Introduction

In the last two decades, child-sensitive social protection (CSSP) became a widely adopted strategy to address poverty, based on the premise that interventions during childhood could interrupt cycles of intergenerational reproduction of poverty (Barrientos and DeJong 2006; Minujin et al. 2017). The appeal for CSSP was supported by research demonstrating the lasting impact of monetary and food deprivation on child development, presenting childhood as a crucial juncture for investments in future human capital (Heckman 2000; Jones and Sumner 2011). Since CSSP aims to ‘maximise positive impacts on children as well as minimise any unintended harm’ (Roelen 2021, p 369), it can encompass any programme that relieves the economic burden of parents to allow poor families to invest in the future of their children. In its most popular form, CSSP incentivises children’s access to educational and health services through programmes such as school meals and conditional cash transfers (CCTs) (Roelen and Sabates-Wheeler 2012).

Social protection programmes targeted particularly rural families, who are overrepresented among the global poor (Fotta and Balen 2018; Hanlon et al. 2010). These programmes’ attention to household dynamics allowed them to reach children at an unprecedented scale, but also led to an oversight of the structural drivers of families’ economic vulnerability (Barrientos and DeJong 2006; Devereux and McGregor 2014). Many well-known drivers of rural poverty—such as quality of soil and water, exposure to disasters and health hazards—are connected to the overarching political issue of land use and distribution (Devereux 2001; Scoones 2021). Yet, the transformation of these structures that promote inequality remains far from the objectives of most social protection programmes (Sabates-Wheeler and Devereux 2008).

In Latin America, a key driver of rural poverty is the expansion of extractive industries (Bebbington and Bury 2013). These activities not only fuel dispossession but tend to worsen the quality of resources available for peasant families (Bebbington and Bury 2013; Robins and Fraser 2020). Nonetheless, governments across the region have relied heavily on extractive revenues to fund social programmes, often disregarding the adverse impact of these industries on the same communities that social protection programmes aim to assist (Lang 2017; Riggirozzi 2020). This study examines this contradiction through a case study of a village in Peruvian Amazonia where social protection programmes coexist with oil extraction.

In contrast to the dominant ‘top-down production of evidence’ (Jones and Sumner 2011, p 6) that informs CSSP, this research centres Indigenous perspectives on and experiences of intergenerational impoverishment and social protection. The methodology involved eight months of participant observation triangulated by interviews and participatory methods, and all findings were validated after data analysis to ensure that they accurately represented the views of the community. Overall, this study underscores the problem of viewing intergenerational impoverishment as an individual behaviour problem when resource scarcity is rooted in environmental degradation. It also shows that dismissing the loss of environmental capital can undermine CSSP’s efforts to address the intergenerational reproduction of poverty.

Although the importance of environmental capital for sustainable rural economies is widely recognised (Devereux 2001; Scoones 2021), CSSP programmes often fail to recognise how the loss of this capital ‘compromises the ability of future generations to meet their needs’ (World Commission on Environment and Development 1987, p 41). Governments and development agencies continue to think of social protection as economic safety nets, rather than mechanisms to promote radical social transformation (Sabates-Wheeler and Devereux 2008, p 65). Integrating an environmental justice lens into these debates is crucial to shifting this mindset. By linking present resource overuse with future scarcity, environmental justice contributes to highlighting the contradictions of a development model focussed solely on economic growth to the detriment of environmental sustainability (Hiskes 2008).

Environmental justice also contributes to foregrounding environmental capital in the analysis of intergenerational impoverishment. The literature shows that resource loss due to overuse and contamination disproportionately affects already marginalised populations, such as income-poor families and peasant communities (Bullard et al 2003; Martinez-Alier 2014). Moreover, Stephens (1994, p 4) aptly noted that children are like ‘canaries in the mines’ who suffer with environmental contamination long before, and at a much greater scale than, adults. Similarly to monetary deprivation, the degradation of food and water sources can have lasting detrimental effects on children’s cognitive development (Satterthwaite 1996). Therefore, the lack of attention to environmental sustainability could end up hindering CSSP efforts to break cycles of poverty reproduction. This reflection is essential for questioning the purpose of social protection, and whether the format of existing programmes can foster social justice (Devereux and McGregor 2014; Hickey 2014; Sabates-Wheeler and Devereux 2008).

