This version of the paper does not include the Abstract, Bibliography or Appendix. Please refer to the PDF Download for the full text.
Cash or Food Aid? Assessing the role of resource-based transfers in achieving female empowerment and gender equity in social protection programmes (70)
Author: Shefali Shah
Chapter One
1.1 Introduction
The end of an era of structural adjustment policies by the nineties marked a discursive shift within multilateral institutions that oriented towards a renewed focus on global poverty. The tendency to associate women and female-headed households as the poorest and most vulnerable has had the effect of making women synonymous with poverty alleviation (Razavi, 1999). However in alluding to a “female face of poverty” across developing economies, this has been to the detriment of a contextual understanding of gender roles and relations at a household and community level (Walsh, 1998:na). It is the purpose of this research to assess the relative merits of cash and food aid in mediating intra-household gender relations and power asymmetries, and the wider impact on female empowerment. As social protection programmes gain momentum across the developing world as a means for poverty reduction, the ways in which household dynamics shape poverty is fundamental to understanding how far resource based transfers strengthen the position of one gender over another. It is this authors view that food and cash transfers as complementary aid instruments reduce intra-household gender conflicts over scarce resources whilst augmenting household welfare.
Despite the UN’s international commitment to gender justice from the mid-Seventies, progress for women has been contradicted by a growing ‘feminisation of poverty’ (Jahan, 1995). This research applies the analytical framework devised by Rounaq Jahan to the Poverty Reduction Strategy Paper that guided the formation of Nicaragua’s Conditional Cash Transfer in 2000. The purpose of this is to assess how far women have been integrated into or set the agenda of development policies at the international level. The importance of this is to contribute to a better understanding of how gender issues at the macro level have been mirrored within social protection programmes on the ground. The view presented here argues international policy has followed an integrationist approach to gender mainstreaming.
Conditional Cash Transfers have grown in popularity since the success of Mexico’s Oportunidades (formerly Progresa) that became a model for replication across Latin America and developing economies. Whilst gender equality was made a key feature of programme design specifically through targeting women with cash, the wider goals of CCTs have sought to change the behaviour of poor households. Evaluations of Oportunidades have been criticised for the risks to women as a result of their explicit focus on investment in human capital. Yet the continued proliferation of such programmes particularly for low-income economies lacking institutional capacity is a cause for concern where programme design fails to promote gender awareness (Molyneux, 2006; Bradshaw and Viquez, 2008). This research thus draws on Nicaragua’s Red De Protección Social, a conditional cash transfer modelled on Oportunidades in promoting poor families’ investment in human capital. This is related to Ethiopia’s Productive Safety Net Programme (PSNP); a Public Works programme providing cash or food for labour, whilst the humanitarian Food and Cash transfer (FACT) initiated in Malawi assesses resource-based transfers within a development and emergency context.
In contributing to a limited literature this study seeks to demonstrate in what context cash over food aid or indeed food aid over cash can promote female empowerment and gender equity. Empowerment is defined as women’s “choice, agency and power” to challenger their subordination (Kabeer, 1997 in Razavi, 1999: 419). It is the purpose of these case studies to provide an in-depth contextual understanding of how resource transfers can define gender relations across Sub-Saharan Africa and Latin America and thus emphasise the study’s reliability. Whilst the case of Malawi reflects a humanitarian emergency initiative, the value of its conclusions has the potential to revolutionise the design of social protection programmes in both development and emergency contexts.
This study is organised as follows. Section two provides a conceptual and empirical analysis of cash and food aid and intra-household dynamics arguments, followed by a background to women in development and the meaning of empowerment. Section 3 adopts a case study analysis of the social protection schemes of Nicaragua and Ethiopia and Malawi’s Food and Cash Transfer in addition to a broader analysis of gender awareness in development policy. Section 4 concludes.
Chapter 2 – Resource Transfers, Household Dynamics, and Gender Equity: A Theoretical Background
2.1 Cash Transfers or In-kind Transfers?
Debates over the relative effectiveness of in-kind food aid as a tool for poverty reduction and food security has led to a decline in its “absolute value and relative importance” since the mid-nineties (OECD, 2006: 11). Critics questioned the cost-effectiveness of food aid and its ability to reach the poorest, whilst empirical support for cash transfers was ascertained by poor families’ tendency to exchange in-kind aid for cash, indicative of households’ preference for non-food items (Barrett and Maxwell, 2005). The fact that foods form a basic human right and cash a means of livelihood security lends support for the implementation of cash and food transfers in both development and emergency situations. Research on the benefits and limitations of cash transfer programmes has grown since the implementation of Latin America’s Oportunidades, though less is documented about the role of food and cash transfers in poverty reduction or mediating gender tensions. Whilst the adoption of cash in development situations has been widespread, there remains little conclusive evidence of the benefits and limitations of cash transfers (Kebede, 2006). Further whether cash or food is responsible for changes in behaviour or this is attributable to conditionality is equally inconclusive. Nonetheless a greater understanding of the benefits of cash and food as complementary aid instruments in promoting female empowerment and gender equity is fundamental to increasing the effectiveness of social protection programmes. To speculate, the limited literature in this area could be attributed to the lack of attention to monitoring intra-household gender dynamics at programme implementation. This was the case in RPS where less evaluative reports discussed progress than that of Oportunidades (Bradshaw and Viquez, 2008). It can be further argued that the explicit attention to human capital and results-based evidence has been to the corollary of gendered analyses.
The economic logic behind food aid purports to its role in protecting families’ investments in the health and education of their children (Barrett, 2002). Critics contend it has a negative impact on local food production, greater potential for misuse by intermediaries, delays in reaching beneficiaries, a cause of heightened aid dependency as well as poor targeting (Gelan, 2006; Barrett, 2002). Food aid was similarly labelled as paternalistic in that it gave less choice to families as consumers whilst cash could be rapidly delivered, involved low administrative costs and offered households greater diversity and choice in terms of expenditure (Standing, 2007). Further it promoted investment and the accumulation of strategic assets at the household level with greater potential for sustainable poverty reduction (Devereux et al., 2006). In Ethiopia, cash transfers generated multiplier effects that benefitted not only recipient households but also market producers and sellers who did not qualify under the social protection scheme (Gelan, 2006). However the potential for cash to be misused based on stereotyped assumptions of men as irresponsible lends support for food aid where evidence points to a higher marginal propensity for poor households to consume food transfers than to spend cash transfers on food (Devereux et al., 2006). Similarly, growing concerns over the efficacy of cash transfers in increasing food price inflation have questioned their ability to mitigate the impact of adverse shocks (Sabates-Wheeler, 2009). The global food price crisis of 2007-08 increased food insecurity in Ethiopia where cash stipends were not increased with rising inflation (Ibid.). The insecurity of cash as a means for survival within unstable developing economies, and particularly within a fragile global economy has important connotations for the effectiveness of CCT schemes.
