Women’s empowerment and social innovation in childcare: the case of Barcelona, Spain

ABSTRACT Social innovation and empowerment are complex concepts that, from an analytical point of view, are not necessarily related. One explicit goal of social innovation is to empower communities, as well as the individuals that are involved in activities within those communities, but this does not necessarily always occur. Here we address the question ‘Does social innovation in early childhood education and care (ECEC) empower women?’ First, we explore whether the projects we examine can be defined as social innovations. Second, we analyse to what extent arrangements that are identified as innovative in ECEC empower the mothers who choose them. We argue that if the characteristics of a particular social innovation project enhance or reinforce the capabilities of the women who participate in it, that experience will most probably empower them; if not, this is unlikely to occur. Our empirical material includes 37 interviews with key informants, educators, and mothers involved in these non-institutionalized projects, collected in Barcelona. Our results reveal the socioeconomic bias in these projects, as well as the costs derived for both sets of participants (mothers and educators). They also show the wider social impact that stems from these projects being under-regulated.


Introduction
Over the last few decades, the study of early childhood education and care (ECEC)that is, for the under-threeshas contributed to enriching the debate on the welfare state, and has affected policy areas such as social investment, equal opportunities, equity, gender equality, or the interplay between education, care and work-family balance (Blasco 2017). Unlike other welfare policies, access to ECEC is not provided on a free and universal basis in most welfare states, but through a mix of public, private, and subsidized private provision which only covers a small part of the demand, and which offers similar institutionalized models. In this context, social innovation servicescharacterized by horizontal and community-based governance and workings in which parents and educators are involvedhas recently proliferated in this area as an alternative to market or public provision, offering a different approach to the education and care of the under-threes. Social innovation projects are characterized by parents being heavily involved in several aspects of the design and operation of the service. They have raised new questions about women's empowerment that intersect with other, more familiar aspects.
Social innovation and empowerment are two complex concepts that, from an analytical point of view, are not necessarily related. In theory, one of the explicit goals of social innovation is to empower communities and the individuals that are involved in the activities within those communities (Baglioni and Sinclair 2014), although this does not necessarily always occur in practice (Blanco and León 2017;Cruz et al. 2017;Maestripieri 2017a). Thus, much of the literature has discussed 'social innovation' and 'empowerment' as two sides of the same coin (Moulaert et al. 2016). For example, some projects have been analysed that are related to community action (Barbieri et al. 2018(Barbieri et al. , 2021, or to the activation of service provision arrangements that are alternatives to the traditional ones used by the majority of people (co-production in areas such as housing, health, dependency care, etc.) (Gallego 2019) among others. In this paper, we explore if social innovation is able to trigger the empowerment of the women who are involved in innovative ECEC services and, if this appears to be true, we seek to understand the characteristics of this empowerment.
To do so, we explore currently active social innovation projects in ECEC services for children under three years old. These began to emerge in the city of Barcelona in the early 2000s as alternatives to existing regulated services. The three types of projects examined are the following: childminders, care groups and free-education nurseries. Despite some differences in how they function, they are all usually set up at the initiative of parents or educators; they explicitly promote alternative practices of education for the 0-3 age group, namely free-education methods 1 ; and their everyday activities require the involvement of the participants, who contribute by volunteering their hours. We pose the question 'Does social innovation in ECEC empower women?'. We analyse to what extent the arrangements we identify as innovative in early childhood education and care empower the mothers who choose them; that is, to what extent they enhance or reinforce their capabilities. The word capabilities refer both to the potential and to the actual power of what a person is able to achieve in terms of valued choices (Gangas 2016;Sen 1985). Empowerment is conceptualized as an increase in women's capabilities and is operationalized through the relationship between three elements: resources, agency and achievements (Ibrahim and Alkire 2007).
We also analyse the type of empowerment sought by women participating in those projects. To this end we study: (i) why mothers choose the projects; (ii) the costs of these options for both mothers and educators; (iii) the profile of the women who opt for social innovation in ECEC. The empirical material collected in the analysis includes 37 interviews with key informants, educators and mothers involved in the projects. Our analysis focuses on the interviewees' discourses, narratives and arguments, which we take as expressions of their perceptions and feelings. These narratives allow us to identify the extent to which the interviewees are conscious of the constraints that condition their preferences and choices, that thereby limit their capacity of acting according to their values and, ultimately, prevent them achieving empowerment. We do not aim to obtain an objective measurement of empowerment, but a subjective one, in which the individual perceptions of both costs and benefits play a role. In the discussion we reflect on whether these socially innovative arrangements in ECEC contribute to enhancing women's capabilities at the individual level, and whether they reproduce patterns of social inequality at the community level.
In the first section, we explore the theoretical relationship between social innovation and women's empowerment, reviewing the (relatively few) articles that have analysed social innovation in childcare. In the methodology section, we present our study and fieldwork. Then we illustrate the specific characteristics of social innovation in ECEC projects in Barcelona; to do so we compare the three types of innovations that we examine in this study. We then analyse women's empowerment (in the 1 Free-education pedagogies are inspired by the work of Montessori, Waldorf, Pikler (and many others).
