The diamond model: a literature review

Gaining recognition of unpaid care as work, and women’s role in providing it, is a long-standing endeavour within feminist economics (England and Folbre Reference England and Folbre1999). The concealment of care in the domestic realm has contributed to the difficulty in assessing its enormous socio-economic contribution (Folbre Reference Folbre2006, 185; Araujo Guimarães Reference Araujo Guimarães, Araujo Guimarães and Hirata2021, 127). Care is a unique type of work, an activity that requires affectionate, relational, and emotional components with a centrality in the provider. Folbre and Weisskopf (Reference Folbre, Weisskopf, Ben-Ner and Putterman1998, 127) criticise a neoclassical perspective on care restricted to a provision of services, without considering motives, attitudes, and shared expectations involved in performing it. Folbre (Reference Folbre2006, 185) argues for the need to go beyond definitions of work based on the site of production (domestic vs market sphere), which result in conventional statistics hiding the costs of unpaid care work.

Models contribute with a comprehensive perspective on how paid and unpaid care happens and is distributed in a given society and time, conveying an architecture and connections between providers. Initial formulations considered three dimensions – State, family, and market – as welfare triangles. Empirical applications made evident the need to incorporate, define, and measure unpaid sources of care beyond the family (Jenson and Saint-Martin Reference Jenson and Saint-Martin2003). The transition from the welfare triangle to the diamond emerged in studies such as Payments for Care by Evers et al in 1994 (Wijkström Reference Wijkström1996), which described the diamond comprising State, market, voluntary sector, and family. This pioneering work examined several case studies, providing an analysis from a macro to a microlevel of individuals and across several cultural, political, and religious settings. It solidified the role, responsibilities, and extent of the ‘family’ and the ‘State’. Yet, it lacked conceptual and empirical characterisation of the market and, especially, unpaid collective forms of care; the latter was portrayed as uncommon, which demonstrates a lack of understanding of ‘basic structure(s) and modus vivendi of society at large’ (Wijkström Reference Wijkström1996, 87).

Folbre (Reference Folbre2006) calls for ‘reimagining care’ from a perspective that recognises other actors, institutions, dynamics, and arrangements involved in the provision of care. By understanding that care may happen in fragmented, simultaneous, and diverse ways, we open the door to consider the weight of an array of actors and institutions involved in care provision, including communities and collective actions (Glenn Reference Glenn, Araujo Guimarães and Hirata2021).

Altogether, voluntary and informal arrangements for care appear to fulfil significant social functions. These arrangements compensate for the decline of services previously offered by highly institutionalised public systems and act as an answer to ‘the crisis of the welfare state and the need of “a new social pact”’ (Wijkström Reference Wijkström1996, 90). However, institutional shortcomings resulting from the retrenchment of welfare regimes are not a new phenomenon if we shift the focus from the Global North to the Global South, and especially to Latin American and Caribbean countries. Among the latter, most never reached the level of institutionalisation of care that many authors refer to as the classical Welfare State.

Razavi’s model considers four main dimensions: State, market, family, and NFP. The author identifies specific forms of care in each dimension and, by doing so, signifies the distribution of welfare and care work as ways in which social fabric and gender inequalities are reproduced. Her argument is that identifying how care happens in each of these dimensions provides a solid understanding of how the social fabric and gender inequality are reproduced (Razavi Reference Razavi2007, 4). Salvador (Reference Salvador2007) uses Razavi’s conceptual framework in her comparative study of the care economy in various Latin American countries. In Figure 1, we bring together both formulations of the diamond and show how Salvador’s work provided details of social actors and institutions in the Latin American case.

Figure 1. Initial formulation of the care diamond by Razavi and advanced formulation by Salvador (Reference Salvador2007).

Source: elaborated by authors based on Razavi Reference Razavi2007, fig 1 and Salvador Reference Salvador2007, 9.