The article will start by elucidating the intricate ties between CSSP and environmental justice in Amazonia, to establish the value of this case study for broader reconsiderations of the purpose of social protection. This will be followed by a methodology section that includes a contextualisation of the studied village and all aspects of data collection. Finally, the findings will describe families’ perceptions of CSSP in a village that grapples with oil pollution amidst a changing climate. This study will then examine the environmental causes of resource scarcity in rural Amazonia to establish environmental justice as a critical lens of analysis of intergenerational poverty.

Social Protection and Environmental Justice in Latin America

The links between CSSP and environmental justice are deep-rooted and extend beyond their shared interest in younger generations. During the period between 2000 and 2014, the expansion of social protection networks in Latin America was closely tied to the economic growth propelled by the commodities boom (Balakrishnan and Toscani 2018). As major exporters, countries in the region experienced a prosperous decade marked by a substantial rise in government income. This economic bonanza, driven by revenues from extractive royalties and taxation of consumption and export, empowered national governments to invest in social programmes (Sánchez-Ancochea 2021). At the time, more than half of the region’s social spending was allocated to cash transfers (Holland and Schneider 2017).

The results of these investments were overwhelmingly positive, as monetary poverty in the region plummeted from 27 to 12% in this period (Balakrishnan and Toscani 2018). This success contributed to building a consensus that extractive development was the route towards more equitable national development, as it allowed governments to fund mass socio-economic inclusion (Svampa 2015; Gudynas 2016). However, it also entrapped governments in a fiscal and political dependency of extractive revenues (Riggirozzi 2020, p 516). This economic model accelerated the pace of resource exploitation and, consequently, the dispossession of peasant families (Arsel et al. 2016). The negative impacts of extractive industries were disproportionately felt by marginalised rural communities, such as Indigenous peoples (Bebbington and Humphreys Bebbington 2011; Svampa et al. 2021).

Poverty and Extraction in the Peruvian Amazonia

The complicated relationship between extraction and social protection is especially evident in the Amazon, a region with historically high poverty rates and significant oil reserves. When commodity prices peaked, the region experienced its own resource boom (Orta-Martínez et al. 2018). A large portion of the forest was allocated for the extraction of hydrocarbons, iron ore and gold, or the large-scale production of soybeans and cattle (Merino 2024). These activities often took place inside or near Indigenous land, threatening the sustainability of these communities (Robins and Fraser 2020). In the Peruvian Amazonia, where this research took place, 80% of the land conceded for oil and gas exploration overlapped with a demarcated or proposed Indigenous territory (Orta-Martínez and Finer 2010).

Indigenous peoples had limited power over this resource plunder. Although their right to land was legally guaranteed, the law also asserted the State’s prerogative to pursue profitable activities in the remainder of the forest (see, for instance, Peruvian Government 1978, Art. 30 and 31). In practice, the law effectively dispossessed Indigenous families of vast swaths of land, limiting their ability to maintain themselves as their populations grew (Chirif and García Hierro 2007). The impoverishing effects of this kind of development are evident in the autobiography of Agustina Valera, a Shipibo woman:

Before we used to find animals nearby and in abundance … Before we did not know the food from [non-Indigenous] people. Now we barely see things from the forest because the White people have killed all our animals (Valenzuela Bismarck and Rojas 2004, p 213—author’s translation)

The context of dispossession was aggravated by the resource contamination that ensued from extractive activities. In the period between 2000 and 2019, Peruvian Amazonia had 474 oil spills that exposed 61% of Indigenous communities to toxic substances like lead, mercury, and cadmium (León and Zúñiga 2020). These contaminants cannot be easily cleaned up, as they infiltrate food and water sources and cause health issues ranging from skin rashes to permanent cognitive damage, particularly among children (PUINAMUDT and Campanario 2016; Whitehead and Buchanan 2019). Subsequently, and contrary to the State’s linkage of extractive revenues with social gains, Indigenous communities found themselves grappling with an incongruency: the supposed economic benefits of oil extraction paled in comparison to the high socio-environmental costs they experienced. This led to heightened conflicts between Indigenous peoples and the State (Fraser 2020; Svampa et al. 2021).