They uphold that a non-institutionalised environment is more favourable to a child's development, and that learning should be respectful with the different learning speeds of all children. Please find extended information on this teaching philosophy here, in an informative page in Spanish provided by one of the associations we have studied: https://educaciolliure.org/es/la-educacion-libre/.
terms defined above) in the projects we have studied, including both mothers and educators. Our results reveal the socioeconomic bias in these projects, as well as the costs (in terms of professional career or labour conditions, for example) for all participants (mothers and educators), and the social impact that stems from the projects being underregulated by public authorities. When socially innovative projects are under-regulated, their possible benefits in terms of empowerment are mainly individualthat is, they only benefit those who participate in those projectsand do not spill over to communities as a whole.
Linking social innovation and empowerment: the role of capabilities Social innovation and empowerment are complex, interwoven concepts that have been the object of extensive research. However, how and why they are interrelated still remains underexplored. Social innovation includes practices that aim to satisfy human needs through horizontal cooperative relations among citizens. Such practices are social because the type of goods and services they generate are conceptualized as basic by stakeholders, namely, citizens with an interest and a stance on the issue (BEPA 2010: 31). Such practices are innovative because they generate alternative arrangements and provision models that differ from those offered by the institutionalized public and private sectors, and which are intended to empower citizens (Grimm et al. 2013). Empowerment includes processes that lead people to feel they not only can, but also have the right to occupy decision-making spaces, which encourage them to act and have influence (Rowlands 1997). Empowered individuals exercise their agency to remove social barriers and obstacles that may hinder their own and others' wellbeing, that is, factors that may constrain and limit individuals' agency (Kabeer 1999;Haugh and Talwar 2016). Although both social innovation and empowerment refer to both processes and outcomes, empowerment has come to be seen as an outcome of social innovation processes (Ibrahim and Alkire 2007). The missing link between social innovation and empowerment has been a concern expressed in the literature on the capabilities approach, which is central to the research on social innovation and on women's empowerment (André 2013;Batliwala 1993;Cornwall 2016;Maestripieri 2017b;Lindberg et al. 2015). In the specific case of women, empowerment involves challenging existing gender relations that shape the structure of opportunities and constraints for women, in order to offer them alternatives that allow them to exercise agency and choice over their lives. Sharaunga et al. define women's empowerment as 'the multidimensional process of increasing the capacity/capabilities (i.e. resources and agency) of individuals or groups to make choices and to transform those choices into desired actions and outcomes' (2019: 5). Citizens' resources are preconditions that influence their agency (action processes); agency in turn influences their achievements (outcomes) (2019: 3). This implies that women use their economic, social, political, familial, legal and psychological capabilities (i.e. resources and agency) to achieve outcomes in their everyday lives. One example could be cash transfers programs in the global south: the responsibility for using money transfers (resources) is entrusted to women (agency), thereby increasing their economic independence to achieve better education and children's and women's empowerment (achievement) (Tabbush 2010).
Outcomes are multidimensional and relational: as well as affecting individual citizens' capabilities in different ways, they may possibly also generate impacts that might have reciprocal influences on capabilities at the community level. Ziegler (2018) argues that capability conversion factorsnamely, institutional, social, economic and policy factors, among othersare contextual and, therefore, changing them may improve citizens' capabilities. In this sense, the studies that have investigated the relationship between women's empowerment and their involvement in social innovation have obtained contrasting results (Maestripieri 2017b;Lindberg et al. 2015;André 2013, Cukier et al. 2018. They highlight that mere involvement in social innovation is not enough to generate community empowerment if it is not accompanied by a systemic change in the underlying social structures that create inequality. In this sense, social innovation and empowerment have been connected through the concept of social transformation, carried out by previously silenced groups in specific territorial contexts (Rappaport 1987). This view sees the relational nature of empowerment as central to understanding social innovation. Following Ayob et al. (2016), it is possible to identify two traditions in the academic debate on social innovation: a weaker tradition, which considers social innovations to be new or alternative practices that achieve an increase in personal utility (individual outcomes), and a stronger tradition, which considers social innovations to be initiatives that produce collaborative co-productions that have an impact on existing power relations (community outcomes) (Ayob et al. 2016;Chiappero-Martinetti et al. 2017). A parallelism can be drawn between the two traditions and, respectively, the concepts of individual empowerment (weaker tradition) and community empowerment (stronger tradition). Individual empowerment occurs when a person, following their involvement in social innovation, develops selfefficacy, personal competence, and the capacity to act according to their own agency. Community empowerment occurs when groups and organizations gain the capacity to influence the distribution of financial, political, and social resources, thus causing an impact on preexisting structures of inequalities (Haugh and O'Carroll 2019).