Salvador (Reference Salvador2007) applies Razavi’s model in her comparative study of the care economy in various Latin American countries. In doing so, she enriches the model with detailed actors involved in each dimension. The author identifies the State as a care provider in the centre, seeing how other dimensions articulate around it. How the State performs welfare outlines a regime in each society and defines the weight of care work in the other three dimensions. Salvador organises care work into familistic and non-familistic models. In Latin America, the former model is dominant, involving care work performed predominantly by women in the family/kin, making it a feminised, unpaid responsibility. Non-familistic regimes, less present, transfer care to public institutions and/or the market, complemented by families and informal networks.

The Mexican context

Before delving into the analysis, it is important to offer a succinct depiction of inequalities derived from gender disparities and policies/institutionalised dimensions. In Mexico, women experience the burden of domestic and care work in the private realm (Valenzuela et al 2022; ECLAC 2022). This hinders their economic inclusion and financial autonomy, maintains gender gaps and power disparities (ECLAC 2022; Garnica-Monroy and Hernández-Reyes Reference Garnica-Monroy and Hernández-Reyes2022, 558). As of 2022, the difference between the labour force participation of men and women in Mexico stands at 32.9 p.p. in detriment of the latter.

It is not uncommon that women’s paid occupation acts as a ‘buffer zone’ for the economy, being ‘activated’ and ‘de-activated’ due to changes in the economic scenario (Duque-Garcia Reference Duque-Garcia2021, 21). Women’s consistently higher informality rate – a strong indicator of accommodating double burden – makes them particularly vulnerable. Wages and the coverage of social security are lower in the informal sector, and informal workers are more exposed to income losses. Precariousness and exposure have a significant impact on families and children’s well-being, especially when not counterbalanced by institutions and policies.

In the 1980s, Mexico experienced an unexpected demand for public childcare due to an increase in women’s participation in the labour market (Juárez Hernandez Reference Juárez Hernández2010). The Secretary for Public Education had difficulties coping with the expenses of expanding childcare. Alternative strategies were adopted – training young educators with community leaders and seeking modernisation of basic education and preschools. Since the late 1980s, Centres for Child Assistance (CAIs) have been created to provide care for children of working mothers from 45 days old up to 6 years old. CAIs are regulated by a federal law, regardless of whether they are public, private, or with mixed funding and may offer full-day care, support for education, and diverse activities.

While the State’s responsibility in providing childcare is defined by legislation, some authors characterise it as highly unstructured, with limited institutional design and articulation, unable to attend to low-income families and women, and hampered by regional inequalities (Juárez Hernandez Reference Juárez Hernández2010; Estrada Reference López Estrada2020). Geographical distances are key to analyse the availability and use of CAIs. In low-income, rural, and indigenous areas, the scarcity of public CAIs means no childcare is available as families lack resources to commute. In 2022, more than 90% of parents and legal guardians from indigenous, low-income, and rural areas who take their children to public CAIs do so on foot (CONEVAL and UNICEF 2022).

Limited public support for childcare is part of a wider reduction of resource allocation in Mexico. According to the United Nations International Children’s Emergency Fund (UNICEF) México (2023), public investment in childhood and adolescence in the last ten years has corresponded to 3% of the GDP, below the regional average of 5%. While public spending in the country has increased in the last decade, social protection targeting childhood and adolescence has lost fiscal space from 2.6% of the budget in 2012 to 0.7% in 2023 (UNICEF México 2023). The retrenchment of the Welfare State is to take the current struggle of many families in receiving and providing care into a crisis (Matos de Oliveira and Alloatti Reference Oliveira and Alloatti2022; Orozco Reference Orozco2006).

Aligned with the temporal scope of our study on childcare provision, we highlight the case of the dissolution of the Programme Childcare Centres (PEI) in 2018. The programme was created in 2007, under the Secretary for Social Development, to care for children up to 4 years of age whose mothers were working, studying, or looking for a job or who were the children of single fathers who were working and studying. The PEI covered 10 states and had a mixed care approach, transferring money to families and providing financial support to public and private CAIs.

The programme faced difficulties in resource allocation in vulnerable and rural areas, reproducing geographical inequalities and exclusion. It also had a gender bias due to the key concept of ‘working mothers’. Single fathers needed to effectively prove they did not have a female member in the household to become beneficiaries, reproducing the existing sexual division of childcare (López Estrada Reference López Estrada2020, 4–8; Mateo-Berganza Diaz and Rodríguez-Chamussy Reference Mateo-Berganza Diaz and Rodríguez-Chamussy2017)).

