Non-contributory pensions and wellbeing

Non-contributory pensions have become increasingly popular in the last decade, and according to HelpAge International, 105 countries, 78 of which are lower- or middle-income countries, currently distribute money in this way (HelpAge International 2013). Their political success owes itself partially to the fact that older people are frequently viewed as being ‘deserving’ of social protection in contrast with other groups such as the unemployed and mothers who are frequently stigmatised (Grosh et al. Reference Grosh, Del Ninno, Tesliuc and Ouerghi2008). Meanwhile, in policy-making circles, the popularity of non-contributory pensions is also due to a strong and growing evidence base suggesting that they invariably have positive impacts on wellbeing, reducing income and consumption poverty amongst recipients and often their family members (e.g. Barrientos Reference Barrientos2005; Gasparini et al. Reference Gasparini, Alejo, Haimovich, Olivieri and Tornarolli2007; Lloyd-Sherlock, Saboia and Ramírez-Rodríguez Reference Lloyd-Sherlock, Saboia and Ramírez-Rodríguez2012). In Brazil, the impact of pensions on poverty has been so significant that later life is increasingly associated with financial security (Lloyd-Sherlock, Saboia and Ramírez-Rodríguez Reference Lloyd-Sherlock, Saboia and Ramírez-Rodríguez2012).

While the impacts of non-contributory pensions on reducing income and consumption poverty are clear and substantiated by evidence, however, an additional set of claims for pensions are frequently made which have contributed to their reputation as a cure-all for development challenges. One of these is that they are frequently invested in micro-activities, leading to additional long-term income gains. This claim is based on studies like those of Martinez (Reference Martinez2004), which finds that recipients of the Renta Dignidad's predecessor, Bonosol, increased their consumption significantly above the level of the transfer itself, because they had invested in fertilisers and livestock to increase the productivity of their smallholder land plots. According to other studies, pensions can also benefit third parties such as traders, who have earned more from increased sales and an all-round business cycle (Barrientos and Sabates-Wheeler Reference Barrientos and Sabates-Wheeler2009; Bukuluki and Watson Reference Bukuluki and Watson2012; Davies and Davey Reference Davies and Davey2008; Devereux Reference Devereux2001; Ibrahim Reference Ibrahim2013; Staunton Reference Staunton2011). Models of a hypothetical non-contributory pension in Bangladesh suggest that such multiplier effects would make positive and large-scale contributions to agriculture, services, manufacturing and transport (Khondker Reference Khondker2014).

However, what is often left out of this celebratory literature is the importance of context. Whereas social pensions can be used for investment, this depends heavily on when and where they are implemented. In the case of the Bonosol, for example, the major income gains from investment were due to the existence of spare capacity in the recipients' landholdings, which had been under-invested due to lack of capital (Martinez Reference Martinez2004), but these conditions do not always hold. For example, Lloyd-Sherlock et al. (Reference Lloyd-Sherlock, Barrientos, Møller and Saboia2012a) found that while pension money in Brazil is invested in areas where recipients own their own land (and therefore possess an asset to invest in), this is not the case in a different region where recipients are not landowners. Similarly, Møller (Reference Møller2011) finds that practically none of the money from South Africa's State Old Age Grant gets invested in micro-enterprise activities, and even though pensions get invested in education, there is no guarantee that this will have the expected long-term benefits in the context of widespread structural unemployment. Regardless of the fact that pensions might in some cases be used for investment, that should not necessarily be cause for much celebration if the investments are in micro-enterprises which have limited prospects for expansion.

The second area where claims for non-contributory pensions are potentially overhyped is with regards to health. As Lloyd-Sherlock and Agrawal (Reference Lloyd-Sherlock and Agrawal2014) note, it is common for claims to be made that pensions contribute to the health of recipients (e.g. Case Reference Case2001; Schatz et al. Reference Schatz, Gómez-Olivé, Ralston, Menken and Tollman2012), based on the understanding that older people face financial barriers to accessing health care in low- and middle-income countries. For example, in Tanzania and Côte d'Ivoire, it has been found that people over the age of 50 are significantly less likely to seek out health care compared to younger people (McIntyre, cited in Lloyd-Sherlock and Agrawal Reference Lloyd-Sherlock and Agrawal2014). Perhaps unsurprisingly, therefore, research suggests that pensions often improve access to health services and lead to better self-reported health (e.g. Lloyd-Sherlock Reference Lloyd-Sherlock2006; Schwarzer and Querino Reference Schwartzer and Querino2002). However, this does not tell us much about actual health outcomes, because spending on ‘health’ may simply mean spending on inadequate and ineffective remedies, and self-reported evidence is not an accurate indicator of health status (Lloyd-Sherlock and Agrawal Reference Lloyd-Sherlock and Agrawal2014).