Thus, in our research, on the one hand, an explanation is required as to why some women choose social innovation projectschildminders, care groups, and free-education nurseries in our studyinstead of institutionalized ECEC, whether a private or subsidized private service. On the other hand, despite the increasing interest in these projects, research into their implications and unintended consequences is scarce, not only in the Spanish context (Subirats and García-Bernardos 2015;Keller-Garganté 2017), but also at the international level (Brennan 1998;Findlay 2015;Mahon and Jenson 2006). Recent comparative studies have analysed the impact of ECEC policy design features on capabilities conversion factors, particularly on gender and class (Yerkes and Javornik 2019). The research we present here will contribute to providing evidence about such implications from the field of social innovation in ECEC. In addition, the results may help identify the changes that these social innovations are attempting to achieve, and help discern the capability conversion factors that might be at play.

Methodology
The goal of this article is to investigate to what extent social innovation in ECEC is able to trigger empowerment in women who use it and work in it. As a second goal, we would also like to find out the type of empowerment these women are seeking.
The empirical results presented in the next sections are part of the wider Primera Infància research project. 2 The aim of this project is to study the participation in the labour market of women in Barcelona who have children younger than three, and to determine to what extent childcare services can favour this participation. The fuzziness of social innovation poses a challenge for the operationalization of this concept. For this research piece, we selected projects that respected the following criteria: they had to be initiatives that (i) were led by citizens; (ii) fostered cooperative and horizontal relationships between participants; (iii) generated alternatives to ECEC services offered by the state or the market, following previous definitions of social innovations (Blanco and León 2017). This selection yielded three types of projects: childminders [llars de criança], who are educators that welcome children in their own homes; care groups [grups de criança], that are cooperative groups of parents and educators, and free-education nurseries [espais de criança], associative non-institutionalized educational playgroups for children. Although childminders have a long-standing tradition in other European countries (e.g. tagesmutter in Germany or assistante maternelle in France), in Barcelona, all three types of projects are relatively new (the oldest project of this kind in the city was founded in 2000). We should add a second point regarding our selection of cases: in the case of childminders, the participation of parents in the functioning of the projects is limited to the preparation of the food for the children (which is sometimes a task that is shared among parents); however, there are horizontal relationships created by the projects, as most childminders organize group events in which parents and children are involved.
We followed a maximal variation sampling (Patton 1990): we selected five social innovation projects in three neighbourhoods in the city (Gràcia, Poble Sec and Horta-Guinardó) which on average have different socio-economic make-ups (respectively: high, middle and middle-low income). In each neighbourhood, we selected two childminders, two free-education nurseries and one care group: for each project, we interviewed one of the educators and at least one mother. We interviewed only mothers because the Primera Infància project was aimed at studying women's decision-making when tackling the childcare and education of their children under three years old.
The data collected came from three different types of sources: firstly, we set up four exploratory interviews with representatives of associations and activists in the field; and secondly, 15 interviews with educators. 3 In both these cases, we used semi-structured interviews that mostly focused on the socially innovative elements of the project, on how the projects functioned, and the extent of the parents' participation. Thirdly, 18 interviews were performed with mothers who participated in one of the aforementioned projects. In this case, we used biographical interviews in order to reconstruct the decision-making processes behind their childcare choices and their participation in the labour market. The questions in the interviews were mostly about the women's past, current and future plans in terms of their children, their relationships with their partners, their participation in the labour market and professional career path, their current involvement in the social innovation project and the motivation for this choice. A few mothers (5 out of 18) had already finished participating in the 0-3 projects, as their children were older. Table A1 in the annex provide full details of the interviews with the mothers.
All the interviews were transcribed in full and analysed using Atlas.Ti 8, and a content analysis was carried out on them. The most important codes used were: the socially innovative elements of the projects, the functioning of the projects, work-family balance strategies, and women's empowerment. The codes regarding social innovation and empowerment were created partly deductively, but the majority of codes were created inductively following the procedure recommended by the grounded theory approach (Glaser and Strauss 1967).
The interviews lasted between 30 and 90 min; the fieldwork was conducted between May 2019 and January 2020, prior to the onset of the COVID-19 pandemic and the subsequent lockdown that closed all childcare facilities in Spain. The project received ethical approval from the 'Commission of Ethics in Animal and Human Experimentation' of the Universitat Autònoma de Barcelona. Representative association names are disclosed as they have a public role and fight for the acknowledgement of their services, so they explicitly gave us permission to cite them in this research.

Socially innovative childcare in the context of Barcelona
Over the last two decades, Barcelona has experienced a strong growth in the supply of institutionalized ECEC services. The model that has been developed is the escola bressol, based on the construction of municipal public nurseries, which over the years have consolidated their reputation for offering high quality services. There are now 102 of these nurseries distributed evenly around the city and serving 8500 children. This is only just over half of those who currently apply for the service (around 14,000) and about 21% of the total children number of under three who are resident in the city (38,377 children in 2020). Around 24% of underthrees attend private nursery schools (of which there are 195) and the rest are either not in education or attend non-institutionalized forms of ECEC services, in socially innovative projects that are similar to ones that exist in other countries (Barcelona City Council, 2021). The municipality itself, in its last strategic plan for early childhood education and care (dated April 2021), recognized the increasing multiculturality and diversity of the families applying for the service, as well as the growing differentiation of needs that can only be partially covered by an institutionalized model of care such as the one provided by the escoles bressol (Ajuntament de Barcelona, 2021).