Nevertheless, the PEI showed several positive aspects. In 2007, 5,504 CAIs existed, and 125,359 children were enrolled. By 2016, the programme had 9,195 CAIs and 315,000 children enrolled. Evaluations underlined a positive perception of mothers, strong indicators of children’s well-being, qualified personnel, improved coverage in rural and vulnerable areas and, most notably, promoting low-income mothers returning to the labour market (López Estrada Reference López Estrada2020, 13).

In 2018, the PEI was replaced by the Programme for Wellbeing of Children and Working Mothers. This programme puts the well-being of children at the centre, leaving the support for working mothers in a second stage. A key change was the reduction of the budget to private CAIs that had vacancies for children, and funds were shifted to direct cash transfers to legal guardians (Garcia 2023). This change in resource allocation compelled families to seek other sources of childcare, such as private CAIs or informal arrangements with nannies. The dissolution of the PEI had several consequences, many of which could only be fully comprehended in the long run. Some authors understand it has led to a process of re-privatisation of care, as a movement that brings childcare from institutionalised spaces back to the private sphere (López Estrada Reference López Estrada2020; Carrión et al. Reference Carrión, Reséndiz and Tejada2022). It is with this background our analysis investigates the current distribution of childcare in Mexico.

Analysis of childcare distribution in Mexico

The market dimension: who provides care and who can afford it?

For the market dimension, we detail characteristics of private childcare provision, including costs, services, and workers. We then summarise recent data regarding private services intertwined with the provision under social security (IMSS) for formal workers. As a general depiction, we refer to the National Survey for Occupation and Employment as it offers detailed information on caregivers of children, people with disabilities, and the elderly in establishmentsFootnote ¹ and in private homesFootnote ² (INEGI ENOE 2023b). In Table 1, we show characteristics of paid private childcare in both spaces. We underline a considerable difference in numbers in favour of caregivers working at households, almost twice those in establishments and, equally important, a significantly higher informality rate.

Table 1. Caregivers of children, people with disabilities, and the elderly in establishments and private homes (Mexico, fourth trimester, 2022)

Source: INEGI. Based on National Occupation and Employment Survey – ENOE.

A comparison of trends between the number of workers in these two groups is shown in Graph 2. There is a visible decrease in the workforce in establishments, simultaneous to an increase of those working in private homes, with the exception of the peak of the pandemic (Figure 2).

Figure 2. Number of caregivers of Children, People with Disabilities and the Elderly in establishments and private homes (Mexico, 2013–2023, trimesters).

Source: INEGI.

Data on private CAIs, prices and fees for childcare services in Mexico are considerably outdated. The CONEVAL (2022) works with information from 2017, which we use to contextualise the private sector. In that year, the minimum wage was around 2,500 to 2,640 Mexican pesos per month. For the same year, 44% of families with children in private CAIs paid up to the equivalent of 1 minimum wage for this service. When the monthly fee moves up to between 1 and 2 minimum wages, there is a significant decrease in the number of children attending CAIs. It drops to 3.95% of the families for this value range, and 0.6% for CAIs charging an equivalent of 2 minimum wages or more.

The IMSS provides childcare as ‘direct provision’ (vacancies in CAIs that belong to the IMSS) and as ‘indirect provision’ (which refers to private CAIs with a contract with the IMSS to provide vacancies for children or a payback arrangement for childcare fees). The IMSS relies heavily on the indirect provision, which has reduced since 2009, from 1,568 CAIs to 1,143 in 2024. The direct provision by the State for 2024 is 126 CAIs, 15,142 beneficiaries, 16,130 enrolled children compared to the indirect provision of 1,143 CAIs, 16,577 beneficiaries and 175,417 enrolled children (IMSS 2018). The decrease in the number of CAIs, vacancies, and beneficiaries and the increase in unmet demand convey a clear reduction of the public provision of childcare. However, we stress an extra layer of complexity due to a previously utilised strategy to increase the number of vacancies among the existing CAIs.Footnote ³ This can lead to overcrowding and does not by itself address regional absences and disparities – regional differences, socioeconomic vulnerable areas of the country, vacancies for babies and children of certain age – a key vector of inequality in Mexico.

