pension-tested Program of food support for adults over 68 years old living in Mexico City Universal (only Mexico City residents) Panama 120 at 65: Special program for economic assistance of older adults (previously 100 at 70) Means-tested Paraguay Food pensions for elderly in a situation of poverty Means-tested Peru National solidarity assistance program "Pension 65" Means-tested Uruguay Noncontributory pension for old age and disability Means-tested NOTE: Some programs included in this table also cover persons with disabilities. Programs and features by c. 2017. SOURCE: Elaborated by the author based on CEPAL (2017b) and Arza (2017a). Rofman, Apella, and Vezza (2013, 10) made an important contribution in documenting the expansion of noncontributory pension benefits in Latin America. They identified two strategies that countries have adopted to expand coverage: the expansion of traditional contributory pension systems, and the development of poverty reduction programs targeted at older adults. To offer a taxonomy of noncontributory pensions, the authors compared some of their key design features, specifically looking at whether benefits were: (1) universal or means-tested; (2) integrated with or independent from contributory systems; (3) permanent or temporary; (4) implemented gradually or immediately; and (5) managed by existing or new institutions (ibid., 38–43; see also Rofman, Apella, and Vezza 2015). In line with that approach, this article aims to move further in the analysis by jointly evaluating a number of dimensions that characterize countries' old-age coverage strategies, including benefit adequacy and the relative role given to contributory and noncontributory benefits. To do this, it considers five main dimensions: (i) the basic benefit amount; (ii) the total old-age coverage rate; (iii) the contributory pension coverage rate; (iv) the noncontributory pension coverage rate; and (v) the eligibility condition for noncontributory benefits (means-tested or non means-tested). On the basis of these five dimensions, Table 2 conceptualizes three alternative strategies for old-age coverage, each combining a particular set of outcomes in the five dimensions. To fit in each model, countries are expected to have a combination of features—i.e., to present a particular value in each indicator, as described in Table 2. Table 2. Alternative strategies for old-age pension coverage in Latin America I—Basic benefit amount II—Total old-age coverage III—Contributory old-age coverage IV—Noncontributory old-age coverage V—Eligibility Country cases Contributory Plus Above average High High Low to mid-range Any Argentina,a Brazil, Chile, Uruguay Universal Minimal Below average High Low High Non-means-tested Bolivia, Mexicob Means-tested poverty relief Below average Low to mid-range Low Low to mid-range Means-tested Colombia, Ecuador, El Salvador, Guatemala, Paraguay, Peru NOTES: The five strategies are defined as follows: I—Basic benefit amount: Noncontributory pension benefits (or similar benefits paid to older adults without a full record of contributions), expressed in 2016 US$ PPP. "Average" refers to the simple average of these benefits in the 14 Latin American countries included in this article (see Table 1). II—Total old-age coverage: "high" (above 80% of the population aged 65 and over receives a benefit); "mid-range" (between 60% and 79%); "low" (under 60%) III—Contributory old-age coverage: "high" (above 50% of the population aged 65 and over receives a contributory benefit); "mid-range" (between 30% and 50%); "low" (under 30%) IV—Noncontributory old-age coverage (or similar coverage expansion program): "high" (above 50% of the population aged 65 and over receives a noncontributory or similar benefit); "mid-range" (between 20% and 50%); "low" (under 20%) V- Eligibility: Means-tested or not means-tested a Contributory coverage (point III), excluding moratorium pensioners, is a few points lower than expected for its group. b Total coverage (point II) is a few points lower than expected for its group. SOURCES: Own elaboration. See notes to Figure 1 for details on reference years, data sources, and estimation details for country cases. The empirical analysis, which uses the most recent data available at the time of writing, allows the classification of 14 Latin American countries into three groups based on the framework outlined above. The first strategy, labeled "Contributory Plus," was implemented in Argentina, Brazil, Chile, and Uruguay. It characterizes countries that have comparatively high coverage rates in the contributory pension system and that use a variety of instruments (means-tested benefits, flexibility for access, rural pensions) to reach close-to-full coverage. The benefits these countries offer to workers outside the formal labor market are relatively more generous than (or at least not as low as) those of other Latin American noncontributory schemes. The second strategy, labeled "Universal Minimal," characterizes countries that have low contributory pension coverage and universal or pension-tested schemes with broad coverage and modest benefits (Bolivia and Mexico). Finally, the third strategy involves countries with a more residual approach to old-age protection, that fit in a "Means-Tested Poverty Relief" model. These are countries with low contributory pension coverage rates, which extended the coverage of older adults with noncontributory benefits that are low, means-tested, and separate from the contributory system. Total old-age coverage in this group ranges from low (in El Salvador and Guatemala) to mid-range (in Colombia, Ecuador, Paraguay, and