In the early 2000s, socially innovative citizen-led projects started to emerge in this context. Some families chose these projects as their first option; others, because they were not able to access public nurseries (unfulfilled demand for them has oscillated between 35% and 50%). Educators and families chose these options because, while their prices were similar to those of private nurseries, they offered the following features: lower ratios of educators per children (3-4 children per educator at 1 year of age) than both the public and private institutionalized services (over 8 at 1 year of age); a non-school environment with a flexible adaptation process that was more like home care or a playgroup; innovative education models; and a social community network (other families, educators, etc.) that gave them support in their parenting experiences (Zechner 2020).
Non-institutionalized ECEC projects can be divided into two broad groups: 'parent-led' groups (care groups) and 'educator-led' groups (childminders and free-education nurseries). All share some characteristics, such as a low adult/child ratios, free-education pedagogical principles and a non-school environment (see Table 1), but each has its specific way of functioning in terms of social innovation.
Childminders (llars de criança) are especially appropriate for very young children. One or two educators open up a private home and look after children from 4 months up to three years old, in a familystyle environment. They have usually studied pedagogy and have specialized in free education; most of them have also taken the course offered by the Llars de Criança association, which began fighting a few years ago for these projects to be officially acknowledged and professionalized. Parents usually have a one-on-one relationship with the educator, although some projects also organize events where families get together.
Care groups (grups de criança) are mostly bottom-up arrangements in which parents participate. They usually stem from childbirth and breastfeeding groups, one of the services offered within the public health services in Catalunya. In these care group projects, mothers selfcoordinate, taking the lead to organize groups in which parents participate actively to help an educator in the day-to-day tasks necessary in childcare activities. The educational project is decided collectively between parents and educators, through assemblies. Most of these projects are informal and self-managed; some are also anti-system, i.e. their activities are held in squatted spaces. A few groups manage to evolve and consolidate into free-education nurseries.
Free-education nurseries (espais de criança) are usually run by two or three educators that take the lead, although parents are required to participate regularly by attending assemblies and helping with logistics (i.e. communication, financial management, organization of open-days and fundraising events, cleaning and food preparation). The educational project is decided by the educators, with parents contributing their opinions and ideas. Most free-education nursery projects that we examined are the evolution of previously established care groups, although this is not always the case. Once the original children are older, and enter the Source: Authors' own elaboration based on interviews and document analysis. a Although in the whole population of projects that belong to this category there are projects that are squatting their locations, we do not have any in our sample. We only have one casebut this is an exceptionthat is using a municipal building for free that has been assigned to them specifically for running their project. The rest of the care groups are renting their spaces on the private market. public pre-school or primary school system, the educators look for new families to enrol in the nursery. A few free-education nurseries have now been functioning for over five to ten years; despite this, the working conditions of the educators and the financial status of these projects tend to be very precarious. The people who run these projects are reluctant to define them as innovative, although they provide alternatives to the dominant models of ECEC service provision (Blanco and León 2017). In line with previous empirical investigations on social innovation (Sinclair et al. 2018;Maestripieri 2020), some participants argue that the cooperative practices at the basis of their projects are contemporary reinterpretations of preindustrial habits in which children were raised in extended families of relatives and neighbourhoods. The real innovation is that that those who run the projects are now competent workers, specifically trained to foster free-education principles.
Apart from being novel, social innovations must be social in their means and ends (BEPA 2010). In terms of being social in their means, these projects respect the principle of free education and, thus, the children are nurtured and respected as they undertake the path of separating from their parents for the first time. This requires a gentle process of adaption to the project that lasts several weeks. This might constitute a barrier to access for working mothers, because not all parents can afford longer unpaid leave or several weeks of the flexibility required in their schedules. A second social element of the projects' means is the participation required of the parents.
As for the social nature of the projects' ends, the majority of our interviewees, regardless of whether they were association representatives, educators or mothers, agreed on the fact that the primary beneficiaries of these activities were the families that participate in them, i.e. the children and their parents. They attest that these projects, unlike traditional childcare services, offer the parents the possibility of maintaining an ongoing relationship with the educator and of being involved in a community of peers, learning from them and receiving help when needed. A second social objective promoted by these projects is to foster new pedagogical practices, generating beneficial impacts on the lives of both the children and their families. Although most of the educational methods used in the projects belong to well-known and established pedagogies such as Montessori, Waldorf or Pikler, only a minority of institutionalized childcare services follow these practices, and so our interviewees perceive they benefit from the added value of accessing these pedagogies when choosing these projects.
On the other side of the coin, there are the costs associated with these innovative projects. They do not receive public funding, and in the majority of cases they have to rent a space on the private market to carry out their work, incurring high fixed costs. For parents, being part of this social innovation implies paying a monthly fee that is higher than in public nurseries, although it is in line with private nursery services. These costs exclude people on lower incomes from accessing these initiatives. Apart from the financial costs, choosing this type of education requires the families to make more of a commitment: the opening times are shorter compared to other more institutionalized alternatives and, especially in the case of nurseries and care groups, parents put time and effort into organizing the projects, as well as preparing the children's food. Many of the mothers involved in these projects opted to reduce their in working hours or take leave in order to ensure they could contribute to their functioning (see next section).