The third area of major claims concerns the impacts of pensions on the social relationships of recipients. There is evidence that, in some circumstances, non-contributory pensions can contribute to a range of other aspects of wellbeing, such as subjective wellbeing, social relationships and dignity. Leisering and Barrientos (Reference Leisering and Barrientos2013), for example, emphasise the positive contributions that non-contributory pensions can make to older persons' autonomy and the respect other people give them. The Bonosol, for example, has been shown to have strengthened intergenerational links and informal support networks in urban La Paz (Skinner Reference Skinner, Aponte, Jemio, Laserna, Martinez, Molina, Schulze and Skinner2007). Claims that pensions bolster social networks are frequently associated with evidence that they are shared (‘pooled’) within households so that other people can benefit from them. In Brazil, for example, numerous studies have found that pension money is commonly pooled within households and shared with younger family members (Delgado and Cardoso Reference Delgado, Cardoso and Camarano2005; Lloyd-Sherlock Reference Lloyd-Sherlock2006; Lloyd-Sherlock et al. Reference Lloyd-Sherlock, Barrientos, Møller and Saboia2012a; Møller and Ferreira Reference Møller and Ferreira2003; Schwartzer and Querino Reference Schwartzer and Querino2002). Such evidence has been employed by the non-governmental organisation HelpAge International to advocate in favour of non-contributory pensions, arguing that they generally have progressive impacts in low-income countries, reducing both poverty in later life and intergenerational poverty (HelpAge International 2014). In Bolivia, the Bonosol programme was defended partly on the basis that it was a ‘prudent investment’, benefiting not just the recipients but also their families and the broader economy (Aponte et al. Reference Aponte, Jemio, Laserna, Martinez, Molina, Schulze and Skinner2007).

Again, however, there is a danger of overhyping the impacts. In reality, the internal dynamics of older people's lives are highly varied, meaning that the outcomes of a pension on social relationships are likely to be differentiated (Lloyd-Sherlock Reference Lloyd-Sherlock2006). It is clearly too simplistic to assume that evidence of ‘pooling’ is therefore evidence of a ‘positive’ impact, because this could occur under circumstances of coercion or social pressure (Burman Reference Burman1996; Sagner and Mtati Reference Sagner and Mtati1999). More broadly, studies that focus exclusively on non-contributory pensions run the risk of selectively identifying positive statements made about the pensions without weighing them up against all the other things that are happening in the lives of older people, and thus exaggerating their impacts.

This paper assesses the impacts of the Renta Dignidad on wellbeing by focusing on four key areas: livelihood security, productive activities, health and social relationships. By identifying the contexts within which older people construct and negotiate wellbeing, as well as the ways in which the Renta Dignidad is and is not used, we demonstrate that while the transfer does contribute significantly to the livelihood security of older people, its influence on the other areas is far more nebulous and limited by a range of contextual factors. We conclude by emphasising the need for a realistic appraisal of non-contributory pensions and the importance of additional policies to improve the wellbeing of older people, particularly regarding health care.

Wellbeing and development

In order to understand the impacts of the Renta Dignidad on wellbeing in its broadest and most holistic sense, we define wellbeing as the interplay between three key dimensions: the material, the relational and the subjective (Gough and McGregor Reference Gough and McGregor2007). The material element takes into account the more conventional elements of development, thus corresponding to the focus on livelihood security, productive activities and health in this paper. Meanwhile, the inclusion of relational wellbeing, and the study of older people's social relationships, is included to reject the methodological individualism underpinning much of conventional development thinking. Analysing relational wellbeing includes the understanding that it is often influenced by conflictive processes that affect those who manage to ‘live well’ and those who do not (Deneulin and McGregor Reference Deneulin and McGregor2010). Thus, a focus on wellbeing necessarily entails understanding inter-person relationships and also trade-offs, because people's ability to live well together is likely to depend on their ability to accommodate each other's ‘systems of meaning and value’, as well as tensions between different individuals and groups who dispute these systems (Deneulin and McGregor Reference Deneulin and McGregor2010). This also means understanding why it is that particular groups of people seem to be particularly vulnerable to ‘wellbeing failures’, which occur when people are denied opportunities for wellbeing due to structural inequalities (Devereux and McGregor Reference Devereux and McGregor2014). Finally, the subjective element of wellbeing acknowledges the reality that humans' wellbeing is not simply generated by a set of objective resources, goals or processes which they are involved in, but rather with the specific meanings which they themselves attach to these (McGregor Reference McGregor2006). The three aspects of wellbeing are not, it must be emphasised, discrete spheres which are separate from one another, and it needs to be recognised that they are ‘co-evolving, interdependent and dynamically interacting’ (McGregor and Sumner Reference McGregor and Sumner2010: 106). As we will show, relational and subjective wellbeing underpin key elements of material wellbeing, which are influenced by older people's relationships with their family members, friends and communities, as well as their subjective understandings of them.