Though limited, the research on food aid and female empowerment has stressed the benefits accruing to children where mothers are food aid beneficiaries (Khogali and Takhar, 2001). This was further supported by household expenditure surveys that indicated women are more responsible and inclined to spend resources on their children (Walsh, 1998; Bradshaw and Viquez, 2008; Soares and Silva, 2010). However like cash transfers, the effect of this on women’s own welfare and their existing obligations has implications for their power, agency and choice. Walsh argues that allocations of food aid must complement women’s work and their responsibilities as a means for alleviating their burden (1998). Similarly, even where women receive food aid there is inconclusive evidence on the equality of its distribution amongst children (Ibid.). In Mexico’s Oportunidades higher incentives were given to promote girls’ enrolment in school, but girls remained more likely than boys to drop out (Bradshaw and Viquez, 2008). How far food transfers succeed in improving the status and subsequent empowerment of women within the household and community is important for the application of these models to other contexts.
Studies have affirmed women’s preference for food or food-for-work schemes as it supplements their role in producing and preparing food. However these trends have been context specific, in Bangladesh female-headed households (FHHs) favoured cash over food aid as they had freedom of control over it (Walsh, 1998). On the other hand, where women had limited access to employment opportunities there was a greater need for cash over food aid (Ibid.). Sen asserted if the control of and access to resources was linked to the individual who contributed most to the household or held most power, women would accrue greater bargaining power from being food aid beneficiaries (1989 cited in Walsh, 1998). Yet in Indonesia the Program Keluarga Harapan (PKH), a pilot conditional cash transfer initiated in 2007, had no significant impact on altering the gendered division of labour or intra-household gender relations by making women cash beneficiaries (Arif et al., 2010). This challenges the assumption that cash would empower women as recipients (IADB cited in Bradshaw, 2008a). Similarly the priorities of poorer families were less oriented towards the education of children than to immediate household needs (Arif et al., 2010).
Food aid in social protection programmes has had the effect of emphasising traditional gender roles such as women’s role in food production (Soares and Silva, 2010). The fact that social norms inform gender roles suggests the need to challenge social stereotypes. Women in Oportunidades reported greater autonomy, status and self-esteem from cash transfers (Latapi and de la Rocha, 2009), however these analyses point to concerns over the dubious assumptions in which giving cash to women have been made, as well as the context to which they have been applied. Whilst food aid has been found to reinforce stereotypes and cash aid increases women’s voice, applying these models to developing economies requires attention to women’s existing roles within the household and society as well as intra-household gendered power relations. As Luccisiano found the attendance required by women at social talks represent social activities and may affect the way women are perceived in society (2006).
2.2 Intra-Household Gender Relations
Addressing the ways in which resource-based transfers can strengthen or weaken the position of women over men has been paid little attention within CCT literature that can be arguably attributed to the limited monitoring of intra-household dynamics in safety net programmes. However to assert this is an easy task would oversimplify the ‘subtle negotiations’ that occur between men and women at the household level (Razavi, 1999:420). Moreover as Dijkstra contends measuring the attainment of bargaining power within intra-household literature remains a significant methodological and empirical issue (2011). Nonetheless without strong contextual analysis, the gendered relations that shape individual experiences of poverty within countries and societies are susceptible to contentious generalisations that do little to augment female empowerment.
Research stipulates the potentially negative impacts of cash transfers on gender relations being; a reduction in women’s ability to control cash in contrast to in-kind aid, and the increased susceptibility of cash to ‘anti-social’ expenditure (Slater and Mphale, 2008:1). In Slater and Mphale’s gendered assessment of World Vision’s cash transfer programme, initiated in Lesotho in 2007, their findings purported to the role of cash in reducing gender conflict (2008:10). This was attributed to Lesotho’s history of migrant labour that has been responsible for women’s management of cash incomes. In the situation of cash transfers, household’s use of a ‘whole wage system’ has meant that women control income regardless of whose name it is in, and have been accepted by men (Ibid.). However the reliance of households on remittances has contributed to increased gender tensions directly relating to men’s redundancy from mining companies and subsequent disempowerment (Slater and Mphale, 2008). This has been paralleled with an increasing trend for women’s entry into paid factory work that has tilted power back towards women. Whilst this has not created new conflicts, it highlights the significance of a contextual understanding to women’s empowerment.
The difference in coping strategies of men and women particularly at times of crisis further implies the necessity of a gendered analysis. Where men reduced their level of consumption to provide for their children rather than their wives, women tended to consume less for the benefit of both their children and husbands (Slater and Mphale, 2008). Whilst all women agreed gender tensions increased during adverse shocks, evidence showed that cash could significantly reduce gender conflict where it allowed for the purchase of food and other needs. Where men are reported to waste cash on alcohol and cigarettes, this was not supported in Lesotho. Instead the differences in coping strategies were more likely to be a cause of tension as men objected to women’s tendencies to share food with their neighbours (Ibid.). Because women controlled resources this was to men’s dissatisfaction (Ibid.).