To reach an equilibrium between viability and accessibility, most of the educators employed in the provision of these services work under unsatisfactory conditions, in a context of under-regulation or no regulation at all. Almost all the educators we interviewed received informal payment for part or all their working hours, and the projects are not recognized as schooling activities but as associations organizing playgroups (with no educational tasks). The educational authorities' lack of acknowledgement of their work worsens the workers' conditions. It's a temporary, discontinuous contract. And it's the lowest category that exists, it doesn't even represent what … I mean, it's like the minimum salary and category, I think it's something like admin assistant category, although it's a teaching assistant job. It's the closest, let's say. There are also other reasons, added to the fact that we haven't yet made it financially. Legally we are in a … we are in a situation that is not legal. (P0, free-education nursery) These projects face acute financial strains, leading to many of them struggling to keep their activities afloat or sometimes even to comply with standard labour market regulations. Most of these educators are female, it is clear that these social innovations place a costly burden on women workers.

Women and empowerment: the role of social innovation in childcare and education
The mothers who participate in these projects have quite a specific profile: most of them had their children after they were 35 years old (the symbolic beginning of 'geriatric' pregnancies), they still have only one child, they are highly skilled and employed in intellectual activities such as research, counselling, health, or education (see Table A1 in the annex). Nevertheless, a majority voluntarily opted to pause their careers during the first three years of their children's lives, giving priority to intensive mothering instead (Damaske, 2013). They say that opting for socially innovative ECEC gives them a series of advantages: (i) it offers a non-school environment, more like a family than an institution; (ii) they feel empowered in their process of becoming mothers (especially if this is their first child); (iii) they become part of a community of peers, with whom they can share their ideal model of parenting and childcare, based on affection, free education and low adult-children ratios.
Surprisingly, their main narrative is not to achieve an easier workfamily balance: short opening times and constant involvement during the adaptation period and in the cooperative organization require a degree of participation which is not easy to juggle with a full-time job. The main priority for them is the possibility of enjoying years of intensive mothering with their children (Lareau 2003;Damaske 2013). These mothers opt for social innovation because it allows them to exercise a primary role as educator and carer in the first years of their children's lives. However, they do not perceive their investment in mothering as being in tension with their careers (Damaske 2013): their discourses rely on the empowerment offered by the social innovation: the empowerment of being mothers over being workers.
When we were looking for a nursery school, there were some that made everything easy for you. If you want to bring the baby at 7 in the morning you can; you can leave the child from 7 in the morning until 6 in the evening if you pay extra, and you can bring them before 9. That's really good if you want to go out to work and forget that you have a child. I didn't have a child for that. (M1, free-education nursery, 38) It was a dream come true to educate our children in a different way and … and I don't regret it at all. I'm proud of having done it, of having participated in this project, having built this reality for our children and my daughter is delighted. (M3, care group) Social innovations step in when institutionalized solutions are not able to respond adequately to a social need (Moulaert et al. 2016). In the case of mothers opting for social innovation, the need that has not been answered is a childcare service which does not prioritize the practical needs of a working mother, but, instead, the welfare of the child. Social innovations offer care which is similar to a mother's in terms of providing affection, protection and respect for the development of the children; families that have these needs accept that they have to pay more for existing services or to actively engage in creating a new project. Social innovation add a bonus: that of finding a group of peers with whom they share these needs. I think it's a community and it definitely feels like one, nobody is in a rush. We're always outside chatting to other parents after nursery. Many days we'll meet when we pick up the kids and go to the park or just outside and I mean it's like a community in that … we all have to be involved in running of *project*. Cleaning, looking after the place, helping organize things; I also think that kind of school attracts the kind of people who are interested in being part of the community, not just dropping the kids off and then picking them up. It's not like a babysitting service, you like this kind of place and choose it because you want to be involved. (M5, free-education nursery) Mothers who participate in the social innovative projects are also very keen to highlight that they have chosen these projects because they do not believe that such young children should attend an institution. They favour the loving environment which all these projects are based on, which is fostered by educators who prioritize creating strong bonds with the children and helping to build a community with the parents, in line with the free-education principles. In fact, the projects offer an open environment for parents to participate in the care of their young children: they reduce the distance between care in the home and care in institutions, thanks to the possibility of caring-with offered by the community that sustain these projects, composed of both parents and educators (Tronto 1994). I wouldn't want my son to go from being with us all day to going to an institution with a headteacher, teacher, cook, cleaning lady. It's a huge system that he can't understand; it's better to first go through something in between before he gets to something more institutional, corporate, school. (M12, free-education nursery) Social innovations also allow mothers to fully engage in the education of their children (Laureau 2003), letting them actively participate in the coproduction of the service they choose (Grimm et al. 2013): helping organize the nurseries/care groups or staying longer with their children at the childminder's. This possibility of participating is valued for several reasons: to learn more about how to educate their children, to be there more for their children in those important first years of their development, and to feel less alone in the endeavour of parenting, thanks to the community of care made up of educators and other parents.