The wellbeing of older people in development processes has attracted growing interest in recent decades (e.g. Barrientos, Gorman and Heslop Reference Barrientos, Gorman and Heslop2003; Barrientos and Lloyd-Sherlock Reference Barrientos and Lloyd-Sherlock2002; Gorman Reference Gorman1995; Gorman and Heslop Reference Gorman and Heslop2002; Lloyd-Sherlock Reference Lloyd-Sherlock2000; Schröder-Butterfill Reference Schröder-Butterfill2004; Schröder-Butterfill and Marianti Reference Schröder-Butterfill and Marianti2006; Varley and Blasco Reference Varley and Blasco2003). Much of this literature responded to a perception that older people had been sidelined from the development agenda, misunderstood or ignored by key policy makers, and portrayed only in terms of their weaknesses and deficiencies (Gorman Reference Gorman1995, Reference Gorman1999).

In response to this, much research on older people has sought to make visible older people and their roles in developing countries. Much of this literature has had the clear intention of debunking myths about older people that are harmful to their interests, and which might limit the political will to respond to their needs, contributions and aspirations. Researchers have questioned assumptions that older people would necessarily be well looked after by their children, highlighting the abuse and neglect that can occur, especially in contexts of deprivation (Gorman and Heslop Reference Gorman and Heslop2002; Schröder-Butterfill Reference Schröder-Butterfill2004; Skinner Reference Skinner2006). An increasing number of researchers have challenged the assumption that older people are net recipients of resources from younger generations, arguing that in many contexts they are independent, productive and often net contributors of resources and services to children and grandchildren (Barrientos, Gorman and Heslop Reference Barrientos, Gorman and Heslop2003; Gorman and Heslop Reference Gorman and Heslop2002; Schröder-Butterfill Reference Schröder-Butterfill2004; Skinner Reference Skinner2006). More broadly, qualitative studies reveal that older people's experiences of wellbeing are highly diverse, both across and within developing countries, ranging from high levels of deprivation and neglect to increasing livelihood security and dignity (Lloyd-Sherlock Reference Lloyd-Sherlock2006; Lloyd-Sherlock and Locke Reference Lloyd-Sherlock and Locke2008; Lloyd-Sherlock, Saboia and Ramírez-Rodríguez Reference Lloyd-Sherlock, Saboia and Ramírez-Rodríguez2012; Lloyd-Sherlock et al. Reference Lloyd-Sherlock, Barrientos, Møller and Saboia2012a).

One theme that stands out in many cases, and which is highly relevant for our own study, is the importance of religious faith in influencing subjective wellbeing (Chida and Steptoe Reference Chida and Steptoe2008; Fehring, Miller and Shaw Reference Fehring, Miller and Shaw1997; Inglehart Reference Inglehart, Diener, Helliwell and Kahneman2010; von Humboldt, Leal and Pimenta Reference Humboldt, Leal and Pimenta2015). In a study of older people's wellbeing in Buenos Aires, for example, Lloyd-Sherlock and Locke (Reference Lloyd-Sherlock and Locke2008) have shown how Evangelism can provide certainty and spiritual fulfilment in later life, allowing older people to overcome grief and attain psychological resilience in the face of adversity. This paper attempts to build on these insights by revealing how Evangelism, which has been growing significantly in Bolivia in recent decades (Gill Reference Gill1990; Lazar Reference Lazar2008), influences the wellbeing of older people in the rural Bolivian Altiplano, in order to understand the potential impacts and non-impacts of a non-contributory pension. We will now present the methodology and basic information about the fieldwork sites.

Methodology and fieldwork sites

This article uses ethnographic data collected between November 2012 and September 2013 during fieldwork in two rural communities. In this time, we engaged in participant observation and semi-structured interviews with community members (from here on referred to as comunarios). Conversations were held in both Aymara, which is the mother tongue of one of the authors, and Spanish, which we both speak.

The two communities, which we will refer to as ‘Community A’ and ‘Community B’ in order to guarantee the anonymity of the people involved, are located in different municipalities in the Department of La Paz. There are 30 households in Community A and 40 in Community B, and many comunarios have double residence in urban areas. Livelihoods are based around smallholder land systems, with peasant families owning small, dispersed plots of land. Families grow potatoes (the staple crop), oca (a sweet-tasting Andean tuber), beans and barley, and use winter frosts to produce chuño (freeze-dried potato) and kaya (freeze-dried oca) for self-consumption, sale and barter. Community A has a micro-irrigation system, and a more dynamic agricultural economy, with comunarios growing and selling peas on a commercial basis. Both communities have access to electricity and drinking water, and unpaved road access to the nearby towns.