Research into intra-household dynamics has important implications for the assumption that making women cash or food beneficiaries will automatically result in the equitable distribution of resources. In Vietnam, giving cash or food to households reduced tensions because it alleviated pressure to obtain food and meet other needs (Holmes and Jones, 2010:22). However in India cash transfers to women had no effect on preventing domestic violence, whilst the conditional entry of women into employment increased household tensions where men resented women’s inability to perform their household duties (Ibid.). In the context of a patriarchal society, this emphasises how women’s empowerment is significantly influenced by the social and cultural norms that may perpetuate female stereotypes. It is interesting no less that in Peru’s Juntos programme, a CCT based on human capital investment, where women and men were incorporated within community meetings to inform both of their rights and reduce domestic violence, women’s enhanced agency, power and reduced dependence on men stemmed directly from their participation in community talks rather than the cash itself (Holmes and Jones, 2010:24). The varied role of resource transfers in heightening women’s autonomy contends to the fact that targeting women in social protection programmes has been based on limited evidence (Dijkstra, 2010). Be it food or cash, each plays a significant role in reducing intra-household tensions were resources are scarce. However where cash is provided to men this can increase conflict particularly where women have greater agency. This would suggest the potentially positive outcomes of conditional employment programmes where men and women both contribute to the household.
2.3 Women and Empowerment
Since the 1960s as women’s movements challenged the social stereotypes that perpetuated their disadvantage in developed economies, economic and efficiency arguments for women’s inclusion across the developing world promoted their productivity as key to their integration (Razavi and Miller, 1995). The United Nations declaration of 1976-85 as the Decade for Women institutionalised the promotion of women’s rights and status within the global community, a turning point in achieving gender equity and female empowerment. However by the early Eighties Ester Boserup, an influential feminist thinker, challenged the welfare approach under development agencies and NGOs that continued to depict women as mothers or wives (1981). Giving rise to the Women in Development (WID) movement, women were promoted as ‘agents and beneficiaries of development,’ such that economic change was central to female empowerment (Jahan, 1995: 12). Proponents argued that the subordination of women in society stemmed from their barriers to the productive sphere (Razavi and Miller, 1995). This was critiqued however for its depiction of women as an already marginalised group that should be added into development policy (Porter and Sweetman, 2005).
Similarly for Goetze, efficiency arguments for women’s integration into development discourse stressed what development required from women, rather than women’s needs within development (1994). Such a view has not been far removed from CCT programmes nor international development policy mandates, whereby women continue to be “working for development,” (Chant, 2008 in Molyneux, 2008:190). Feminists alike concurred that integrating women into exploitative and unequal systems did little to challenge their subordination, emphasising the need to address the social construction of men and women’s roles within development agendas. By the end of the ‘Decade’, gender mainstreaming in development represented a paradigmatic shift from integrating women in development to putting them at the centre of it. Though how far development policy and practice has moved in line with this shift has been central to women’s gains within social protection programmes.
The UN Economic and Social Council define gender mainstreaming in development as a method of making “women and men’s concerns integral…to the design, implementation, monitoring and evaluation of programmes” (1997 in Porter and Sweetman, 2005:2). This was supported by Gender and Development (GAD) discourse that argued gendered hierarchies restricted women’s opportunities and access to resources (Sharma, 2008). However Elson states the implementation of GAD in development policy has been in addition to existing policies rather than dictating it (2004). Rounaq Jahan conceptualised two distinct approaches to gender mainstreaming to challenge the claim that women’s empowerment and gender equality were at the centre of development policies (1995:12). Her pivotal study offers greater substance to the present analysis of CCT programmes in understanding how far women benefit under social protection programmes in terms of their status, autonomy and decision making power. In her gendered analysis of the macroeconomic policies of the World Bank (WB), United Nations Development Program (UNDP), Canadian International Development Agency (CIDA) and the Norwegian Agency for Development Cooperation (NORAD) she argued that the means for addressing WID and GAD had been incorporated within the project frameworks of both Canada and Norway, yet the WB had not prioritised gender at each level of programme design, monitoring or evaluation (1995:68).
This paper uses Jahan’s work to apply her framework to the context of Nicaragua to assess how far the incorporation of gender issues into development policies and programmes has followed an integrationist approach or an agenda-setting approach (1995: 13). Under gender mainstreaming, an integrationist approach reflects the framing of gender issues within development policy so that women are a feature but not a primary goal. An agenda-setting approach would imply the re-conceptualisation of development policy entirely from a gender perspective, prioritising female empowerment through directly targeting women’s decision-making power (Jahan, 1995). Donors and policy practitioners claim to have adopted both an integrationist and agenda-setting approach to mainstreaming gender, prioritising integration and mainstreaming before gender equality and women’s advancement. This they contend is based on the need to ‘set the agenda’ [through women’s integration] before changing it (Jahan, 1995: 21). To interrogate this view, an integrationist approach implies women have been treated solely as “participants and beneficiaries in development policies,” (Jahan, 1995: 24) rather than decision-makers through their enhanced agency, voice and power.
To strengthen the integrity of such an argument this study incorporates the analytical framework of Molyneux and Thomson (2011) to examine how decisions at the policy level dictate programmes on the ground. They assert that sensitivity to gender in CCTs would imply the allocation of resources “directly seeks to strengthen women’s capabilities; social and economic empowerment is an explicit goal; family-friendly policies that account for women as care givers; promote gender equity by sharing responsibility amongst men; women’s voice in the programme design, implementation and evaluation” (Molyneux and Thomson, 2011: 199; Molyneux, 2008). As the PRSP that guided Red de Protección Social pertained to being “country owned, designed through national participatory processes with the assistance of the WB,” (Bradshaw, 2008a: 196) it would be expected that gender is a higher priority particularly as it is modelled on the gender aware Oportunidades. The inattention to gender no less would give credence to the contention that RPS has been based on an integrationist approach to mainstreaming gender in development. With the replication of CCTs based on existing, successful models such as RPS this analysis at the micro-level permits a greater understanding for the universal applicability of CCT schemes to other developing economies. A limitation to Molyneux and Thomson’s framework arises in the subjective definition of empowerment, thus its meaning here is discussed in 2.2A. The extent to which women’s positions within safety net schemes has been an outcome of a gender blind political economy has important consequences for women’s empowerment.