And I also liked the idea of being able to spend time in the space and get to know it and see the routine. […] And I wanted to know what went on, and I saw that. And I shared this experience with other parents … So there is a relationship created there. I think I wanted to know what went on, yes. When I understood that it was a safe place, I felt free to look for a job. And I found one. (M14,Care Group) What can mothers do to respond to their wish to carry out their first years of mothering in a fulfilling way? The institutional context does not make it easy for these women to act according to their values. When the interviews took place (end of 2019-beginning of 2020), the statutory maternity leave in Spain was 16 weeks from the birth of a child, paid at 100% of the mother's salary. Additionally, fathers could take up to 8 weeks of paternity leave; this is the father's individual and non-transferrable right. Since 2019, it has been compulsory for fathers to take at least 2 weeks leave, and this was extended progressively. In 2021, when the reform had been fully implemented, the mothers' and the fathers' leave was the same length (with at least 6 weeks compulsory for the father as well). The father's leave is not cumulative with the mother's leave. No paid extension is foreseen for maternity leave. In addition, there is an option for both parents to request unpaid parental leave until a child is three (Koslowski et al. 2020). The mothers criticize the short 16-week maternity leave: they affirm that it is not long enough to fully experience mothering in the early months of a child's life, and that they could not bear to leave their child in an institution such as a nursery at 16 weeks old. At the same time, they say that extending the father's leave does not respond to their wish for more active, present mothering, at least over the first two years of their children's lives. They claim the prevalence of the mother's role in the early years.
He has now a lot of work and we are sharing childcare based on the work we have, not because I am the mother, and he is the father. Even more now that they are 1 and 3 years old. When children are younger, it's trickier: I should be the one who stays with them, because they need me, because if I try to leave them with him and then go out, then that night they don't sleep. Because they are little and they need their mother. (M9, free-education nursery) Because of this, some mothers opted to extend their maternity leave with some unpaid leave, but only for a few months or at most until the child was one year old. When their leave came to an end, the majority of the mothers interviewed went from full-time to part-time work in order to participate more in the care of their children and to combine their working activity with their involvement in the care projects. I really like my job. The truth is that I like it, but right now, on my scale of priorities, my daughter comes first. Work is very important, but … I don't want it to be … Because in the end, if the day has 24 hours in it and I work 8 of them, I'm with her 2 to 3 hours. And it seems very unbalanced. OK, she goes to sleep at 8 in the evening, she doesn't have our schedules, it's not like *partner*, who I see for more time. I think that at this moment in her life … 3 or 4 years from now, I think it'll be different, but at this moment I feel that I have to be there for her more. (M1, free-education nursery) It's true that it is a break in my professional career, I mean, there have been things that happened while I was on leave of absence, promotion possibilities that I could not opt for, because I was on leave. But of course, um, … that's like not being in the right place at the right time. So, yes, perhaps I would have had internal promotion opportunities that I could not take due to this leave of absence. (M17, free-education nursery) However, most of the mothers seem unpreoccupied about the possible outcomes of their intensive mothering choices, both in terms of their future agency and in terms of gender equality. On the one hand, they risk excluding fathers from the education of their children by taking on a predominant role in parenting their children, made possible by choosing a reduction in working time. On the other, the impact of extended leave on their careers is underestimated, also in terms of their future independence from their partners. In fact, to fulfil their wish to spend more time with their young children, some mothers decided not to return to work until their children reach three years of age, opting instead for unpaid leave, despite the consequences: economic dependence on their partner, damage to their careers and less income for the household.
Since no options for extended paid parenting leave are contemplated in the Spanish welfare system, some mothers use unemployment benefit strategically to have some income while caring for their children. This implies practices that are bordering on illegality, such as asking a complicit entrepreneur to sign a contract with them and paying the social contributions themselves in order to reach the level of contributions required for claiming unemployment benefit, or asking their previous employer to dismiss them instead of them resigning. They do not feel ashamed about these practices; instead, they feel like their right to be a committed mother has been betrayed by the current system of leave which deserves to be taken in for its incapacity to grant them the right to mother their children.
In order to bring up *child*, I was working with my sister-in-law, and since I had accumulated unemployment benefit from previous jobs, I took it. It's a way of getting some income while I raise him.
[…] The last 7 months of working with my sister-in-law, I paid the social security expenses myself; on top of the twenty hours I worked, I paid twenty more of social security, to have 40 hours paid up and have the option of unemployment benefit. (M7, free-education nursery) So I found a job for a month or 2 months and with that I was able to claim unemployment benefit. Now I am running out, and now that I have one child at school and the other at nursery, I am looking for work again. But I decided to put my working life on hold to be able to spend at least the first year, the first 2 years with the children. (M9, free-education nursery) The choice of putting work on hold to have a primary role as carer is the privilege of mothers of a certain class and economic status. Women who can choose this option have family backgrounds that can support their choices, with funds generally coming from either their partners or their parents. In other cases, women who work from home or are selfemployed have the flexibility and the reduction in hours they need in order to participate in caring for their children. Many of the participants in these projects are expats (8 of 18 mothers in our sample): they often have international salaries that are higher than Spanish salaries or more flexible schedules as freelancers, allowing them to access and participate in these types of projects without putting the financial security of their families at risk.