Table 1 details the older informants who contributed to this research in the two communities: five of whom we interviewed a considerable number of times, four whom we interviewed fairly frequently and four whom we interviewed occasionally over the ten months of the fieldwork. Finally, we also had occasional conversations with a small number of younger families and over 20 other people of varying ages during the course of the year. Spending a long time in the communities allowed us to interact with and observe older people in social situations and on their own, so we could see both how they related to their other family and community members as well as in individual situations when they could express themselves more freely. By interviewing younger people as well as older ones, and using participant observation, we were also able to undertake a degree of triangulation so we were not totally dependent on people's first-hand accounts of their own lives. For example, we were able to establish that some older people had sources of wealth and savings which they had neglected to discuss with us in earlier interviews. We were also able to interview people on different sides of disagreements rather than relying exclusively on the accounts of our closest informants, which allowed us to draw more nuanced portrayals of them than would have been the case if we had relied solely on their own subjective accounts.

Table 1. Informant households

Notes: 1. Spending entire days with and eating dinner with them at least once a week when we were in the relevant community. 2. Speaking with them at least once every week and spending the entire day with them at least once every two weeks. 3. Speaking with them at least once every two weeks.

Early on in the fieldwork period we attempted to talk to as broad a sample of people as possible in order to assess some general trends and consider who might be the most appropriate informants for the study. Following Spedding and Llanos (Reference Spedding and Llanos1999), we felt it was better to focus on getting more detailed data from a smaller number of respondents rather than attempting to obtain a larger sample of responses and risk getting unreliable data, although at the same time we continually aimed to interview as broad a cross-section of older people as possible. For example, we often interviewed frail older people with cognitive impairments, as well as with their carers, although we are unable to develop the rapport and understanding with them comparable to with our most key informants (Table 2).

Table 2. Basic details of key informants

The Renta Dignidad

Bolivia's non-contributory pension has its roots in the 1990s, when the government of Gonzálo Sanchez de Lozada declared that it would distribute some of the proceeds of the privatisation of public companies directly to people over the age of 65 through a social programme called Bonosol. In 2008, under the government of Evo Morales, the Bonosol was replaced with the Renta Dignidad, now financed out of the surplus generated by gas extraction and profits from renationalised public companies. The amount was raised from 1,800 Bolivianos per year with Bonosol to 2,400 with Renta Dignidad, and this was increased again in May 2013 to 3,000 Bolivianos (US $434) a year for people lacking a pension. It is worth emphasising that while the Renta Dignidad can represent a significant amount of money for poor recipients in rural areas, the amounts paid out are still modest by regional standards. While countries such as Brazil pay out non-contributory pensions of equal value to the minimum wages in those countries, in Bolivia the Renta Dignidad was equivalent to just 17 per cent of the official minimum wage at the time of writing. In addition to the increase in the value of the transfer, another key difference is that whereas Bonosol was paid out as an annual lump-sum, the Renta Dignidad can be withdrawn at any frequency as desired by the recipient, provided they do so at least once a year. We will now proceed to assess its impacts on four key areas of wellbeing: livelihood security, productive activities, health and social relationships.

Livelihood security

Across all of the households, as people get older, they become progressively less able to generate income and produce food. This happens in a number of ways. Physical weakness makes it harder for people to work, so they gradually cultivate in fewer (or smaller) land plots, thereby limiting their agricultural surpluses and ability to trade. People distribute their land to their children when they marry, such that older people often have relatively less land remaining. This is clearest in the case of the oldest interviewees (Juana, Marta, Fausto and Fortuna, Pedro and Laura), who had hardly any land left, whilst some of the younger households in the cohort (Francesca and Álvaro, Juan and Valentina) continued to dispose of significant land plots. However, increasing physical impairments meant that these last two households had a reduced ability to reach rural markets to sell and barter their products, leaving them more dependent on selling products to traders within the village at a lower price.

Older people are negatively affected by reduced ability to walk and run and increased frequency of accidents. Juana, Valentina, Francesca and Marta had all suffered from accidents sustained when looking after cows in the recent past, and most of the respondents commented on the increasing difficulty of maintaining flocks of sheep, leading them to sell off the sheep as they get older. Meanwhile Juana, Valentina and Saturnina all commented on the difficulty of continuing to knit clothes and bedcovers for sale or as gifts due to visual impairments, while Juan, Álvaro and Jose had all given up their previous part-time jobs as travelling musicians or builders, which they had used as key off-farm sources of income when younger.

In this context, the Renta Dignidad is generally allocated to five areas: foodstuffs, services (such as gas and electricity), health care, hiring tractors and costs relating to communal obligations in the sindicatos. Of these, foodstuffs are by far the most common use of the money. This observation is supported at the macro-level by data from a recent survey by the Bolivian Government's UDAPE, which shows that in rural areas, 58.1 per cent of the money is spent on food items, while 6.9 per cent is used for health-care products and 4.7 per cent for services (Chumacero Viscarra, Escobar Loza and Mendizabal Reference Chumacero Viscarra, Escobar Loza and Mendizabal2013).