2.2a Defining Female Empowerment
Empowerment has become a buzzword within development discourse and policy circles, alluding to notions of enhanced self-esteem, self-actualisation and status with participatory governance. However as Jahan argues, self-empowerment or improving women’s access to resources is not a suffice means of challenging the structures that perpetuate gender inequality in the first place (1995). Women must become their own agents of change rather than the “passive recipients of welfare” stereotyped under the Women in Development discourse (Razavi and Miller, 1995: 4). The shift to GAD re-conceptualised the definition of empowerment as a ‘process of awareness raising and struggle’ to challenge gendered hierarchies (Sharma, 2008: 7). For Rowlands the ambiguity in the term power itself has connotations for the way in which empowerment is conceptualised and addressed within development programmes (1999 in Campbell, 1999). Kabeer similarly identifies it under categories of power; disempowerment and empowerment arguing individuals cannot be empowered without having been disempowered (2001). Rowlands adds to this where the definition of power is thus conceptualised as a means of challenging oppression and inequality, only then can empowerment be realised (1999).
Sharma debates the neoliberal conception of empowerment as a de-politicization of poverty versus a means of re-establishing power where it belongs, the former viewing poverty as something to be technically managed (2008: xx; Ferguson, 1994). For her, empowerment represents a governmental strategy to control and direct the behaviour of women towards specific ends (2008). The UNDP specifically works to ensure gender equity and women’s empowerment is met not just as a human right but a means of eradicating poverty as part of the Millennium Development Goals. This is based on the idea that women’s access to education can reduce fertility promote growth and poverty reduction. However this views women as a means to an end within development circles rather than an end in itself, with little concern for women’s welfare (Jahan, 1995). This supports the contention that definitions of empowerment within international development agencies have not sought to give power back to women but instead promoted a ‘feminisation of poverty alleviation’ (Chant, 2003 in Bradshaw and Viqeuz, 2008: 827). The expression of empowerment used within this paper expands on these author’s contributions in affirming the view that empowerment is not something that can be “done ‘to’ people or ‘for’ people…[but must be] undertaken with women,” (Rowlands, 1998 in Sharma, 2008: 7). Empowerment from here on is thus defined as power, agency and choice (Kabeer, 1997 cited in Razavi, 1999); the ability of women to challenge their subordination through self-actualisation, bargaining power and decision making power.
The claim that CCTs empower women in any given context has been a strong mechanism in gaining international support for conditional cash and food transfers. Giving cash directly to women is justified by the rationale that women are more responsible than males and subsequently more inclined to spend the cash subsidy on their children (Bradshaw and Viquez, 2008). Such imperatives have been reinforced without question by programme designers, based on an economic rationale that argues the investment in the human capital of children is key to economic growth (Fiszbein and Schady, 2009). However whilst one cannot dispute promoting the welfare of children across developing economies, there has been a level of criticism directed at the negative impact upon women. Molyneux defines what has emerged as a ‘maternal model of care’ that reinforces the gendered responsibilities of women as caregivers and within the household (2006: 22). Women in Mexico reported greater status and self-esteem as a result of their participation within Oportunidades yet Latapi and de la Rocha’s ethnographic study found women’s attendance required at talks and meetings, their participation in activities in addition to family and work obligations was a burden borne solely by mothers (2009). This suggests women gain from being cash beneficiaries but such benefits are negated by the conditions attached to cash. Importantly can any feelings of empowerment reported by women in CCTs truly equate to those felt given the opportunity to earn such money within the labour market? (Molyneux, 2006).
Critically, the assumption that men are irresponsible in terms of their use of resources has only heightened the burden placed on women to meet the conditions of cash transfer programmes. Evaluations of Oportunidades have examined the objectives of children’s enrolment in education and health programmes, overlooking the effect on women and in transforming intra-household gender relations (Adato et al., 2000). Where the male role of ensuring their household’s survival is threatened, this has contributed to increased tensions (Molyneux, 2006). However whilst the burdens placed on women have forced dropouts, the low rates of attrition in Oportunidades suggests a recognition by women of the value of CCT programmes (Latapi and de la Rocha, 2009). Nevertheless this strand of thought alludes to the reality that CCT schemes are not a one size fits all and fail to account for the complex structures of households that do not conform to stereotyped perceptions.
The intra-household allocation of resources as part of food and cash transfers has its roots in the theory behind household dynamics. The unitary approach argues that households operate as an independent decision-making body such that targeting resources by gender has little impact on programme success (Haddad et al., 1997). The collective approach on the other hand endorses a gender perspective that stipulates cash beneficiaries play a distinct role in how cash is utilised allocated (Khogali and Takhar, 2001). Whilst CCTs pertain to a collective approach in giving cash directly to women this has come only as a result of their stereotyped image as more inclined to their children’s welfare. Furthermore it does little to suggest that women’s individual needs were built into programme design, being seen as instruments in development only. Latapi and de la Rocha argue that households cannot ‘be reduced to [a] nuclear/non-nuclear dichotomy ,’ though they function best under nuclear structures (2009). Similarly the distribution of resources within households has been influenced not only by gendered power relations but social and cultural norms (Barrientos and de Jong, 2004). It follows that where decision-making power is dependent on the income contributed by a member of the household, women have greater potential for empowerment. However in Ecuador’s Bono de Desarollo Humano CCT, women had no opportunity to contribute to the decision-making process in programme implementation or reflect on how they were being affected (Molyneux and Thomson, 2011). Furthermore it is contradictory in itself that CCTs promote women’s needs and choice whilst constricting them with conditionality. Adato argues there is difficulty in ascertaining whether women’s enhanced autonomy can be attributed to the programme itself or rather the absence of males due to out-migration (2006 cited in Molyneux, 2006). A fundamental concern within conditional cash transfers is the extent to which families have been convinced of the need to invest in their children’s education or whether receiving the cash has been a greater incentive (Molyneux, 2006).
Chapter 3 – Social Protection, Gender Equity and Female Empowerment
This chapter is divided according to two main areas of analysis. First is a critical assessment of the country specific case study to determine the relationship between cash and food aid and intra-household dynamics. This is analysed in relation to their capacity to augment women’s power, agency and choice and with it gender equality. Section 3.2 examines Nicaragua’s Poverty Reduction Strategy Paper as a means of assessing at a broader level, how gender issues have been recognised and addressed within development policies. Significant attention is paid to RPS because as an advance on Mexico’s Oportunidades, it suggests the extent to which these models have acknowledged and redressed gender bias.