Because Amazonia is a resource-rich territory, different governments across the region have attempted to frame Indigenous opposition to extractive development as a form of ‘backward’ thinking that needed to be corrected. In Ecuador, president Rafael Correa (2007–2017) argued that Amazonian countries ‘cannot be beggars sitting in a sack of gold’ (cited in Lang 2017, p 80). Similarly, in Peru, president Alan García (2006–2011) likened native communities to the fable of the dog in the manger, accusing Indigenous peoples of impeding national development for lacking the ability to profit from Amazonia’s valuable resources.1 By conflating extractivism with poverty eradication, these governments obscured the role that resource extraction plays in impoverishing rural communities by degrading vital resources (Gudynas 2019). The classification of Indigenous peoples as poor was also instrumental in framing their assimilation into the market economy as a matter of social justice.

Child-Sensitive Social Protection in the Peruvian Amazonia

In Peru, the creation of CCT Juntos, in 2005, and the school-feeding programme Qali Warma, in 2013, happened amid disputes over oil exploration in Amazonia. The assessment of multidimensional poverty conducted by the government, based on families’ access to goods and services, concluded that Indigenous peoples were in a state of severe deprivation (Correa Aste et al. 2018). Despite this, the programmes were not immediately welcomed by Indigenous communities, who watched with distrust the arrival of government staff offering them money (see Sarmiento Barletti 2015). In some contexts, parents suspected that the sudden government interest in their children could conceal harmful intentions against their communities (Santos-Granero and Barclay 2011). Indigenous families also worried that government benefits were an inducement to facilitate the advancement of resource extraction into their territories (Correa Aste et al. 2018, p 164). These concerns align with Gudynas’ (2016, pp. 107–108) analysis of CCT programmes as a form of compensation to reduce the social tension in communities affected by extractive activities.

However, even after the proliferation of CSSP, the realities of children in Peruvian Amazonia remained challenging. In 2019, only 35.9% of students completed their secondary education and 59% of children still suffered from anaemia (INEI 2019). Research suggests a link between environmental degradation and these poor child indicators. For instance, Espinosa and Ruiz (2017, p 32) argue that climate change aggravates school evasion because families who lose crops to extreme floods need their children’s work to recover from this shock. Likewise, studies of malnutrition in Amazonia relate the persistence of child anaemia to declining fish stocks that are causing a shift in local food systems in favour of chicken farming, and exacerbating iron deficiencies (Heilpern et al. 2021). Despite this evidence, environmental justice issues remain largely neglected by current social programmes.

Over the past two decades, existing models of social protection have faced criticism for addressing monetary vulnerabilities instead of transforming the unequal structures that cause them (Hickey 2014; Sabates-Wheeler and Devereux 2008). The case study of Peruvian Amazonia adds to this debate by underscoring the limited understanding of intergenerational poverty dynamics within current forms of CSSP.

Methodology

Overview of the Field Site

This research was conducted in a Shipibo village with 600 residents on a tributary of the Ucayali River in Peruvian Amazonia (Map 1). The choice of this field site was motivated by the presence of amenities such as a medical unit, a water well, and three school levels (early childhood, primary, and secondary). These features are rare in native communities but were pivotal in ensuring my safety as a solo researcher conducting extended fieldwork.2 Additionally, they likely provided access to a wider sample, since students could complete their education within the village.

Map 1
figure 1

Location of Shipibo native communities in Ucayali, Source: author3

Full size image

While this village is not situated over an active oil drilling site, it is exposed to water contaminants stemming from abandoned oil fields in the Ucayali basin (Cépeda and Lossio 2021). A government report from 2018 identified that lead levels in the river surpassed ten times the acceptable limits for human consumption, affecting the quantity and quality of available fish and game (National Authority of Water 2018). This, in turn, impacts the ability of families to make a living from the forest. The environmental degradation of Ucayali is compounded by other economic activities such as illegal mining, commercial plantations, and timber extraction, which have caused the region to have the worst deforestation rates in the country (Ramírez et al. 2022). In addition to worsening climate change at a global level, deforestation has intensified the scale and duration of seasonal floods in Amazonia (Barichivich et al. 2018). This has had a profound impact on the Shipibo, as 90% of their villages are annually inundated during the rainy season (INEI 2017).

Consequently, the Shipibo community has grown increasingly dependent on the market economy. This reliance has led them to look for alternative sources of income, such as selling handicrafts, offering ethno-tourism services, or seeking waged job opportunities elsewhere through seasonal migration (Sherman et al. 2016). Villages have also altered their crop production, with plantain becoming the primary harvest due to their fast growth in wet terrains and easy marketability (Collado Panduro 2021). However, this agricultural change has reduced the diversity of local diets, which are now frequently supplemented with Qali Warma food parcels. Given the modest household earnings, Juntos’ bi-monthly payment of s/200 (US$51), though less than 10% of the national minimum wage, can easily become a family’s main source of income.