The ways in which people talk about the Renta Dignidad reflect its use: the most common expression used is ‘it really saves us’, which reflects its use for essential daily spending, while at the same time recipients frequently comment that ‘it doesn't cover everything’, reflecting the fact that amounts paid out are limited and not enough, alone, to live off. By contributing to daily spending, the Renta Dignidad contributes to older persons' livelihood security, which Wood (Reference Wood2006) has argued should be given greater prominence in discussions of human wellbeing and development, because insecurity is often a defining characteristic that distinguishes the lives of the poor from those of the non-poor. Due to their livelihood insecurity and political marginality, poor people often suffer from ‘the fear which arises from not being able to control or significantly influence their immediate or longer term operational environment for survival’ (Wood Reference Wood2006 32). Thus, although the Renta Dignidad does not cover all of people's needs, it nonetheless makes a valuable contribution to meeting them.

At the same time, while the Renta Dignidad is mainly used to purchase foodstuffs, particularly non-local ones, such as pasta, rice, sugar and salt, which are consumed frequently, as well as chicken and fish, the provision of cash alone does not guarantee food security or nutrition. While obesity is not a significant problem for older people in rural areas due to their active lifestyles, they consume large amounts of sugar in their tea and salt in their soups, and this almost certainly has negative impacts, such as diabetes, which is on the rise in Bolivia (Pagina Siete 2016), and frequently goes undiagnosed. Moreover, while the combination of the money provided by the Renta Dignidad, savings and home-grown food production means that food scarcity is not a significant problem for older people, dwindling appetites and in some cases the lack of motivation to cook and eat pose threats to older people's diet. This is a psychological and emotional challenge as much as one of material resources or physical capacity. Many of the respondents, particularly Valentina and Francesca, commented that they became more motivated to eat when we went to visit them in evenings so that they could share food with us, but that in our absence they and their husbands often neglected to cook properly. Thus, while the provision of the Renta Dignidad massively increases older people's ability to purchase important complements to their diets, it does not guarantee food security or nutrition. While Bolivia lacks solid data on nutrition amongst older people,Footnote ¹ the government has recently announced a programme of providing nutritional supplements to them, suggesting a degree of recognition that cash alone is inadequate to address food security and nutrition among older people (Sinembargo 2015).

Productive activities and investment

Older people in the two communities maintain active and productive lives as far as they possibly can, in spite of declining health and physical strength, and do not attempt to ‘retire’ by ceasing work. This is not driven solely by material needs but also by ethical and cultural norms. Therefore, although the Renta Dignidad is used primarily for day-to-day consumption, it is nonetheless injected into livelihoods that continue to contain significant ‘productive’ elements.

The most obvious way in which the Renta Dignidad contributes to productive activities is through the hiring of tractors for ploughing and harrowing, which have become an indispensable activity for farmers in recent decades. Not only do peasants need to hire tractors to farm, they have to do so at very specific moments, or else they are forced to sow later in less favourable weather conditions. Juana, Lucia and Marta all noted that the regular secure income provided by the Renta Dignidad made it easier for them to hire tractors at particular times. However, at the same time they also pointed out that they had often missed out on hiring tractors at key moments due to their lack of mobility and clout vis-à-vis younger farmers.

More broadly, we found no evidence for the Renta Dignidad contributing to major increases in productivity via purchases of fertiliser and livestock as Martinez (Reference Martinez2004) finds for the Renta Dignidad's predecesor, Bonosol. This finding is also supported by the UDAPE data which show only 2.6 per cent being spent on agricultural inputs (Chumacero Viscarra, Escobar Loza and Mendizabal Reference Chumacero Viscarra, Escobar Loza and Mendizabal2013). Respondents in most of the 13 households commented that fertiliser investment is unattractive because earlier experiments led to short-term increases in productivity followed by exhaustion of the soil and the spoiling of the taste of potatoes. Meanwhile, Renta Dignidad recipients are unlikely to invest much in livestock because, as mentioned earlier, the physical impacts of later life make it increasingly hard to look after animals. Regarding multiplier effects, it is clear that the Renta Dignidad has benefited urban-based and travelling traders in small urban towns by providing them with a steady flow of income. At the same time, the multiplier effects do not accrue within rural villages themselves, because all the recipients spend the majority of money almost entirely on products purchased in towns, most of which are not produced locally (e.g. pasta, rice, etc.), while local and regional products are more commonly obtained through self-production and barter. In sum, the impacts of the Renta Dignidad on productive activities are relatively limited.