Well, of course, first, you need to be a family with a decent income; second, you need to have a job that allows you to do fewer hours, because with an 8-hour workday you can't take part in a project like this. So it is elitist, really, if you have those two conditions. So it's like that, you don't need to be rich, but you have to earn enough to say: well, 70 euros more doesn't matter, it won't change my life … But of course 70 euros for some families is a lot … (M6, Childminder) These projects are only accessible to families that have a high enough income. But some middle-class families decide to cut down on other types of consumption (e.g. travel, cars, houses) in order to afford the increased costs of the childcare incurred by these types of projects. I don't think everyone can afford it, but a lot more people could, really. Because it depends on your values and your priorities. You can have a very old car; we have a crappy car that we can't even use now, with this new pollution thing, you know? We go on holiday to the family village in Spain; it's a question of values and priorities. There are families of four who fly off on holiday every year. Well, we don't like that kind of tourism anymore; also, we use the money for other things. For education. (M18,Childminder) In sum, socially innovative projects are only available to those who can pay their price: individual-level sacrifices for mothers (in terms of their careers) or sacrifices at the family level (in terms of total household income).

Discussion
Our results show that the ECEC projects we have analysed can be defined as social innovations, both in terms of their means and their ends. On the one hand, they are processes built upon the collaboration, collective strengths and participation of their members (Tiwari 2017). On the other hand, these projects can also be conceived of as forms of empowerment since, through these processes, participants can acquire agency, that is, the ability to act according to their own values (Sen 1985;Gangas 2016). However, the study reveals a surprising kind of individual empowerment: rather than merely allowing mothers to further their careers, balance work and family life or earn a living, these projects allow mothers to experience fully involved parenting, learn mothering skills in a community and participate fully in the care of their children. This is their main reason for choosing these innovative projects, and it is in this sense that the mothers in the in the social innovation feel empowered, even though they are aware of the costs involved on other fronts. The mothers who choose these projects are highly educatedthey all hold university degrees. They value their jobs but are willing to sacrifice their careers (at least partially) to spend time with their children during their early years. Additionally, in the light of the analysis of this study, some educatorsmost of them femalethat decide to engage in these innovative collaborative projects could be said to be exercising agency, because of their personal and political commitment to these innovative projects.
The need for social innovation in ECEC arises because the institutional context does not allow for prolonged maternal care. In fact, the projects studied here fulfil mothers' (or parents') needs and demands that are not covered either by the market or the public sector. These demands concern their wish to be involved in the highly time-consuming activities of caring for and educating their children in their early years. The process of empowerment cannot occur merely at the individual level: a favourable institutional context may act as a facilitator (Ibrahim and Alkire 2007), or when institutional support is lacking, as we argue in this paper, the role of facilitator can be fulfilled by social innovation (Ziegler 2018). Citizen engagement in these social innovations in public policy is high: in the case of care groups, they actively work to ensure the provision of the services they need, with the risk of (self-) exploitation that co-production sometimes implies. Community participation in these initiatives may palliate the risks derived from the re-privatization of the responsibilities for social reproduction.
As we point out above, these innovative projects not only have costs for mothers, in terms of their careers and the corresponding loss of financial and personal independence, but also for the educators involved. An overwhelming majority of childminders, care group teams, and educators leading free-education nurseries are female; in addition, their field is under-regulated and suffers from poor labour conditions. Thus, these social innovation projects do not empower the women that work in them.
This situation deepens both the traditional gender divide and inequalities in society. Our analysis shows that these social innovation projects may contribute to keeping social reproduction tasks both gendered and invisibletasks traditionally performed by women in the household are thus transferred to the community, where other women continue to perform them. This reinforces the conclusions from the existing literature (Bezanson and Carter 2006;Keller-Garganté 2017). But because there is little or no policy framework of reference for these kinds of social innovation initiatives as service provision models, these projects find it hard to become resilient and viable. Thus, while these arrangements favour individual empowerment (achieved by the mothers), organizational and community empowerment are not clearly promoted (Haugh and O'Carroll 2019); workers are not empowered and there seems to be no impact on social structures of inequality.
Even though we see capabilities developing on the basis of the relationship between resources, agency and achievements (Gangas 2016), these socially innovative projects reveal a clear socio-economic bias. Because of their price (similar to private provision) and their associated costs in terms of halting career progression, only mothers who are sufficiently well-off or self-employed (resources) have the option to participate in projects (agency) that allow them to fulfil their values concerning motherhood (achievement). Since this is their main reason for choosing these innovative projects, it is in this sense that they feel empowered.
