Summarize/Analyze Journal Articles

World Development 113 (2019) 352–367

Contents lists available at ScienceDirect

World Development

journal homepage: www.elsevier.com/locate/worlddev

The power of lump sums: Using maternity payment schedules to reduce the gender asset gap in households reached by Brazil’s Bolsa Família conditional cash transfer

https://doi.org/10.1016/j.worlddev.2018.08.012 0305-750X/� 2018 Elsevier Ltd. All rights reserved.

E-mail address: gmorton@bard.edu 1 Other conditions vary by program, but they can include attending health

education sessions, obtaining prenatal care, or getting an ID card. For reviews of conditional cash transfers, see (Valencia Lomelí, 2008; Fizbein, Schady, & Ferreira, 2009; Lagarde, Haines, & Palmer, 2007; Ranganathan & Lagarde, 2012; Saavedra & García, 2012).

2 Bolsa Família officials preferentially sign up women for the benefit and card to them; in a household where a woman is not available, however, a sign up to receive the money (Lindert, Linder, Hobbs, & de la Brière, 2007,

Gregory Duff Morton Bard College, 30 Campus Road, Red Hook, NY 12504, USA

a r t i c l e i n f o

Article history: Accepted 26 August 2018 Available online 27 September 2018

Keywords: Bolsa Família Maternity benefits Conditional cash transfer Gender asset gap Salário Maternidade Household budgets

a b s t r a c t

Can cash assistance have an influence on gender relations inside a household? What are the processes through which this influence occurs? The present article investigates the everyday uses of money that women receive from two gender-targeted social programs in rural Brazil. Bolsa Família is a conditional cash transfer that disburses money to women every month. The Maternity Wage is a program that gives a sizeable lump sum to women when they become pregnant. Drawing from two years of ethnographic research in two villages in Northeastern Brazil, I show how these different payment schedules can lead to different patterns of investment in assets. I find that women typically spend monthly cash assistance on items, like clothing and furniture, that correspond to local stereotypes about feminine property. By contrast, lump sums are used by women to purchase income-generating assets, like cows and fields, that would normally be held by men. Monthly money reinforces gendered stereotypes about assets, while lump-sum money challenges those stereotypes. Lump sums thereby enable women to become the own- ers of wealth that generates a flow of income over time. I identify two key qualities that underlie this change: a payment’s large size and its unpredictability. These qualities affect the mental accounting that beneficiaries use to understand their money and the institutions through which they save it. By outlining such processes, the article brings the literature on conditional cash transfers into dialogue with studies on the gender asset gap. Lump sums can help to re-gender a household’s assets. This finding suggests that cash assistance policy, particularly in the case of conditional cash transfers, might be able to have an effect on gender equity by making use of targeted lump sums.

� 2018 Elsevier Ltd. All rights reserved.

1. Introduction

Conditional cash transfers have become important tools for fighting poverty in middle-income countries. Their importance derives, in part, from their promise to tackle poverty and gender inequity at the same time. Conditional cash transfer programs (CCTs) provide modest, regular cash payments, which are usually delivered to women in families with low incomes; in exchange for the money, children must attend school, receive vaccines, and comply with other human-capital ‘‘conditions.”1

In the case of Brazil’s Bosla Família, the world’s largest CCT, money arrives from the federal government, every month, on a

debit card whose secret code is chosen by the woman receiving the benefit.2 Policymakers intend for this delivery system to change gender relations inside the family (De Brauw, Gilligan, Hoddinott, & Roy, 2014, p. 487; Gil-García, 2016, p. 451; Barrientos, 2012, p. 15; Adato, De la Briere, Mindek, & Quisumbing, 2000, p. 46). CCTs thus raise important questions about what happens to cash once it reaches a household. Who holds onto the money, who gives orders about it, and who spends it? Do households use some of this money to buy assets? If so, who owns them?