Methods

Data Collection

The main method of data collection for this research was participant observation, which can be explained as ‘a production of knowledge by being and action’ (Shah 2017, p 45). The main fieldwork for this project was conducted by the author from July 2019 to March 2020, when the project was abruptly interrupted by the first outbreak of the COVID-19 pandemic. During this time, I joined 73 children at the primary school (with ages ranging from 6 to 16) and regularly visited families in the village in the afternoon. The observations of school and family dynamics were supplemented by unstructured interviews that offered a deeper understanding of events (Hockey and Forsey 2012). I also conducted semi-structured interviews with representatives at the local Indigenous organisation and a focus group with Shipibo teachers. In such cases, achieving a balanced gender representation was challenging due to the prevalence of men in power positions, but efforts were made to sample participants based on their roles and experiences.

To ensure the credibility of my findings, I later organised a second round of data collection conducted by two Indigenous research assistants in Shipibo language. This multiple triangulation—the variation of informants, investigators, methods and language of data collection—was done to validate and challenge my original fieldnotes (Laws et al. 2013). Given the limited time and budget for this research, I developed a toolkit indicating the methods and questions that should be used in each section. We opted for using drawing with children due to the success of this activity as a method of inquiry in my former fieldwork (de Carvalho 2021). Table 1 gives an overview of the samples and methods for this second round of data collection, which took place in July 2021:

Table 1 Overview of samples and methods of the second fieldwork
Full size table

Given the specific challenges of doing research with children, the research toolkit indicated a specific order of exercises that should be pursued to make children more comfortable. The order was based on the fact that group-based work can empower children to speak up and reduce age-based power imbalances (Boyden and Ennew 1997); however, recordings of in-depth interviews in large groups are generally incomprehensible. Individual interviews also facilitate the observation of children’s creative processes, which is essential to the interpretation of drawings, and are more respectful of privacy around sensitive topics. For the collective exercise, we used collaborative mapping to study children’s thoughts about the environment (Alerby 2000). This article also draw insights from the prompt ‘draw a child with a good/bad life’, used during individual interviews to discuss children’s perception of well-being (Camfield 2010).

Ethical Aspects and Limitations

Before each stage of data collection, we obtained verbal informed consent from all participants in communal assemblies with the support of the village chief. Consent from children was initially obtained at the school and, later, during individual interactions. All researchers received age-sensitive training to note and respect silent signs of distress or withdrawal during research activities (e.g. when a child erased a drawing). To protect the anonymity of my informants, all names have been replaced with pseudonyms in this article. Additionally, I have avoided any disclosure of the village’s location, respecting the safety concerns of the community.

As a non-Indigenous researcher, I am aware that my identity and background may have influenced how families responded to my questions and how I interpreted and presented their data. To ensure methodological rigour, I made changes in the research team and language of inquiry during the second fieldwork, as described above. The data recorded by the two research assistants was transcribed and translated by a third collaborator—a Shipibo linguist—to ensure an unbiased report of parents’ views. Furthermore, in August 2022, we conducted a validation workshop in the village. This not only fostered a sense of transparency about how we were using people’s data but also warranted that research findings were accurately representing the community’s perspective.

Findings

This section compiles the results of both rounds of fieldwork by themes. First, it describes participants’ perceptions and experiences of CSSP programmes. Then, it considers how environmental injustice is linked to impoverishment in rural contexts.

Perceptions of CSSP Programmes

From August 2019 to March 2020, 55 out of the 73 primary school children in the Shipibo community where I conducted fieldwork were enrolled in the Juntos programme, and students of all ages received breakfast and lunch from Qali Warma during weekdays. Parents seemed to embrace the pervasive presence of social protection in their children’s lives and complained if a teacher’s absence from school prevented students from receiving meals, or if Juntos’ payments were delayed. However, when asked if there were any projects in their village to help children, 8 out of 11 parents replied negatively. Even when the question was reformulated to inquire directly about the usefulness of Juntos and Qali Warma, their answers did not change. The words of Rosa, a mother of seven, may help explain this paradox: ‘There is nothing wrong with feeding children, but when they finish school, they will have nothing’.