The lack of institutional support may not only involve individual costs but also social ones, as society as a whole is precluded from the potential benefitsnamely, enhancing women's capabilities irrespective of their socio-economic levels. Empirical studies are challenging the rhetoric of social innovation (Pol and Ville 2009). They imply that rather than being democratizing services, they generate a Matthew effect for which only certain mothersthe ones who were initially more empowered, active and well-offaccess these projects and benefit from them. Furthermore, the most innovative and disruptive casessuch as the care groupsare only able to involve a very small number of participants, thus showing limited capacity for scalability (Cukier et al. 2018). This puts into question whether the proposed solutions can be implemented to meet the needs of the proposed targets on a wider scale.
The idea of social innovation also stresses the involvement of multiple actors and the participation of citizens in the definition of social policies and delivery of welfare services, especially at the local level. The rhetoric of social innovation has proclaimed to democratize welfare through the increased engagement and participation of recipients, but has also been applied instrumentally in the context of retrenchments to mask the progressive de-responsibilization of public actors in the provisions of goods and services (Oosterlynck et al. 2013). The support that municipalities give to innovative welfare provision initiatives might become problematic when accompanied by a reduction in direct public investment in public ECEC solutions. The reduced role of the state is supposedly partially compensated for by civil society and private actors, but as we see in this paper, it comes at the price of favouring those who are already wealthy.
In conclusion, in a context in which institutional support is absent, our analysis shows that socially innovative services can be chosen only by mothers who have the opportunity to enjoy flexible working hours or extended leave, those who can afford to take a break in their careers, and those who can pay the high prices charged. This situation reproduces and deepens social inequalities instead of mitigating them.

Conclusions
Social innovation in early childhood education and care is on the rise in the city of Barcelona, in a comparable way to similar but longer-standing projects in other cities across different countries. Although these various initiatives have different characteristics, they all offer mothers who choose them a way of fulfilling needs and demands that institutionalized (public and private) services do not fulfilnamely, they allow mothers to remain highly involved in the early care and education of their young children. In that sense, these women are empowered because their capacities to pursue their values are enhancednamely, through the virtuous resources-agency-achievements circle as formulated in the capabilities approach. However, because these innovative projects have no public or institutional support, women who would like to participate in them face a trade-off between pursuing their values regarding their children's early care and education and the high costs for their professional careers and financial independence. Thus, participation in these projects is not inclusive of all socio-economic profiles, as only parents who can afford to choose them (resources) may end up participating in social innovation (agency).
We should also highlight that the empowerment offered by these projects remains at an individual level: in line with previous research, we see that participation in social innovation is not enough to generate a systematic change that transforms the underlying social structures that create gender inequality. Thus, individual empowerment (weak tradition of social innovation) does not lead to community empowerment (strong tradition of social innovation). Socio-economic and gender inequality is deepened because not only are most educators in these services women, but also because they carry out their activities in unstable, under-regulated contexts. Thus, women (as educators or mothers) are still the ones who take care of children, reproducing the model of male breadwinner/female carer. All these individual costs are also costs for society as a whole, since they reduce the capacity of women to contribute to economic growth, and they maintain traditional gender inequities in both paid and unpaid work.
The social innovation projects studied in this research could be understood as pursuing change within the individual sphere and through informal channels. Self-selected citizens who participate in them see their capacities to pursue their values concerning care and education of their younger children enhanced because they already have the resources and capacities to do so, and can afford to bear the high costs incurred to their careers. Far from considering the benefits from social innovation a unique case, further research could analyse to what extent mothers taking their children to traditional public or private institutionalized pre-school services feel empowered.
Keeping these innovative projects within an individualized sphere makes them invisible to society as an alternative that could potentially question some aspects of current institutionalized models of care and education, and could ultimately influence public policy. Some of these aspects include participation, democratization and empowerment regarding public and private services. Thus, further research could explore the extent to which change at the institutional level could increase women's resources and opportunities, working as a trigger to enhance their capacity to pursue their diverse valuesboth as individuals and as a collective. Of course, such formal changes would have to include a problem-based policy design, namely, an approach that is multi-sectoral. In this sense, changes in ECEC regulation cannot empower women if key transformations are not made in labour market regulations and in many other areas where gender inequalities are particularly acute.
Coordinator of the Master in Public Management organized by UAB, Universitat de Barcelona, and Universitat Pompeu Fabra. Her research interests are: public policy analysis, public administration and public management reform, welfare policies, welfare state and decentralization. She has worked as free-lance international consultant for the Inter-American Development Bank. Lara Maestripieri holds a Ph.D. in Sociology and Social Research, granted by the University of Trento (2011). She is distinguished researcher 'Ramon y Cajal' in the Department of Political Science at the Universitat Autònoma de Barcelona. She is also a member of the Research group Analysis, Management and Evaluation of Public Policies (AGAPP), through her affiliation to the IGOP (Institute of Government and Public Policies). Her main interests of research concern social change in post-industrial society and in particular: public policies and intersectionality, economic insecurity in Southern European countries, social innovation and emerging professions.