An outpouring of ethnographic research, in Brazil and beyond, has begun to answer these questions by documenting the mecha- nisms through which CCT cash circulates inside households (Streuli, 2012a,b; Adato et al., 2000; Pires, 2014; Rego & Pinzani, 2013; Suárez & Libardoni, 2007; Pires and da Silva Jardim, 2014).

issue the man can p. 17).

4 Among survey respondents in rural areas, De Brauw et al. (2014) find a negative relationship between receipt of Bolsa Família and women’s reports that they are the exclusive decision-makers in household decisions about food, employment, and children’s school attendance. Ethnographic evidence from Rio Branco and Maracujá (the villages analyzed in the present article) suggests one possible explanation from

G.D. Morton / World Development 113 (2019) 352–367 353

The present article contributes by focusing on payment schedule: how often the money arrives.3 Does it matter that women receive this money every month? Would the money have a different effect on the household if it came, instead, as a lump sum?

This article is based on ethnography carried out over eighteen months, from 2011 to 2013, with 51 households receiving Bolsa Família in the sertão region of rural Brazil. The article relies on the contrast between Bolsa Família and a different social program, the Rural Maternity Wage (Salário Maternidade rural). While Bolsa Família enters the household each month, the Maternity Wage is a lump sum that women can obtain in a single payment when they become pregnant.

As it considers the uses of money from these different sources, the article concentrates on one aspect of household relations: the ownership of assets. Ethnography is used to document which assets – like cows, fields, and appliances – are considered to belong to a man or a woman. The article investigates the impact that social assistance programs can have on the gendering of these assets. In so doing, the article brings CCT scholarship into dialogue with the literature on the gender asset gap (Deere & Doss, 2006; Doss, Deere, Oduro, & Swaminathan, 2014).

I find that women buy assets with some portion of the money they receive from both Bolsa Família and the Maternity Wage, but they use the money from each program differently. Bolsa Famí- lia’s monthly money often goes towards items that correspond to local stereotypes about feminine property – like appliances and furniture. By contrast, the lump sum from the Maternity Wage allows women to buy assets that, in these villages, normally belong to men – income-generating assets like cows and fields. In other words, monthly money reinforces gendered stereotypes about assets, while lump-sum money challenges those stereotypes.

Why do women spend money from the two programs differ- ently? Anthropological investigation demonstrates that the one- time lump sum is profoundly disruptive to a household’s normal budgeting practices. The lump-sum schedule influences people’s habits of mental accounting and the institutions available to help them save. Because of its disruptiveness, I suggest, a lump sum cre- ates an opening for household members to reorient the gendering of assets.

The article identifies two reasons for the disruptiveness: pay- ment size and payment unpredictability. First, the Maternity Wage delivers a sum whose size is often ten times larger than the per capita monthly income of a household at the research site, and because the payment is so large, households cannot save this money through their habitual savings practices. Instead, household members develop new practices of mental accounting. They make use of unaccustomed modes of saving—modes such as having women own masculine assets.

Secondly, the Maternity Wage is granted with great unpre- dictability. More than half of women who apply for the benefit are denied. Because of this unpredictability, existing institutions do not allow a household to take out credit in anticipation of the lump sum. If the benefit does arrive, the household must manage it as an unanticipated positive shock. Women tend to respond to this shock by acquiring assets, such as cows and fields, that serve as an ongoing source of income. Through the new assets, the household smooths out the sudden increase in cash by spreading it over time. These income-generating assets are the property of

3 There have been several scholarly appeals calling for greater attention to timing in the analysis of CCTs. Fizbein, Schady and Ferreira noted, ‘‘Timing of payments is another potentially important design feature. To date, however, few programs have experimented in that direction” (2009, p. 133). Rabinovich and Diepeveen echoed this theme six years later: ‘‘The impact of dimensions such as the timing and amount of transfers and delivery agents and mechanisms remains largely unexamined” (2015, p. 638).

a woman, which changes intra-household distribution over the long run.