Ethiopia’s PSNP draws on the literature of Devereux et al., (2006) and Holmes and Jones (2010) and their analysis of the successes and limitations of the public works programme through a ‘gender lens’. The PSNP has been accorded a highly gender sensitive design, thus it is the intention here to relate this evidence specifically to the role played by cash aid in mediating household tensions and increasing women’s empowerment. As a cash for work scheme it provides a better understanding of the role of employment in increasing women’s status. Malawi’s unique food and cash transfer programme facilitates the assessment of food and cash together as a modality for equalising intra-household gender relations and gendered power asymmetries, particularly where conclusive evidence is lacking.
3.1 Empowering Women: Cash, Food, Food and Cash?
Ethiopia’s Productive Safety Net Programme (PSNP)
The Government of Ethiopia with the support of donor financing, initiated Sub-Saharan Africa’s largest safety net programme in 2005 with the objectives of alleviating food insecurity amongst the ‘predictably food insecure’ and creating assets at the community level (Devereux et al., 2006:1; Gilligan et al., 2008). Diverging in design from traditional CCT programmes, the PSNP centred on the view that ad-hoc appeals for food aid and emergency assistance did little to overcome the asset depletion of households. Similarly it would permit a reduced dependence upon emergency humanitarian aid for those made vulnerable by their poverty (Devereux et al., 2006). Ethiopia’s PSNP functions as a Public Works and Direct Support scheme, tying cash or food to beneficiaries’ supply of labour in labour-intensive infrastructural community projects to increase household purchasing power (Holmes and Jones, 2010). For poor households unable to contribute their labour, their choice in unconditional food or cash forms their direct support (Gilligan et al., 2008).
Women’s occupation in agriculture and food security as well as their low representation in paid labour and gender-wage disparities have been incorporated within the PSNP such that one third of women are required to engage in paid labour opportunities (Holmes and Jones, 2010). This is with the intention of enhancing women’s autonomy, linked to the generalisation that access to an income equates to notions of empowerment (IADB, 2003 in Bradshaw, 2008a). An explicit feature of programme design has been to account for women’s household and child obligations through the provision of crèche services and flexible working hours to overcome their time poverty (Holmes and Jones, 2010). The PSNP is a significant advance on earlier CCTs in its recognition of the barriers to women’s empowerment. However whilst gender sensitivity has been a pervasive feature of the PSNP, there has been inconsistency in the prioritising of these services on the ground (Holmes and Jones, 2010). Furthermore the lack of attention to the role of intra-household gender relations in resource allocation questions the extent to which improving women’s welfare has been addressed.
Three quarters of the programmes 8 million beneficiaries received cash or food aid in exchange for their labour, with one quarter unconditionally (Devereux, 2006). Yet even where female labour was exchanged, stipends were directly distributed to the heads of household regardless of gender differences (Holmes and Jones, 2010). This infers programme practitioners have viewed households as a single unit and as a result overlooked the reality in Ethiopia that men and women make use of assets and resources differently. The fact that the unitary approach to intra-household resource allocation adopted here was challenged for its failure to account for gender asymmetries is evidence of an inconsistent approach to advancing women’s welfare within the PSNP. The provision of cash in this instance has served to strengthen the position of males over women in male-headed households. Women have faced uneven access to finances whilst their opportunity for enhanced voice was limited to a 30 per cent provision in the decision-making of community assets, giving credence to the view that the benefits to women as cash or food beneficiaries is confounded by unequal gender relations as well as social norms. The finding that male unemployment increased household tensions in Lesotho can be applied to Ethiopia. Though women and men entered into employment the fact that women did not necessarily receive the cash income suggests this could fuel tensions where men do not allocate resources in the same way as women. These findings suggest cash can increase gender conflicts where women are not beneficiaries, supporting the role of food aid in reducing conflict over scarce resources.
The limited literature on household preferences for food, cash or food and cash has emphasised the difficulty in knowing in what context providing one or the other is better suited (Kebede, 2006). Fifty-four per cent of households participating in the PSNP reported a preference for food only, compared to 36% for half cash half food, and 9% favouring cash only. However Devereux et al., assert this preference may have stemmed from the fact this is what they had already been provided and thus had nothing to compare it to (2006). The difficulty of implementing a control group within such programmes would explain the limited literature on the gendered allocation of resources within households. Of significance was the preference amongst wealthier households for cash, whilst poorer households favoured food and cash. This would imply that the need for food and cash has strong links to the livelihood strategies adopted by poor and wealthier households, as well as potential inclusion errors. Nonetheless, the majority of food receiving households reported consuming all of the food at home, whilst only 7% sold a proportion of it. From this it can be inferred that food aid is less likely to create household tensions as it meets immediate needs, however the fact that food only recipients were more likely to sell than those receiving both food and cash transfers may suggest otherwise. Where limited by a lack of research, it is difficult to say with certainty the reasons for some household’s sale of food aid, and in particular on what cash was spent. The fact that men and women utilise cash and value assets in distinct ways is fundamental for a better understanding of the value of food and cash transfers.
Households that expressed a preference for cash recorded their greatest spending on staple foods, groceries, clothes, health and education. Here it can be argued families’ lack of investment in education results from their level of poverty rather than the World Bank’s idea of misguided beliefs. Households in receipt of food transfers enjoyed greater income growth than those given cash, whilst transfers of cash plus food were found to offer self-reported food security, income increases as well as the ability to purchase and keep livestock (Sabates-Wheeler, 2009). This lends the support to the value of food aid when supplemented with cash as a means for achieving long-term development. This evidence also shows the value given to food aid providing greater validity for the provision of cash and food to poor families. Where Devereux et al.’s study is limited by a lack of disaggregation of resources by gender; this may be due to the non-differentiation between genders in resource transfers. In spite of this, some gender analysis can be applied to the response of participants who expressed a preference for food, cash or both (See Appendix 1). Women reported fears that cash would be spent irresponsibly thus preferred food transfers only, whilst concerns over high food prices, difficulty getting to market, and the fear that cash aid artificially augmented food prices were similarly reported by food receiving households. A limitation of food aid expressed by those favouring cash was that food aid had to be collected and brought home, suggesting a male tendency to sell food along the way or that poor households lacked sufficient labour. Cash and food together on the other hand met the food and non-food needs of households, allowed for greater food security when food prices were high whilst cash was more useful after harvests and reduced the need to sell food for cash (Devereux et al., 2006). It is clear food aid supplements livelihood strategies whilst cash permits a more effective route to sustaining livelihoods and investing in health and education through increasing purchasing power. Support for cash as a complement to food aid is justified where cash transfers are increased in line with food prices and delivery of food aid is more efficient. Of course the institutional capacity of countries is an important factor in addressing these issues.