The reaction of parents was distinct from other studies that documented a rise in aspirations among CSSP recipients (Jones 2022). But this scepticism was rooted in a discrepancy between programmes’ goals and their children’s needs. Both Juntos and Qali Warma assumed that improving children’s school attendance would lead to improvements in their cognitive capacity and, consequently, in their human capital (Vásquez 2020). Nonetheless, parents argued that incentives for school attendance were pointless given ‘the lack of State support to excel in education’ (Nathan). Instead of paving the way for children to achieve new levels of education, CSSP seemed to be focussing on a task that parents felt capable of doing, as argued by Lizzy: ‘I can support my children. If they want to study, I can make them study as it is’. Believing there were more effective ways to improve children’s education, Jacob questioned: ‘If they have all this money, why don’t they pay for at least one of our children to get a good education?’. These parental concerns echo the findings of programme evaluations indicating that CCTs effectively increase children’s school attendance but have minimal impact on their educational achievements (Baird et al. 2013; Gaentzsch 2020).

The perception that programmes were poorly aimed was exacerbated by misleading messages during Juntos’ monitoring. An incident at the local primary school can illustrate how easily the actual assessment of conditionalities could deviate from programme guidelines. Instead of adhering to Juntos’ rules, the front-line officer who was visiting the school chose to inspect what he perceived as indicative of misuse of CCT funds: the quality and cleanliness of students’ uniforms and school supplies. Students failing to meet his invented conditionalities were warned of the risk of losing the benefit. This punitive form of monitoring, which itself was contrary to programme rules, seemed to convey that adopting new consumption habits, resembling those of children in an urban school, was the utmost goal of the CCT.

Fotta and Balen (2018) have contended that CCTs reinforce the discrimination against rural lifestyles by suggesting that the inclusion of recipients in the market economy is a positive and desirable change. This stigmatisation is often symbolised by the enforcement of what Cookson (2018) calls shadow conditions. This phenomenon was not exclusive to this village, as there is substantial evidence of it occurring elsewhere (Correa Aste et al. 2018, p 172; Cookson 2018, p 128, Streuli 2010). Shadow conditions create confusion about the purposes of programmes, especially since they can be monitored more frequently than official rules (Escobal and Benites 2012). Moreover, they can express discrimination against Indigenous recipients (Ramírez 2021). As a father once told me, the behaviour of front-line officers during Juntos’ monitoring visits was illustrative of ‘how Peruvians treat forest peoples’ (José).

A similar message seemed to be present in the Qali Warma programme. Its attempt to replace the local diet with a less nutritious alternative suggested that Indigenous eating habits were inadequate for children (see Ricaud Oneto 2019). Ricardo, then president of a large Indigenous organisation in the Ucayali, felt offended by the main premise of the programme:

Nobody says that two cups of masato [fermented manioc beverage], two cups of chapo [sweet plantain juice] and three carachama [Amazonian catfish] with plantain is wealth. But the children who eat this every day are not hungry. They are strong and healthy. Why is this not enough? We know that our children were strong. Now that children eat food from stores, we have more sickness. They should change the formula that measures poverty. What generates poverty is the big industries. (Interview on 5 August 2019)

At the heart of Ricardo’s discontent with CSSP lay a fundamental difference in the understanding of the root causes of poverty. From his perspective, the impoverishment of Shipibo families was intricately tied to the depletion of food resources within their territory, caused by the ‘big industries’ operating in Amazonia. Similar perspectives on poverty have been reported by other Indigenous peoples in the region (see Lang 2017, p 84; Sarmiento Barletti 2015). In contrast, CSSP assumed that malnutrition and poverty were related to a cultural deficit that could be addressed with incentives for new consumption patterns.

The same individualistic view of poverty led CSSP programmes to assume that recipients would achieve self-reliance once their children reached the age of 19 or completed secondary school. While this rule may be logical from an administrative perspective, it was out of touch with the realities of Indigenous recipients, whose livelihoods were unlikely to improve with time. After years of relying on the CCT as a source of income, families would find themselves without any support to face their unchanged circumstances. As expressed by Jessica, a mother: ‘after [children] finish school, they have nowhere else to study, and no [Juntos] money either’. This pessimistic view of their children’s future arose from an awareness of the deteriorating conditions in their village, as summarised by Eduardo: ‘the future is bad, the world is barren and not like before’. To unpack this idea of a ‘barren world’ for children, the next section examines the links between environmental degradation and intergenerational impoverishment.