Health

In general, older people are more exposed to ill-health than other age groups (Lloyd-Sherlock Reference Lloyd-Sherlock2006), and this can have major effects on other areas of wellbeing. In the two rural communities focused on in this study, health problems play a major role in contributing to wellbeing failures among older people. In Bolivia, the vulnerability of older people to health problems and disabilities is demonstrated by a significant quantity of data collected in the survey by UDAPE that shows both are more likely among older age segments (Chumacero Viscarra, Escobar Loza and Mendizabal Reference Chumacero Viscarra, Escobar Loza and Mendizabal2013).

Older people suffer massively from muscular problems (reported by virtually all respondents), eye problems (Juana, Fortuna, Valentina, Saturnina), skin irritations (Lucia, Juana) and foot pain (Lucia, Valentina). These problems are strongly related to manual work, revealing the extreme differences between the dilemmas of older people living in rural Bolivia, and those in more affluent societies, where sedentary living patterns are increasingly seen as problematic, and where physical activity is assumed to have positive impacts on health care in later life (Landi et al. Reference Landi, Onder, Carpenter, Cesari, Soldato and Bernabei2007; Steptoe, Deaton and Stone Reference Steptoe, Deaton and Stone2015; Steptoe, Demakokos and de Oliveira Reference Steptoe, Demakokos, de Oliveira, Banks, Nazroo and Steptoe2012). Francesca, Fausto, Fortuna, Lucia and Saturnina all blamed their current health problems and physical weakness on having worked excessively when they were younger, a view which is supported by broader evidence across Latin America which shows how childhood social and health conditions influence frailty in later life (Alvarado et al. Reference Alvarado, Zunzunegui, Béland and Bamvita2008). In addition, the practice of cooking with fuelwood and manure (the most common method of cooking amongst all respondents) also causes health problems (World Health Organization 2014). While most comunarios combine gas with fuelwood and animal excrement to cook, older people generally prefer the latter, because the fuel is freely available, heats up the kitchen and the food cooked with them is said to be tastier than if it were cooked with gas. Both Juana and Marta stated that they avoided using gas altogether due to a lack of familiarity with it and a fear that the canister might explode. In sum, many key aspects of older people's lives in rural areas make them vulnerable to health problems.

To treat themselves for minor maladies, comunarios employ a mixture of homemade remedies, palliative naturista (‘non-pharmaceutical’) products purchased in the intermediate city and, in some cases, pain killers and eye drops acquired from doctors and local health clinics located 20 and 45 minutes' walk from the villages, respectively. All of the older people in the two communities are registered for free health insurance that is provided by the government for older people, but they generally do not use it due to lack of time (cited by Marta, Juana and Ilda) to go to the clinics. Meanwhile, those with more significant physical impairments (Roberta, Pedro, Fortuna and Saturnina) were unable to reach the clinics, and felt it was not worth having their carers take them there. Many also expressed scepticism over the effectiveness of the treatment provided, with Fausto, for example, telling how he had once been prescribed ineffective eye drops, and Gabriela saying that the medicines from the clinic had made her hypersensitive to cold water. Meanwhile, in Community A many of the respondents discussed a recent case where a woman's condition had been made drastically worse following an injection, generating widespread distrust in the clinic. Moreover, all of the older women except for Lara, Rufina and Mariela lacked confidence in Spanish and in engaging with state bureaucracies more broadly. As a result of these factors, there was a widespread view that the treatment provided by the clinics was inaccessible and inadequate.

In this context, the contribution of the Renta Dignidad is providing money which often gets spent on the previously mentioned non-pharmaceutical health products bought in town centres. In particular, Valentina and Francesca both reported spending a large proportion of their Renta Dignidad money on these products on a regular basis. These products serve a palliative purpose at best, and both Valentina and Francesca acknowledged that their impacts are often only temporary in nature. For more comprehensive health solutions, Francesca, Valentina, Juana and Felipa had turned to Evangelism, which they perceived to have cured them of apparently incurable maladies. This interpretation of health problems as potentially having spiritual causes and solutions is rooted in Aymara culture, where medicine and spirituality have historically been linked (Spedding Reference Spedding1994: 167).

In sum, the potential for injections of cash to make significant contributions to people's health is heavily constrained by the fact that many of the health problems suffered by older people are deeply embedded in their livelihoods and life experiences, and the inadequacy of local health services. These problems are not exclusive to Bolivia, but rather are indicative of a wider global challenge to health systems which are poorly prepared to respond to the needs of people in later life due to a combination of misplaced priorities and a lack of investment in outreach programmes (Prince et al. Reference Prince, Wu, Guo, Gutierrez Robledo, O'Donnell, Sullivan and Yusuf2015). The effectiveness of social protection for improving health depends heavily on the availability of services, which as we have seen are poor in the communities of this study, leaving older people dependent on spiritual and palliative remedies.