Support for sensitivity to intra-household relations and cultural norms in the design of social protection programmes are warranted where polygamy is a norm. Where first wives were recognised as main beneficiaries in the transfer of resources, this was at the expense of second wives regardless of their participation in public works or the depth of their needs (Holmes and Jones, 2010). In terms of overall empowerment, Holmes and Jones argue that the PSNP had positive impacts on women at the individual level in terms of the enhanced opportunities for their participation in economic activities, increased knowledge, skills and self-esteem as well as contributing to male attitudinal changes (2010: 20). Forty per cent of women entered into paid labour works with both sexes having wage equality. Women reported the benefits of such employment as a greater alternative to the domestic and abusive work they often engaged in. However societal attitudes had not changed towards women workers such that male labour was more valued and given greater incentive to work (Holmes and Jones, 2010). It can be argued that women’s participation at this level has not contributed to enhancing their voice, agency or power. Similarly where women reported increased confidence and self-esteem as a result of their paid labour opportunities, this did not alter their gendered roles and responsibilities within the household. This evidence further suggests that where programme officers are not informed of the benefits of enhancing women’s empowerment, there is little hope for advancing women’s welfare. In effect no aspect of the productive safety net programme was effective in making women agents of development. The fact that these programmes seek to promote long term benefits once the programme ends is indication of a need to better assess the gendered allocation of resources within households. Women reported increased feelings of worth and gained more respect as a result of their participation in public works, whilst some noticed men shared the burden of women’s duties. However of the 8 million programme beneficiaries, these changes were witnessed amongst a minority of households. Rather than becoming the agents of change made synonymous by development agencies since the UN ‘Decade’, the advancement of women in development has been hindered by a failure to challenge social stereotypes and men’s attitudes.
Food And Cash Transfer (FACT), Malawi
The implementation of Malawi’s food and cash transfer (FACT) scheme in 2005-06 formed part of an international response to its food crisis. With CARE International, the unique design of FACT represented a humanitarian relief intervention where the provision of cash plus food offered “all the benefits of both while avoiding the limitations of each,” (Devereux et al., 2006a: vii). The innovative programme was designed to meet 50% of rural household’s food needs through a 25% food aid provision of maize, beans and oil, with 25% through a cash amount equivalent to the cost of buying that food (Ibid.). The decision was made to provide both cash and food aid in case of potential market failure to ensure nutritional requirements were met and that food reached those less able to obtain it such as female-headed households and the elderly.
The necessity of providing cash followed previous assumptions that contend it is more empowering for women; it is cost-effective and has positive impacts on local markets. Families reported the benefits of food and cash transfers in ensuring food security and to their children’s education (Devereux et al., 2006a). Under FACT men and women generally behaved in similar ways to households in Lesotho; women were more likely to share food transfers, whilst men were more inclined to lend cash to other men. Where both were received, few reported domestic violence though this is not to say it did not occur (Devereux et al., 2006a). Though women’s gendered roles have traditionally remained in the production and preparation of food, FACT had little effect on changing men’s attitudes towards shared responsibilities within the household. Even where cash and food transfers were given to the women, in many male-headed households they willingly gave the cash to their husbands (Devereux et al., 2006a). It can be argued that where the gender division of labour reinforces women’s household domain and the stereotypical male breadwinner, women have little self-esteem to assert control over cash transfers thus surrendering it to their husbands.
It is important to note that assumptions that men use money selfishly cannot be generalised to all males. In polygamous households cash was spent on their second wife and family, which couldn’t be defined as misuse. As Devereux et al., argue such situations require cultural sensitivity and attention to the complex meaning of ‘household’ (2006a). Nonetheless it was apparent that men asserted full decision making power where they were given the cash by their wives, but in 60% of cases families reported mutual discussion of how best to spend the cash. However the fact that it is difficult to ascertain the level of women’s participation within this process, for example whether they were able to assert a voice or they passively agreed does little to suggest cash transfers offer greater female empowerment when gendered power relations prevail. There is strong support for the gendered targeting of resources based on FACT and Ethiopia’s PSNP particularly in the benefits to children, however greater analysis is required to avoid encouraging a trade-off between women and children. As found in Ethiopia, the value of cash to households was significantly reduced where it was provided to men. So whilst the provision of cash to women has the potential to enhance their power and status, it cannot be asserted that this will challenge gendered power asymmetries.
Based on the studies of Ethiopia and Malawi food aid is less likely to reinforce unequal gender relations as it is entirely consumed, however in terms of empowering women it has only reinforced a traditional division of labour. Food and cash together contribute to alleviating the burdens placed solely on women with the potential for shared decision-making within the household. However where there is a concern over men’s behaviour, it is necessary to incorporate men into social protection programmes to permit a greater level of intra-household cooperation. That employment can be more empowering for women has potential for women in Ethiopia were women receive the cash, and men earn their own income. Devereux et al. suggest achieving female empowerment within cash and food transfers requires direct attention to addressing gender issues, enhancing the knowledge of community actors to support women and promote gender sensitisation training in the wider community (2006a).