Lump sums, then, have a special, disruptive capacity to change the gender asset gap inside the household. This capacity means that they are potentially powerful tools for social policy.

This article speaks to the extensive literature that documents the gender effects of conditional cash transfers. CCTs are, at heart, gendered policy. The programs were originally motivated by theo- retical (Chiappori, 1992; McElroy & Horney, 1981; Lundberg & Pollak, 1993; Agarwal, 1997) and empirical (Quisumbing & Maluccio, 2000) research challenging the ‘‘unitary” view that all members of a household share a single set of preferences. CCT pio- neers found inspiration in studies suggesting that women are more likely than men to spend additional income on children (Thomas, 1990; Hoddinott & Haddad, 1995), although subsequent results have not always been as tidy (Handa, Peterman, Davis, & Stampini, 2009; Yablonski and Peterman, 2016; for a review, see Yoong, Rabinovich, & Diepeveen, 2012).

More recent empirical work has gone beyond the question of women’s expenditures on children, examining a range of gendered CCT outcomes (van den Bold, Quisumbing, & Gillespie, 2013; Chap- ter 3 in Holmes & Jones, 2013). In large-scale surveys, Mexico’s Progresa/Oportunidades/Prospera is associated with lower levels of violence against women (Rivera, Hernández, & Castro, 2006; Angelucci, 2008) and an increase in women’s reports that they decide how to spend their own money (Attanasio & Lechene, 2002; Adato et al., 2000). A similar Brazilian survey demonstrates that Bolsa Família has a positive impact on women’s reports that they are the sole decision-makers in several areas of household life, including contraception and children’s school attendance, but these results hold only among urban households (De Brauw et al., 2014; also see Soares & Silva, 2010).4 Qualitative researchers find evidence linking CCTs to female control over family expenses (Streuli, 2012a; Morton, 2013; Adato et al., 2000; Adato & Roopnaraine, 2010), men’s endorsement of women’s spending power (Maldonado, Nájera, & Segovia, 2006), and women’s expressed sense of autonomy (Rego & Pinzani, 2013; Suárez & Libardoni, 2007). The qualitative research also documents extensive debates inside house- holds over which household member can rightfully claim a share of the CCT money (Pereira & Ribeiro, 2013; Corboz, 2013), and researchers describe women’s efforts to divide CCT cash among chil- dren and spouses (Morton, 2015b, p. 1298; Pires, 2013; Ahlert, 2013; for an overview, see Villatoro, 2005). In a critical vein, scholars note that CCTs may reinforce stereotypes about women’s obligation to do child-rearing labor (Gil-García, 2016; Mariana & Carloto, 2009). CCTs may also impose time demands on women, increase women’s work (Molyneux, 2006; Molyneux, 2009) without asking for ‘‘greater involvement by men” (Gomes, 2011, p. 78), and fail to create space for women’s public involvement (Gomes, 2011, p. 77; Suárez & Libardoni, 2007, p. 126) or women’s participation in wage labor (in contrast to crèche programs; see Lavinas & Nicoll, 2006).

this puzzling result. Among small farmers at Rio Branco and Maracujá, ‘‘shame” (vergonha) is a highly prized virtue, a positive quality held to be characteristic of the countryside. People who exhibit ‘‘shame” are modest and collaborate with others. They are unlikely to report in a survey that they make decisions fully by themselves. However, not all small farmers successfully master the display of shame. Those who do not demonstrate appropriate shame may have difficulty securing the approval of welfare program administrators and complying with the bureaucratic requirements necessary to obtain Bolsa Família and maintain the benefit over time. In rural areas, then, the refusal to report independent decision-making may be correlated with ‘‘shame” and hence with an increased likelihood of receiving Bolsa Família.