Peru), mainly because of variable efforts in the development of noncontributory pensions, and in all cases a considerable share of the elderly has no benefits. Two countries (Panama and Costa Rica) are situated between this group and the "Contributory Plus" group and are therefore considered to be mixed cases.2 Contributory plus Argentina, Brazil, Chile, and Uruguay currently have the highest contributory pension coverage in Latin America. In Argentina, total coverage rates increased from 69 percent to 91 percent of the population aged 65 and over between 2005 and 2015 (IDB 2017). Not all of these benefits were fully contributory, however, and the increase in coverage rates reported above was largely the outcome of a "pension moratorium" that made contributory requirements for getting a pension more flexible (Arza 2012a; Arza 2013; Danani and Hintze 2011; Rofman, Fajnzylber, and Herrera 2010). This program made older adults who lacked the contribution records needed for a normal retirement benefit also eligible to a pension. From the new benefit, a monthly deduction was made to compensate for the contributions missing. Open to all, the pension moratorium produced a massive increase in coverage rates, especially among women (Arza 2017b). By 2015, pensions paid to older adults who joined the moratorium represented about 47 percent of all pensions paid,3 and the benefit amount was normally close to the legal minimum pension. In 2016, a pension reform created the Universal Pension for Older Adults (PUAM, for the acronym in Spanish), a pension-tested benefit for older adults with no other pension, which progressively replaces the pension moratorium as the policy instrument to incorporate older adults lacking contributory benefits. The Argentine strategy for coverage expansion had two key features. First, older adults not previously covered were incorporated into the same pension system as formal workers, and shared with them the rights to regular benefit indexation and health insurance. This contrasted with the most common strategy in other countries, which was to create a separate noncontributory pension program to incorporate the "outsiders." In fact, a separate noncontributory social assistance pension program also existed in Argentina (Bertranou, Solorio, and van Ginneken 2002), but most older adults availed themselves of the pension moratorium instead, because the conditions were better. Second, the pension moratorium was conceived as a temporary measure (workers could use it only to compensate for past years of informal work performed up until 1993, and later extended to 2003). In terms of financing, the pension moratorium involved a relatively high share of the budget compared with noncontributory pension programs in the region. This was because, unlike those programs, the moratorium also included older adults with some contribution records, and paid more and higher benefits. It was feasible in a context of growing social security revenues and the nationalization of individual pension accounts in 2008, which redirected all workers' contributions and accumulated pension funds to the national social security administration (Arza 2012b). In the case of Chile, one of the aims of the 2008 pension reform was to improve the protection of the population not covered by the contributory system (Arenas de Mesa 2010; Berstein et al. 2009; Rofman, Fajnzylber, and Herrera 2010). The reform created the Basic Solidarity Pension for Old-age (PBSV, for the acronym in Spanish) to provide benefits for older adults who had no other pensions and who belonged to the lowest 60 percent of households in terms of income. It currently covers about 29 percent of the elderly population (IDB 2017). The same reform also created the Solidarity Pension Contribution for Old-age, a noncontributory pension supplement for older adults with some pension savings but low benefits. It is estimated that contributory and noncontributory pensions jointly reached about 90 percent of older adults in 2015 (IDB 2017). Benefits were established as rights for all those meeting eligibility conditions, with no quotas or waiting lists. The PBSV is, however, relatively low—the lowest basic benefit amount in the Contributory Plus group, and well below the basic pension in Brazil or Argentina. In Brazil and Uruguay, coverage rates were already high 20 years ago. In the case of Brazil, this wide-ranging social security system was part of new social rights established in the Federal Constitution of 1988. By the mid-1990s, the old-age coverage rate was (as in Uruguay) already above 80 percent, and in 2015 about 84 percent of older adults received a pension. Two benefits are important to reach the populations that are hard to cover with purely contributory benefits: a social assistance pension (Benefício de Prestação Continuada—BPC) and a Rural Pension (Previdência Rural).4 Following the 1988 Constitution, both benefits were set at the value of the minimum wage, which is the minimum amount for all types of pensions in Brazil. In the case of Uruguay, old-age pension coverage rates have long been high. A noncontributory pension program was created in 1919 (Amarante and Vigorito 2012; Bertranou, Solorio, and van Ginneken 2002), but the large majority of pensioners have a contributory benefit. From 2005, the government has tried to expand coverage of older adults in various ways: first, through the Plan de Equidad Social and the creation of a new noncontributory subsidy for older adults without pensions at ages 65–69 (which complements the existing noncontributory old-age pension for persons 70 and over), and