Red de Protección Social (RPS), Nicaragua
Nicaragua became one of the first classified low-income countries in Latin America to implement a social protection strategy in 2000 as part of its Poverty Reduction Strategy Paper (PRSP) (Regalia and Castro, 2007; Bradshaw and Viqez, 2008). Implemented across two phases, Red de Protección Social was designed to alleviate short-term poverty within rural Madriz and Matagalpa whilst tackling long-term poverty through human capital formation (Maluccio and Flores, 2004). The success of RPS, determined through its objectives for achieving food security, health and nutrition as well as education dissolved concerns over the applicability of such programmes from middle to low-income countries. As the second poorest country in Latin America, RPS was designed according to Nicaragua’s own poverty and policy context and thus differed from Mexico’s Oportunidades in its economic growth focus, particularly where gender parity in education was not promoted through greater cash incentives to girls (Bradshaw, 2008a; Molyneux, 2006). Instead the Pay For Performance Model (P4P) incorporated private providers, as well as a supply side logistic that incentivised both cash beneficiaries and private services in exchange for performance conditions (Bradshaw and Viqeuz, 2008). This has been an innovation in addressing where the supply and quality of services fail to meet the growing demand for education and health services, but also suggests the promotion of GDP growth through the private sector that will trickle down to the poor.
Targeting conditional cash transfers to one member of a household inevitably risks increasing tensions between generations, gender relations, and relatives (Adato and Roopnaraine, 2004:68). In Nicaragua, the limited analysis of gender relations within the household has meant affirmative conclusions are difficult. Nonetheless, in some cases there were increases in intra-household gender tensions where men perceived women’s new income as cash beneficiaries as replacement for them (Bradshaw, 2008). Forty-three per cent of women beneficiaries reported their husbands withholding earnings for their own use, amounting to as much as 50% of their income in some cases (Ibid.). Moreover, Bradshaw found there was a greater tendency amongst women who earned their own income through paid labour to report their husbands (2008). Whilst this suggests these women may have greater levels of confidence, this can be the cause of increased conflict between men and women, particularly where women depend on their partner’s income.
In Bradshaw’s analysis of poverty in Nicaragua, women reported the causes of intra-familial violence as alcohol, economic problems and the behaviour of men and women within households (2002: 17). Whilst this was not confined to the regions in which RPS has been enacted, it can be inferred that men’s socially constructed behaviour has the potential to negatively impact household welfare whilst the scarcity of resources can create conflicts. The extent to which cash transfers increase tensions over how to spend resources is less documented but no less important. Targeting cash to women has raised women’s social standing at a community and household level, whilst their opportunity for enhanced knowledge and participation has altered intra-household gender relations (Adato and Roopnaraine, 2004). It is interesting that in Adato and Roopnaraine’s analysis of RPS impacts, men vehemently supported the targeting of cash directly to women in that men saw food production as the woman’s domain (2004). Men similarly asserted that male tendencies were for the anti-social expenditure of cash, thus warranting the need to provide cash to women (Ibid.). However men also stated that where food was not a priority then cash should be provided to men (Bradshaw, 2002). This reinforces the social construction of gender that has dictated men and women’s roles. This also suggests if food aid was being provided, men may demand to be cash recipients as this denotes their roles. A majority of respondents reported that cash eased conflicts in relieving the stress of limited resources, which purports to the valuable role of cash transfers supported by such occurences in both Ethiopia and Malawi. In some cases there were reports of men demanding cash from women (Adato and Roopnaraine, 2004). The fact that this has not been well documented is not to suggest domestic violence or gendered conflicts do not occur due to resource transfers, but points to the sensitivity of such an issue and the repercussions for women and children. The fact that other reports claimed money was rarely spent on vegetables (Quisumbing and Maluccio, 2000) suggests the important role of food transfers in such programmes, but also the variation in household priorities relative to how cash is allocated. This is not to advocate conditionality but to recognise that households cope with poverty in varying ways.
3.2 Integrationist or Agenda-Setting? Mainstreaming Gender at the Policy and Programme Level
A gendered analysis of Nicaragua’s PRSP finds support for the contention that international development policy has followed what Jahan (1995) defines as an integrationist approach to mainstreaming gender. Understood as the means by which women have been identified as participants and beneficiaries in development, without the explicit goal of redressing the structures that perpetuate their subordination, the objectives of the PRSP have not paralleled a confident shift towards GAD discourse. The UN Economic and Social Council definition of gender mainstreaming defined making gender issues a feature of the ‘design, implementation, monitoring and evaluation of development’ programmes (1997 cited in Porter and Sweetman, 2005: 2). However the PRSP has pushed for labour intensive economic growth as a solution to reducing poverty with little reference to gender goals (Bradshaw, 2008a). This lends support to Elson’s (2004) argument that GAD policies have not re-shaped development agendas. The promotion of labour-intensive growth instead resembles WID arguments for women’s entry into the productive sphere. Poverty alleviation is addressed from a ‘perspective of greater generation of wealth’ through GDP growth and employment (IMF, 2006:19) yet the RPS is not designed as a conditional work programme. Though this could be the result of low investment in infrastructure and limited job creation.
Early literature on the Poverty Reduction Strategy (PRS) process of Nicaragua discussed the integration of civil society to contribute to its design and monitoring (Dijkstra, 2011). With this the participation of women’s organisations would permit greater attention to attaining gender justice within developing economies (Ibid.). Gideon argued poverty reduction strategies needed to ‘promote women’s voice to influence development agenda’s, maintain a level of cooperation between women’s movements and policymakers, ensure transparency and accountability and make explicit the goal of gender justice’ (2006 cited in Dijkstra, 2011: 297), however whilst in theory there has been an opportunity for women’s voice and transparency this has not translated into practice (2011: 297). This supports the notion that poverty has not been viewed in the context of gender roles and relations, supporting the view that human capital and poverty reduction do not go hand in hand.
A significant trend within the PRSP is the discussion of making women beneficiaries in health and education, but this is not synonymous with action on the ground level (Dijkstra, 2011: 293). RPS has promoted two of four pillars of the IMF’s ‘strengthened’ PRSP; the investment in human capital, and social protection for the most vulnerable (IMF, 2006). Under this women have been portrayed as a vulnerable group in need of social welfare alongside that of children and the elderly. This is supported where Dijkstra asserts women have not been identified as ‘economic agents’ at any point within the paper (2011: 293). Whilst the targeting of cash to women within RPS is not confirmation of a gender approach, it purports to failure to embed gender justice within the PRS process. The idea that women have been made decision makers merely by fitting them into CCT programmes is corroborated by the application of Molyneux and Thomson’s 2011 framework to RPS. The criteria laid out by the authors purports to an allocation of resources that directly seeks to strengthen women’s capabilities, the goal of social and economic empowerment, family-friendly policies, shared responsibility amongst men and increasing women’s voice in regards to programme design, implementation and evaluation (2011). The targeting of cash to rural women in RPS has had the effect of increasing the obligations of women with a lack of attention to their existing responsibilities. Meeting the conditions for cash transfers, their attendance in community meetings and talks on how to become better mothers has had the impact of augmenting their time poverty and lowering their self-esteem as mothers (Bradshaw and Viquez, 2008). Cash allows women to meet the additional needs of their family with the benefit of meeting children’s in education, however this has been to reinforce women as what Bradshaw describes as “objects of reproduction” (2008:199). Interestingly, women beneficiaries reported significant improvements in household gender relations as a result of cash transfers, whereby the paucity of resources was a cause of conflict (Adato and Roopnaraine, 2004). This further reinforces the value of resource transfers in meeting the survival needs of poor households, rather than just their socially constructed needs.