Considering the above-mentioned increase in the IMSS target population of around 14% between 2015 and 2022 (CONEVAL 2023b), we stress three main points around the State regulation of private services. First, the heavy reliance on ‘indirect provision’ without conditions for geographical location and enrolment prioritising ethnicity or socioeconomic condition. Second, the vulnerability of this provision is that private parties can suspend contracts unexpectedly. Third, the expansion of vacancies in already existing CAIs leads to overcrowding. Altogether, the reduced State regulation indicates the possible reproduction of inequalities and vulnerabilities.

NFP initiatives: gender, community, and provision of childcare

As discussed in the literature review, data on NFP initiatives in Mexico, as in many countries, is hard to gather as many community-based and collective actions are temporary, non-institutionalised and informal. Also, they frequently occur among hard-to-reach populations and involve an array of actors, communities, networks, and neighbours’ associations. To provide a proper depiction of this dimension, we examine national statistics and delve into some experiences to deepen our understanding.

The NFP sector in Mexico

According to the satellite account for NFP institutions in México (CSISFLMFootnote ⁴ ) for 2022, the GDP of the sector represented 2.9% of national GDP (INEGI 2023c). We focus on associations classified by criteria that consider the service and its social objective as social services. This includes centres and actions providing daycare, supporting child development, and orphanages for babies. As a caveat, we state that in the INEGI data we analyse, NFP social assistance services may include some form of basic healthcare attention. In Graph 3, we analyse the distribution of paid and unpaid positions in NFP social assistance and health services by sex between 2019 and 2022 (Figure 3).

Figure 3. Number of workers in health and social assistance at the NFP sector according to sex, paid, and voluntary work – Mexico (2017–2022).

Source: INEGI. National Account system. Satellite account Not-for-profit institutions.

We underline two points aligned with existing literature. First, voluntary work surpasses paid work by almost twice the number of total positions. This could produce a vulnerability in the provision of services, as volunteers may have a less stable participation and/or fluctuation in the hours dedicated to them. Second, there is an overrepresentation of women in paid positions. This aligns with studies that show it is frequent for women to have a higher participation in paid positions in NFP childcare arrangements in México (Fraga 2022).

While women may have work in paid positions may be positive, it does not exclude inequalities and precariousness. Available data does not indicate formal employment, remuneration, or social security benefits. Moreover, the extreme need to access these services via collaboration in these spaces or associations may hamper women’s autonomy in pursuing more beneficial jobs and financial independence. Additionally, participating in NFP associations or actions, even when paid, can be seen as an extension of domestic and care work, underestimating women’s workload.

We stress that it is difficult to single out childcare for analysis within NFP spaces. Studies in the region indicate that NFP initiatives address several needs simultaneously, such as care for children and the elderly, meals in communal kitchens, protection for women from gender-based violence, access to hygiene products, and health services (Fraga Reference Fraga2022; Fournier Reference Fournier2022; Carrión et al. Reference Carrión, Reséndiz and Tejada2022). Women participating in collective kitchens may benefit by feeding their own children, which alleviates the pressure on budgets, cooking time, and decision-making.

Experiences on geographical and social context as a vector for NFP actions

In Latin America, according to Fraga (Reference Fraga2022) and Fournier (Reference Fournier2022), women and families experiencing poverty, exclusion, and unmet basic needs in a shared space are the main actors in developing and sustaining NFP actions. The latter must be understood as anchored on a territorial scale: neighbourhoods, communities, rural areas, indigenous communities, and areas of extreme poverty. Low-income families and working mothers experience spatial segregation in a plethora of ways, including long commutes to their jobs, limited mobility, and restricted or absent public services (Garnica-Monroy and Hernández-Reyes Reference Garnica-Monroy and Hernández-Reyes2022).