Environmental Degradation and Rural Impoverishment

Indigenous families who rely on hunting, fishing, or subsistence agriculture often define ‘poverty’ as a scarcity of vital resources (Lang 2017, p 84; Sarmiento Barletti 2015). Environmental risks are another well-known contributory cause of rural poverty, as they increase the likelihood of economic shocks (Devereux 2001). The expansion of extractive industries in Amazonia is known to have reduced the availability of resources Indigenous families and their resilience against environmental risks, increasing their dependency on cash (Arsel et al. 2019; Fraser 2020). This cycle of immiseration can be attributed to (i) the pollution of water resources caused by remnants of hydrocarbon extraction and heightened boat traffic; and (ii) the loss of local food sources due to extreme weather events resulting from climate change and deforestation. Both these phenomena affect Shipibo territories, as detailed below.

Water Pollution

Given their proximity to the river, Shipibo families are particularly exposed to water pollution. As mentioned in an earlier section, a government assessment has detected high levels of heavy metals in the river (National Authority of Water 2018), which impacts families’ health and livelihood strategies. All participants in this research acknowledged this problem. During a mapping workshop, all groups of children drew pictures of the village’s riverbank and placed post-its over it saying ‘Today we don’t drink water from the river anymore. It makes us sick’ or ‘There is a lot of trash [in the water]. Lots of plastic and motor liquid’.

The economic repercussions of this pollution can be illustrated by an event that took place in the second week of January 2020, when dark mud with a foul smell appeared in the river after a flood. At first, fish became easier to catch as they swam with their heads partially out of the water, and some even floated dead on the river’s surface. Toddlers and small children went playing with bows, arrows, and spurs to practice their skills, catching abundant fish without needing a canoe. Children would throw fish back into the water if they were already dead but would otherwise take their catch home to eat. Nonetheless, over the next few days, the local nurse would caution mothers that the colour and smell of the water were indicative of heavy metals in the river. People then stopped eating that fish, opting instead for preparing chicken, when they had it, or the canned meat provided by Qali Warma.

The toxic mud incident exemplified how environmental degradation drives families into financial hardship. In this scenario, ‘poverty’ did not stem from a lack of financial resources, but rather from losing the village’s primary food source due to heavy metal contamination. The lack of transparent information about abandoned oil and gas sites obstructs efforts to prepare families to cope with resource contamination more appropriately (Cépeda and Lossio 2021, p 45), even though crude is an important source of lead pollution in soil, water, and wildlife throughout Amazonia (Cartró-Sabaté et al. 2019; Guzmán-Gallegos 2019). While impacted families can resort to food and cash transfer programmes as reactive interventions (Devereux 2001, p 515), the absence of supplementary actions to address the root causes of food scarcity suggests that recipients may be left economically vulnerable once they lose access to these programmes.

Extreme Weather Events

Another critical environmental challenge faced by Shipibo families is the frequency of extreme weather events on their land. While this is an expected consequence of climate change, what is often overlooked is how extractive development directly contributed to aggravate this issue (Hughes 2013). Apart from oil being a key source of global carbon emissions, the expansion of extractive activities in the Amazon led to a loss of the forest cover that protected local communities from disasters (Cepek 2012; Nobre et al. 2016).

Although seasonal inundations are common in the Amazonian floodplains, changes in global weather patterns are exacerbating the severity of these events, and posing a threat to livelihood security in affected Indigenous communities (Sherman et al. 2016). The repeated occurrence of disasters that destroyed local food sources has increased families’ dependence on cash, particularly during the rainy season. However, families have little means of generating income within their territories (Torres-Vitolas et al. 2019). This prompts concern about the long-term sustainability of climate-affected rural communities, as expressed by Billy, an adolescent father: ‘I worry a bit because of the economy. And everything economically, the floods… Our parents do not have work and that’s why we all suffer’.

The increasing unpredictability and scale of floods threat local subsistence agriculture, rendering the cultivation of less water-resistant crops—such as cassava—unviable. It also makes hunting and fishing harder, as animals tend to move further into the forest, while families depend on canoes for basic locomotion. This precarity motivates many parents to migrate in search of waged jobs, to ensure their children can purchase food at the local shop. The extent of this issue is such that Collado Panduro (2021), who did a comparative study in four Shipibo communities, noted that 75% of households had at least one relative living elsewhere.