Gender equity and female empowerment has been asserted as an indirect effect of making women cash beneficiaries as opposed to an overarching goal of CCTs. This would imply gender was not excluded in the design of RPS. However the fact that the RPS has recognised men’s stereotyped behaviours in terms of wasteful spending (Bradshaw, 2008a), their failure to address it implies an integrationist approach to gender in policy and RPS. As in Malawi and Ethiopia, male attitudes remain a significant barrier to the empowering potential of cash and food transfers. Whilst it is of credit to Ethiopia’s PSNP in recognising the need for crèche facilities the failure to implement these is further symbolic of an integrationist approach in international policy, but even less so where women in Ethiopia have not been made beneficiaries of cash transfers. The fact that RPS does not seek to encourage gender parity in education implies there is little recognition of how gender shapes poverty. Though Bradshaw and Viquez suggest this may be because primary school attendance is not unequally gendered (2008: 833). RPS identifies women according to the social construction of women’s needs in other words dictating what is best for them. The corollary of which has been to reinforce their status as “passive recipients of welfare” (Razavi and Miller, 1995: 4). This supports Goetze’s argument that WID discourse did not promote women’s strategic gender interests such as challenging social stereotypes, but what development required from women. This would support the argument that international policy has not changed with the shift from WID to GAD discourse.
In assessing whether women have the ability to be empowered based on Kabeer’s distinction between power, empowerment and disempowerment, is supported by women’s subordination under gender divisions of labour and social stereotypes. RPS has identified women solely as cash beneficiaries and not with the intention to challenge the gendered asymmetries of power that perpetuate other forms of disadvantage. The demand for public accountability and quantitative measures of progress within today’s aid industry could explain the focus on education and health targets as a definition of programme success. However this would only reinforce the need to address gender inequality from the top down. The need for effective monitoring and further research into gendered power imbalances and women’s social relations is pivotal to addressing gender equity and enhancing women’s status through CCTs. Women represent tools of development under RPS, which gives credence to the argument that policy practitioners have failed to move from an integrationist to an agenda-setting approach in mainstreaming gender. The autonomy exuded to women as cash beneficiaries has been further questioned with conditionality. If women fail to meet the conditions for which they qualify, they risk their entitlement for that period with consequences for the ability of households to meet their immediate needs. As this research has found, this augments gendered conflicts suggesting the greater importance of resource transfers than conditionality in poverty reduction, if not the “morally atrocious” nature of conditionality (Freeland, 2007).
Chapter Four
4.1 Conclusion
This research has discussed and analysed Nicaragua and Ethiopia’s social protection programmes as well as the humanitarian relief intervention that was Malawi’s Food and Cash Transfer. In seeking to assess whether food and cash transfers can better mediate the unequal gender relations and power-asymmetries within poor households, it finds cross-country comparative support for the hypothesis that food and cash transfers together have a greater capacity to reduce household gender tensions. Food and Cash together can mediate gender tensions where intra-household conflicts stem from a scarcity of resources. The case studies used show the value of resource based transfers in promoting food security in both a development and humanitarian context, as well as the ability for households to meet important non-food needs, provided women remain resource beneficiaries. The fact that men and women use resources differently is a significant finding and thus highlights the importance of a contextual understanding to gendered roles and relations at both the community and household level. The significance of this is to appreciate how resource transfers and their allocation can strengthen the position of men over women when women are not beneficiaries. As both the literature and case studies show, where men and women do not have access to their own income, this can be a cause of heightened disagreements. Similarly male disempowerment can also increase tensions in providing cash to women. This identifies the need to readdress men’s exclusion from social protection programmes to promote gender equity and mutual decision-making, otherwise food and cash aid may reinforce traditional divisions of labour where social stereotypes are not challenged.
Critically the provision of cash and food-based resources to women within CCT programmes has done little to increase women’s agency, choice and power in a context of gendered power asymmetries. Whilst women reported enhanced status and self-esteem in Nicaragua, this has been within a context of their own gendered roles. The explicit promotion of investing in the health and education of poor children has been at a corollary to the welfare of women and the recognition that poverty is shaped by gender roles and relations. The extent to which women’s reinforced division of labour and time poverty stems from a failure to address gender equity and female empowerment at a broader level is to challenge the rising profile of women under the UN since the mid-Seventies. Addressing gender justice within development programmes on the ground level has been contingent on the prioritisation of gender equity and female empowerment at the policy level. The impact of this on poverty reduction schemes at a time when the effectiveness of development aid is under intense scrutiny has severe implications for developing economies and the poor reliant on assistance. This research highlights the importance of making gender equality and female empowerment an explicit goal of social protection schemes in order to promote long-term, sustainable poverty reduction.
At a broader level, social protection programmes have displaced the role of the state in the public provision of goods solely onto women in leading to what Chant has identified as the ‘feminisation of responsibility and obligation’ (2006 cited in Bradshaw, 2008a). The long-term consequences of the inattention to gender justice at both levels to paraphrase Jahan will only serve to make women’s progress in development more elusive (1995) and perpetuate existing gender inequalities in the future.
Cash or Food Aid? Assessing the role of resource-based transfers in achieving female empowerment and gender equity in social protection programmes (70)
Tags: Conditional Cash Transfers, Economics, Empowerment, Gender, Nicaragua, South America