second, through a reform of the contributory pension system that made access more flexible, reducing the minimum number of contribution years required and creating facilities for women with children and for persons of advanced age, among other measures (Papadópulos 2013, 478–480). Thus, with a range of instruments, applied on structurally different pension systems, these four countries share three key features. First, they are among the best achievers in terms of old-age coverage (with total old-age coverage rates above 80 percent). Second, in all of them, contributory pension coverage represents a high share of total coverage. Finally, the benefit amounts that they offer to older adults with no contribution records are above the Latin American average and higher than the basic benefits in every other country included in this article. Universal minimal In most other Latin American countries, contributory pension coverage is significantly lower than in those discussed above. Therefore, these countries have relied more heavily on noncontributory pensions to expand coverage. A couple of them have even implemented universal or pension-tested noncontributory benefits aimed at the entire population of older adults (Bolivia) or at every older adult with no other pension (Mexico), without focusing on the poor or extremely poor only.5 Since contributory pension coverage in these countries is low, noncontributory programs provide the majority of benefits. However, benefit amounts are very modest (hence the label "Universal Minimal"). The case of Bolivia is particularly interesting, because the universal pension reaches almost every older adult in the country. The first universal pension was established in 1996/97 as part of the structural reform of the contributory pension system. That benefit, called Bono Solidario (or Bonosol) was created with the aim to distribute the public shares of capitalized (i.e., partly privatized) public companies among the population. Bonosol would fulfil that aim by entitling all Bolivians who were aged 21 or older by 1995 (when those companies were privatized) to a benefit as of age 65 (Aponte 2006; Molina 2006; Müller 2009). In 2007, President Evo Morales expanded the universal pension system, replacing Bonosol with Renta Dignidad, a benefit that maintained several key features of Bonosol but was largely financed by hydrocarbons tax revenues, had a lower age of eligibility (60 instead of 65), and no cohort restriction. Coverage increased from 78 percent of older adults aged 65 and over in 2003 to 96 percent in 2015, and the yearly benefit amount rose from about US$237 to about US$564 (or US$43 per month, plus a 13th-month payment) in 2017.6 Despite that increase, the benefit value remains comparatively low. In Mexico, current old-age pension coverage rates are the outcome of a process that started with the creation of a universal pension for residents of Mexico City in 2001. This policy was followed by another noncontributory and non–means-tested benefit (70 y más), which was first established in 2007 for older adults in rural areas and small towns and was later expanded to urban areas (Flores-Castillo 2013). From 2013, the Older Adults' Pension was established as a noncontributory benefit for Mexicans aged 65 and older who have no other pension. By 2014, noncontributory pensions covered 52 percent of older adults (and paid more than 5.7 million benefits in 2015; CEPAL 2017b), well above the 26 percent coverage rate of contributory pension benefits (IDB 2017).7 In both Bolivia and Mexico, noncontributory benefits have a high coverage rate and, together with contributory pensions, cover the majority of the population (almost every older adult in Bolivia, due to the universal design, and about 73 percent of them in Mexico). However, in both cases, noncontributory benefit amounts are very modest and well below contributory benefits. This also contributes to explaining why labor force participation rates of the older population remain high in both countries (especially in Bolivia) and are substantially higher than in Argentina or Brazil (see below). Besides that, the choice of universal (or pension-tested) benefits in both countries (rather than means-tested pensions, which was the most common choice in other countries) makes them part of a particularly broad-ranging coverage expansion strategy. Means-tested poverty relief Most other countries that expanded old-age pension coverage opted for benefits targeted to older adults living in poverty or extreme poverty. These programs are means-tested, are separate from the contributory pension system, and pay low benefits. Some emerged as a part or complement of other social assistance programs, as was the case in Ecuador (Bono de Desarrollo Humano), Colombia, and Peru (Juntos) (Albornoz and Oleas 2013; Lavigne 2013; Mina Rosero 2013, 15). Those are examples of relatively wide programs, but in other countries, like El Salvador, Guatemala, and Honduras, noncontributory pensions have a more limited scope, in terms of both coverage and the resources involved (Mesa-Lago 2012a). Colombia is a good example of a coverage expansion strategy based on means-tested benefits. The contributory pension system in Colombia has long had limited coverage. A structural pension reform in 1993/94 introduced a system of individual accounts operating in parallel to a reformed public system. Despite some increase over the past decade, coverage of contributory pensions remained low, and by 2016, benefits reached only about 27 percent of older adults (IDB 2017). Over recent years, policies to expand old-age covera