This pattern of parental migration has immediate consequences for children (see de Carvalho 2024). As described by Freddy (a father): ‘There are many children that were abandoned by their parents, that live with their grandparents, and these children don’t live well in their homes’. Apart from the emotional toll of family separation, children who stayed in the village described assuming a heavier workload to support their household subsistence. Andrea, a 12-year-old girl, explained that when a child lives alone with their grandparents ‘they must take care of the fields all alone with their siblings’. Some adolescents also migrated with their parents to contribute to supporting younger siblings and older kin. These strategies of adaptation can potentially motivate school evasion and children’s engagement in hazardous labour, directly hindering the achievement of two crucial objectives in existing CSSP programmes.

As argued by Espinosa and Ruiz (2017, p 32), policymakers often interpret the persistence of high numbers of school dropouts and working children in Peruvian Amazonia as a result of the ‘economic difficulties’ faced by Indigenous families. This simplified vision of the problem is often used to justify the importance of cash and food transfers as a form of ‘economic protection’ (Sabates-Wheeler and Devereux 2008, p 65). There is some value in this approach, as CSSP programmes certainly reduced the toll of losing local food sources. However, the programmes did not prevent adolescent engagement in hazardous labour or children’s heavier subsistence workload.

CSSP programmes operate under the assumption that reducing the financial stress experienced by parents can improve the choices they make for children, enabling the next generation to study and thrive (Vásquez 2020). Notwithstanding, this premise may inadvertently lead policymakers to overlook how environmental vulnerabilities restrict parental decisions. That is why Gilson, a Shipibo teacher, stressed the importance of observing the broader context of a village before judging the conditions in which children are raised:

The first thing I do [when working in a new village] is see the context: where and how the population lives, and in which conditions the children are living. In many cases parents have left them, and children are living with their grandparents … If the village doesn’t have fish, there will be none for them either. If we don’t buy food, we will sleep with our belly making noises. (Focus group on 27 February 2020).

To break this cycle of impoverishment in Amazonia, governments should aim to transform the contextual factors that render Indigenous livelihoods unsustainable (Sabates-Wheeler and Devereux 2013). This raises the critical question of what this transformation would entail. One possible approach is proposed by Nathan, a father, when asked about his aspirations for his children: ‘I would like us to be able to reforest, to still have our fishes and animals, and our medicinal plants to at least teach children that’.

Discussion

This study highlights the limitations of CSSP as a strategy to reduce intergenerational poverty in contexts where environmental degradation heightens the risk of impoverishment for younger generations. Currently, CSSP is rooted in the notion that poverty persists across generations ‘because parents fail to invest in the human capital of their children, for reasons of incapacity, irresponsibility or ignorance’ (Devereux and Mc Gregor 2014, p 301). Programmes aim to break cycles of reproduction of poverty by promoting behavioural changes that would, supposedly, secure better lives for younger generations. However, this simplistic view of poverty obscures broader political and environmental causes of impoverishment in rural contexts. Since the root causes of impoverishment remain unaddressed, recipient families are at risk of becoming more economically vulnerable once their children age out of eligibility for assistance.

Environmental injustice is a key driver of intergenerational impoverishment in Indigenous Amazonia. The subsistence strategies of Shipibo families are severely impacted by water pollution and extreme floods, both of which are aggravated by the extractive development of the forest. This finding is not exclusive to this study. Literature on CCTs in Amazonia has argued that monetary solutions may not be the best way to address the economic vulnerabilities experienced by families who rely on subsistence fishing, hunting, and agriculture (Lang 2019, p 79; Correa Aste and Roopnaraine 2014; Verdum 2016, p 39). This aligns with claims for more transformative social protection that go beyond economic aid to promote social justice in the long term (Hickey 2014; Sabates-Wheeler and Devereux 2008).

Integrating a notion of environmental justice into social protection could contribute to this, especially considering the predominantly rural nature of global poverty (Dercon 2009). Environmental justice debates focus on how environmental degradation disproportionately impacts socially and economically vulnerable families, emphasising the political and economic drivers of resource scarcity (Bullard et al. 2003; Martinez-Alier 2014). If initiatives were oriented towards promoting environmental justice, programmes would have to transcend the boundaries of monetary poverty and pay attention to the heightened environmental risks and vulnerabilities experienced by rural families (Devereux 2001).

Although Hickey (2014, p.335) cautions that social protection may not always be the best approach towards promoting radical structural transformation, environmental justice is key to ensuring the effectiveness of child-sensitive programmes, particularly in a changing climate. Ideally, CSSP programmes should be designed acknowledging the importance of the environment for ensuring reasonable living standards for younger generations (Hiskes 2008; Satterthwaite 1996). This alone could spark a conversation about the rationale behind and the best means of financing social protection.