Social relationships

Wellbeing is strongly linked to people's quality of social relationships (Camfield, Choudhury and Devine Reference Camfield, Choudhury and Devine2009; White Reference White2010), and this has often been found to be particularly the case for people in later life (Black and Dobbs Reference Black and Dobbs2014; Heehan Reference Heehan2010; Litwin and Shiovitz-Ezra Reference Litwin and Shiovitz-Ezra2006). Evidence of non-contributory pensions suggests that they can influence social relationships in particular ways. Neves et al. (Reference Neves, Samsom, van Niekerk, Hlathswayo and du Toit2009) find instances where people use non-contributory pensions to make expenditure on events such as weddings, which allows them to maintain moral claims vis-à-vis their friends and family. These ‘investments’ can be understood both as acts motivated by feelings of love and responsibility whilst also having an instrumental value in terms of strengthening kinship networks (Barrientos and Lloyd-Sherlock Reference Barrientos and Lloyd-Sherlock2002; Du Toit and Neves Reference Du Toit and Neves2009; Sagner and Mtati Reference Sagner and Mtati1999). In Zambia, pension recipients comment on the fact that other people in the community have begun to call them ‘bosses’ as a mark of respect (Kidd Reference Kidd2010). Similarly, Leisering and Barrientos (Reference Leisering and Barrientos2013) emphasise the positive contributions that non-contributory pensions can make to older persons' autonomy and the respect other people give them. The Bonosol has been shown to have strengthened intergenerational links and informal support networks in La Paz (Skinner Reference Skinner, Aponte, Jemio, Laserna, Martinez, Molina, Schulze and Skinner2007).

The flipside to this, however, is that, as mentioned earlier, pension recipients may be pressurised into giving or lending money to other family members, in what is known as ‘coercive pooling’, which as we have seen has been found in some cases in South Africa (Burman Reference Burman1996; Sagner and Mtati Reference Sagner and Mtati1999), and in urban Bolivia (Skinner Reference Skinner, Aponte, Jemio, Laserna, Martinez, Molina, Schulze and Skinner2007). The existence of coercive pooling draws attention to the reality, revealed by Lloyd-Sherlock and Locke (Reference Lloyd-Sherlock and Locke2008), that while older persons' relationships with family members can be sources of support and pride, they can also generate disappointment and resentment, and convert into sources of wellbeing failure. Feminist literature in particular raises questions about the nature of relationships between spouses, which are often characterised by asymmetries of power which then manifest in forms of violence directed against women (Chant Reference Chant1998, Reference Chant2002; Kabeer and Joekes Reference Kabeer and Joekes1991). While asymmetries of power within families are acknowledged in the literature on older people, references to abuse or maltreatment of older people focus almost exclusively on the role of younger people as perpetrators of abuse, neglect or disappointment (e.g. Barrientos, Gorman and Heslop Reference Barrientos, Gorman and Heslop2003; Castro Mantilla Reference Castro Mantilla, Salazar de la Torre, Castro Mantilla and Medinacelli Monrroy2011; Gorman and Heslop Reference Gorman and Heslop2002; James Reference James1992; Lloyd-Sherlock and Locke Reference Lloyd-Sherlock and Locke2008; Oddone Reference Oddone2001; Salazar de la Torre Reference Salazar de la Torre, Salazar de la Torre, Castro Mantilla and Medinacelli Monrroy2011), missing the possibility that other older people could also be perpetrators of abuse.

Adjusting to later life is a challenge for older people, who have to adapt to the impacts it has not only on their own body but also on their spouses, who may experience it at a different rate or intensity, generating bitter quarrels amongst many of the older spouses in the study (Valentina and Juan, Fausto and Fortuna, Francesca and Álvaro, Lucia and Emilio, Rufina and Octavio). This reveals how people's bad health in later life can have negative impacts on their relations with family members who become frustrated with them, as has been found in other contexts (Mudege and Ezeh Reference Mudege and Ezeh2009; Schatz Reference Schatz2007). The following statement by Francesca is typical:

Physical violence and bad behaviour is frequently associated with drinking alcohol, which is a central aspect of peasant culture. Although the ethnographic data collected in this study support assertions that domestic violence is mainly a feature of young marriages and tends to reduce as people get older (Spedding Reference Spedding1994), any transition is uneven and it should not be assumed that domestic violence ends just because people reach 60 years of age. In one incident in Community A, when a drunk man in his late fifties hit his wife during a fiesta, it was treated as an embarrassment, but he was not vilified. During the same fiesta, Valentina told us that she stays away from Juan when he drinks, because ‘he can be harsh’. Even when physical violence has ceased altogether, there may still be an uncomfortable legacy of it in older persons' households, which in some cases women talked about explicitly to us. Juana, Lucia, Valentina and Fortuna, for example, complained openly about the violence they used to suffer at the hands of their husbands. Some spouses respond negatively to the physical debilitation of their spouses as they become more frail, for example in the case of Fortuna, who told us that Fausto often insulted her on the basis of her blindness. Although this goes both ways (i.e. women scold men as well), overall the testimonies for women were far worse, with Fortuna and Francesca occasionally expressing a desire to die to end the suffering of their physical debilitation and the harshness of their husbands. One does not have to take these statements literally to appreciate the severe emotional pain these women were undergoing.