A geographical analysis connecting vulnerability, poverty, and NFP initiatives is outside the scope of this study. Nevertheless, we examine selected noteworthy experiences in México. Juárez Hernández considers the efforts to address a higher demand for childcare in the 1980s (Reference Juárez Hernández2010, 1–2) by training community leaders among indigenous groups and vulnerable communities. According to the author, this qualification – which provided guidelines for child development in and outside the household – contributed to and strengthened social movements and actions since 1990, supporting the creation of NGOs that offer childcare in community centres employing mothers.

Fraga (Reference Fraga2022) explores community care amid increased gender violence in México in the aftermath of the COVID-19 pandemic. She examines the Sorority Network (Redes de sororidad), an indigenous community in the state of Chiapas, which created safe spaces for women escaping gender violence. Simultaneously, the community shared knowledge on growing poultry and crops using natural pesticides. This endeavour provided support for urgent needs, such as food for families, and knowledge to sustain these practices over time (Fraga Reference Fraga2022, 22–23).

NGOs have supported migrant children and families at several points in México, providing shelter, meals, and school support to children of migrant families (Estrada Reference López Estrada2020, 16). Montes and Paris Pombo (Reference Montes and Paris Pombo2019) and Valenzuela et al. (Reference Valenzuela, Scuro and Vaca Trigo2020, 55) examine care chains and care networks. The latter emerges as a horizontal and distributed involvement of individuals in caring for children, aligning with studies that highlight the importance of networks involving friends and non-family individuals in socioeconomically vulnerable families and communities (Carrión et al Reference Carrión, Reséndiz and Tejada2022, 34; Orozco Reference Orozco2006).

Findings and discussion

Applying the diamond model allowed us to capture the provision and distribution of childcare happening in Mexico along each dimension (family/household, market, public provision, NFP). In Figure 4, we offer a synthesis of our findings using as a base Razavi’s (Reference Razavi2007) original formulation and Salvador’s (Reference Salvador2007) detail of actors and institutions to depict the Mexican case.Footnote ⁵ With the original proportional diamond as a template, we fit a non-proportional diamond shaped by our assessment of childcare provision in Mexico. The disproportionality of our – inside – diamond expresses the reduction of some dimensions, as well as the interactions among dimensions. To better express these interactions, we use two types of arrows indicating if childcare is being transferred from one dimension to another due to restriction or articulation/collaboration between dimensions.

Figure 4. The diamond care model applied to Mexico.

Source: Own elaboration.

We stress three main findings to discuss. First, the State is the most significantly reduced, both as a dimension in childcare provision and as an actor considering its effective regulation of other dimensions. It presents diminished funding in public CAIs, policy changes involving cash transfer to families, and high reliance on third parties for childcare provision (indirect provision). This results in the transfer of childcare from the State to other dimensions – to the family/household as re-privatisation of care, to the market, widening the demand for private caregivers and CAIs, and to the NFP sector as unfulfilled needs driving community actions. As an actor, the State’s interaction with the market dimension, mostly through its reduced regulation, is linked to reproducing inequalities by not guaranteeing the service to those who are more vulnerable.

Second, the interactions between the market, family/household, and NFP indicate degrees of blurring between private and public domains regarding childcare. Following the changes in public support of childcare, mainly through cash transfer to families, we identified an increase in caregivers working in private homes. We frame this phenomenon within a long-term debate by feminist economics that care in the private realm conceals informality and exploitation due to barriers to regulate working relations within the domestic sphere.

Between the family/household and NFP, the blurring of domestic and public domains happens as women take their own children to be part of community-based on collective spaces and non-market logics, such as solidarity and reciprocity, organising work and collaboration. While community-based childcare provision may be seen as transferring responsibilities from the family to NFP spaces, the latter emerge predominantly due to unfulfilled basic needs that fall under the State’s responsibility. The predominance of women in structuring, functioning, and sustaining NFP initiatives conveys the weight of childcare on them, enhanced by precarious labour market participation and public support. While extreme needs are felt by families and communities, their weight is highly feminised as they are seen as women’s and mothers’ responsibilities (Carrion et al Reference Carrión, Reséndiz and Tejada2022, 22). Favourable collaborations happen between the family/household and NFP dimensions, mostly due to direct and indirect benefits for women participating in these collective spaces, such as meals, safety from gender violence, and empowerment. Even when women receive monetary compensation, they face constraints in getting another job as they depend on community-based and collective sources of childcare.