This study has shown that CSSP serves as a coping mechanism for rural families pushed into monetary poverty due to environmental vulnerabilities (see also Devereux 2001). However, more could be done to ensure that recipients would be better off after leaving these programmes. In addition, complementary strategies could be used to address environmental vulnerabilities directly. To ensure a more transformative and sustainable action against intergenerational impoverishment, policymakers should consider how to combine CSSP with intersectoral measures aimed at enhancing the sustainability of rural livelihoods.

Two main actions could contribute to this goal. Firstly, policymakers need to confront the inherent contradictions within the current model of development in Latin America. As previously discussed, the proliferation of social protection programmes in the region was closely linked to the commodities boom, and this fact is still used to justify the expansion of extractive activities as a means to fund poverty eradication (Svampa 2015). However, extractive development has contaminated rivers and reduced the forest cover that protects rural and Indigenous communities from weather extremes. Hence, policymakers must ensure that the provision of social protection does not inadvertently support a consensus favouring extractive development (Arsel et al. 2016). Although providing a solution to achieve this objective is beyond the scope of this paper, some economists have suggested that progressive tax reforms could offer a more sustainable strategy for wealth distribution (Holland and Schneider 2017; Sánchez-Ancochea 2021).

In addition, governments should consider that addressing the loss of environmental capital can complement and improve the effectiveness of traditional social protection in rural contexts. This would imply setting up rigorous laws to protect the lands of families who depend on subsistence agriculture and their surroundings. It would, however, require a radically different relationship between policymakers and programme recipients, in which beneficiaries would be actively consulted and heard about their needs and struggles.

Conclusion

The current framework of CSSP fails to recognise an important cause of rural impoverishment. The existing programmes focus on promoting behavioural change but disregard how environmental risks and vulnerabilities restrict parental decisions about their children. This study shows that environmental degradation is the main cause of economic hardship for families working on subsistence agriculture. Water pollution, alongside the heightened environmental risks prompted by climate change and deforestation, is exacerbating the impoverishment of Shipibo families and hindering the effectiveness of existing CSSP programmes. The insights from this study transcend the boundaries of Amazonia since rural families are globally overrepresented among those in poverty (Dercon 2009).

Considering that the rise of social protection in Latin America was linked to extractive revenues, this research examined the contradictions of providing CSSP to rural families while also worsening their environmental vulnerabilities. In such contexts, the pollution and depletion of these resources cannot be dismissed as a side effect of economic progress; rather, they represent a development paradox. To imagine more sustainable CSSP programmes, policymakers should recognise the connection between the prosperity of rural families and the quality and accessibility of vital resources, such as land and water (Devereux 2001). Engaging with environmental justice is crucial to acknowledge and address the disproportionate burden of environmental risks borne by socially marginalised families (Bullard et al 2003). It can also draw attention to the importance of environmental sustainability in combatting youth impoverishment.

In practice, being attuned to environmental justice issues would require a shift in focus from individual behavioural change to structural transformation (Sabates-Wheeler and Devereux 2008). Rather than perceiving rurality as a signifier of poverty (Fotta and Balen 2018), CSSP programmes would have to foster the prosperity of rural youth in any context, including their own communities. In practice, this would entail creating additional interventions to strengthen rural youth’s political and environmental resilience while safeguarding the resources they need to thrive in abundant subsistence economies. As a crucial part of this intersectoral approach, governments should re-examine whether the immediate profits of extractive development justify its high and long-term costs for the most economically vulnerable populations.

Notes

  1. (1)

    García Perez published his views in El Comercio (The Commerce), one of the main newspapers in Peru, on 28 October 2007.

  2. (2)

    According to the Peruvian National Institute of Statistics (INEI 2017), only 42% of Shipibo villages have a medical unit and secondary schools, and only 24.7% have a water well. Because the Shipibo are the second largest Indigenous population in Amazonia, and their villages are easier to access than those of other peoples, they are relatively well-serviced in relation to other ethnic groups.

  3. (3)

    This map was produced with data from Global Administrative Areas (GADM); Consejo Shipibo-Konibo-Xetebo (COSHIKOX—Shipibo-Konibo-Xetebo Council); Instituto Nacional de Pesquisas Espaciais (INPE—the National Institute of Spatial Research in Brazil) and Humanitarian Data Exchange (HDX).