In addition to curing themselves physically, both Valentina and Francesca suggested that conversion to Evangelism was potentially one way of making their husbands quit alcohol consumption, which unlike in the Catholic church or ‘traditional’ peasant culture, is prohibited in Evangelism. Evangelical Christianity has often been identified as a female-led effort to instigate a transition away from cultural practices that are associated with male drinking and domestic violence (Brusco Reference Brusco, Garrard Burnett and Stoll1993; Drogus Reference Drogus, Cleary and Stewart-Gambino1998; Gill Reference Gill1990; Lazar Reference Lazar2008). The link between spirituality and the prohibition of behaviour that is not desirable for women is noticeable in the following statement by Francesca, commenting on the efforts of Valentina to convince her husband to convert, and in response to finding out that her own husband had been involved in a drunken fight:

The Evangelical challenge to traditional structures and practices creates tensions within households, because many men resist conversion, often stating how much they look forward to fiestas as an opportunity to drink beer, and they complain about the rise of Evangelism in the community and its opposition to traditional customs.

In contrast to inter-spousal conflicts, intergenerational disputes and issues relating to the abandonment, exploitation or neglect by younger generations which have been strongly emphasised in the literature (Castro Mantilla Reference Castro Mantilla, Salazar de la Torre, Castro Mantilla and Medinacelli Monrroy2011; Salazar de la Torre Reference Salazar de la Torre, Salazar de la Torre, Castro Mantilla and Medinacelli Monrroy2011) are not as significant. Those disputes that do occur between generations are of a different nature, with both Marta and Valentina commenting that they had had major arguments with their children who were attempting to force them to stop working, e.g. by buying them expensive clothing and inviting them to live in the city. Meanwhile, both Juana and Saturnina had fraught relationships with their (female) carers who struggled to deal with the debilitating impacts of later life, leading to quarrels and backbiting.

The other kinds of conflicts that occur are intra-family disputes over land, some of which are decades-long. These are generally played out between siblings and cousins, rather than between children and their parents, but can be intractable and contribute significantly to wellbeing failures. Lucía, in particular, had suffered throughout her life because her brother had seized land she thought should have been hers, to the extent that they had not spoken until recently. Similarly, Rufina, who, had given up her baby to be looked after by her sister, told us that the sister had then proceeded to take her land (unfairly in Rufina's view) as some form of compensation for looking after the child. However, the most severe land dispute during the fieldwork period was between Mariela and Raul and Mariela's nieces, who were now living outside the community and accused her of seizing land that was theirs. The dispute, which had roots going back to the 1970s, flared up during a land-titling programme in Community B. Mariela had a reputation among comunarios of acting aggressively in protecting her interests and was regarded with suspicion by many. The accusation against her was that she had expanded her house over land owned by her now-deceased brother, thereby depriving her nieces who had been living in different communities for some time. The dispute quickly descended into a war of insults traded by both sides, many of them deemed excessively personal and vulgar by other comunarios. For example, Mariela argued that it was in fact her own father and not the nieces' father who had brought them up, and that therefore she had a more legitimate claim to the lands. A community meeting called to resolve the issue lasted all day but did not resolve the issue, although Mariela was fined for insulting the sindicato leaders, and both parties were obliged to sign an agreement to cease trading insults or be whipped. The affair was highly polarised, leaving Mariela feeling that some key family members had failed to support her.

Given the severity of many older people's wellbeing failures due to problematic relationships with their spouses and family relatives, the potential of the Renta Dignidad to mitigate them significantly is limited. On the one hand, older women in relatively better-off households, such as Valentina and Francesca, have more financial autonomy than those in poorer ones, such as Lucia, allowing them to offer gifts and loans to their children and grandchildren without requiring permission from their husbands. However, it is also clear that such tensions cannot be reduced to a lack of income and do not have a straightforward cash-based solution. The legacy of abusive relationships, unequal access to education and off-farm opportunities, combined with the effects of later life on health and physical strength, mean that women generally seem to be more likely to suffer from wellbeing failures than men. In particular, older people's wellbeing is often affected by the conflictive side of their relationships with their spouses or other family members residing with them. Because wellbeing failures are usually generated by long-standing historical inequalities and ingrained behaviour patterns, as well as the difficulties older people have in agreeing with their relatives about the ways in which wellbeing should be achieved, the potential for the Renta Dignidad to contribute significantly to overcoming them is limited. The types of behavioural changes required to prevent these practices cannot be brought about by pensions, but rather through a substantial cultural change through improved education, greater awareness of alternative values and the enforcement of national laws against domestic violence.