Third, and relatedly, we underline the importance of qualifying the interactions between family/household and State through choices around public childcare. Low-income women and families may face many challenges, yet they also exercise their decision of not utilising the public service. Reliance on family members, networks, and NFP spaces for childcare is connected to perceptions and assessment of (i) who should be responsible for raising children (mostly mothers), but also, (ii) the lack of trust in public services and the overall absence of services to address the needs of low-income women.

Limitations in public childcare provision impact families differently. For those with resources, this has increased the demand for private care in establishments and at home. The latter are difficult to regulate, allowing for exploitation and abuse. For rural, indigenous, and low-income communities, unpaid care work at home is bound to increase. While it is not possible to affirm there is a re-privatisation of childcare (Estrada Reference López Estrada2020; Carrión et al Reference Carrión, Reséndiz and Tejada2022), our analysis conveys an increased in-household burden for those with less resources. This movement may also be channelled via NPF initiatives.

As we sought to understand childcare through the four dimensions, looking at actors, institutions, and transferring of responsibilities, we circle back to our question: what are the gender consequences of childcare distribution in Mexico? Unpaid care work is a permanent and considerable phenomenon with significant negative impacts, especially among women with limited resources. As it exists in the family/household and NFP dimensions, it is particularly opaque, with restricted support and regulation from the State. We highlight three summarising points: i) gender norms and roles seem ubiquitous to all four dimensions conveying that women are seen as responsible for childcare; ii) even in countries without a strong welfare system, the State still occupies a central position, as ‘That which the state does not take on is left to markets, families or communities’ (Jenson and Saint-Martin Reference Jenson and Saint-Martin2003, 81); (iii) the current organisation of life has generated a crisis of care that cannot be satisfied by a sole actor (Orozco Reference Orozco2006; Carrión et al Reference Carrión, Reséndiz and Tejada2022; Fournier Reference Fournier2022).

Final remarks

We have synthesised childcare provision, distribution, and arrangement in Mexico based on Razavi’s diamond model. Our version of the model organises findings into (i) actors and institutions in each dimension and (ii) interaction between dimensions. Therefore, the disproportionality of our version of the diamond expresses the reduction of some dimensions, identified at each vertex, as well as their interaction with other dimensions. Conveying the existence of diverse and simultaneous forms of childcare in different dimensions directly challenges the traditional idea that care is women’s responsibility and that it should be carried out in the private sphere.

We have shown the consequences of a compromised public supply of childcare, which produces the transfer of childcare from the State to other dimensions – to the family/household, overburdening women, to the market, widening inequalities between families/women of different social classes, and to a limited NFP sector. Moreover, we make the case that simply increasing the existing public childcare would not – automatically and/or necessarily – satisfy the existing demand or alleviate pressure on women. Alternatively, community-based actions and NFP initiatives providing care have the potential to provide best practices for the State in effectively boosting vulnerable communities.

Three points are important. First, the reduction in public childcare services impacts heavily on women and families without resources. Second, the overburdening and transferring of obligations is linked to (and may reinforce) social perceptions that mothers should care for young children. Third, in both the family/household and the NFP dimensions, absorbing the demand of childcare is particularly opaque, with limited support and regulation from the State. These points support a clear need for policies, measures, and strategies tailored to the interdimensional character of unpaid childcare.

At a conceptual level, our study contributes to the development of models to analyse welfare and childcare distribution in the Global South, especially among societies that historically lacked the strong welfare regimes of some of their counterparts in the North. A key advantage resides in granting analytical space for the household and NFP sector – as well as their interactions – as constitutive dimensions and fundamental sources of childcare, as it has been in many countries in the region. We echo Razavi in avoiding a linear path of analysis in ‘which all countries move with an inevitable shift from “private” (family and voluntary) provision of care to “public” provision (by the State and market)’ (Reference Razavi2007, iv). Consequently, it is key to explore ways in which a society can better appreciate and leverage different sources of childcare provision and their articulation, and avoid reproducing or strengthening the idea of care as women’s responsibility.