354 G.D. Morton / World Development 113 (2019) 352–367

The gender lens has not been yet been focused on CCTs and women’s asset ownership.5 Although evidence suggests that CCTs can increase household ownership of productive assets (Gertler, Martinez, & Rubio-Codina, 2012)6, little research has been conducted on how CCTs affect the gendering of these assets inside the house- hold. Beyond the context of CCTs, however, a substantial literature considers the gender gap in assets and wealth. Scholars have demon- strated female disadvantage in the ownership of varied assets (Doss et al., 2014; Deere, 2010; Moser, 2010a; Antonopoulos and Floro, 2005), particularly land (Agarwal, 1994; Deere and León, 2003). Asset ownership has been shown to have profound implications for women’s health, longevity, and resilience in crisis (Deere and Doss, 2006). Women’s asset ownership at the time of marriage, in some contexts, is associated with variations in the amount that the household spends on food, education, alcohol, and tobacco (Quisumbing & Maluccio, 2003) and with women’s expectations about household power (Fafchamps & Quisumbing, 2002; Anderson & Eswaran, 2009). A policy granting assets to women rather than men can lead to changes in household consumption and spouses’ time allocation (Wang, 2014). The gender asset gap, however, does not look the same in every context; its size varies across nations and across income groups inside nations (Doss et al., 2014), depending on labor market conditions, legal frame- works, and social norms. Hence the literature includes repeated calls for attention to the institutions and practices that determine owner- ship at the local level (Deere, 2010; Deere & Doss, 2006; Moser, 2010b).

Research in this area often intersects with scholarship on the ‘‘asset approach to poverty:” the effort to redefine poverty not as a lack of income, but as a lack of assets (Sherraden, 1988; Sherraden, 1991; Sherraden, 2005; Bailey, 2010; Ssewamala, Sperber, Zimmerman, & Karimli, 2010; Cramer & Shanks, 2014; Moser, 2008). Researchers in this area emphasize that assets have a transformative effect on long-term life plans in contexts of impoverishment (Moser, 2010a) and conclude that ‘‘the accumula- tion of assets might ultimately be more important for household wellbeing than pure income measures” (Moser, 2010b, p. 394). In its focus on the difficult conversion between short-term income and long-term assets, this literature overlaps with anthropological insights about money. Anthropologists have often observed the creation of cultural systems that distinguish between the realm of transient gain and the realm of enduring social reproduction (Parry & Bloch, 1989, Introduction; Bohannan & Bohannan, 1968), the latter sometimes being associated with constructed signs of femininity (Taussig, 1980; Weiner, 1976).

The present article investigates households by considering con- ditional cash, asset ownership, and gender dynamics in two vil- lages in northeastern Brazil. The results are ethnographically specific to a single time and place, but they point toward more gen- eral processes. The article argues that positive budget disruptions can help households overcome the gender asset gap, and the article

5 There are some exceptions. For data that demonstrate that women use Progresa CCT payments to purchase small livestock, see Rubalcava, Teruel, & Thomas, 2009. For evidence that women are more likely than men to invest Progresa CCT payments in business and agriculture ventures, see Davis, Handa, Stampini, & Winters, 2002. Neither of these articles, however, provides new data to answer the question of which family member, inside the household, is considered to be the owner of a given asset. For an analysis of a Zambian unconditional cash transfer associated with a sizable increase in women’s asset acquisition, see Natali, Handa, Peterman, Seidenfeld, & Tembo, 2016.

6 For contrary evidence, see Maluccio, 2010; for a review, see Kabeer, Piza, & Taylor, 2012, p. 24.

notes the usefulness of lump sums as a tool for creating such dis- ruption. After the introduction, Section 2 describes methods. Sec- tion 3 details the ownership of assets by gender at the field site, and Sections 4 and 5 consider the effects of Bolsa Família and the Maternity Wage on ownership. Section 6 examines the gifts that women give with Maternity Wage money. Section 7 discusses the intra-household processes that underlie the effects observed, and a final section concludes.

2. Methods

This article is drawn from a fieldwork project designed to explore the expansion of Brazil’s rural welfare state under the Workers’ Party. Fieldwork was conducted between 2005 and 2016 in the rural area of Vitória da Conqusita, Bahia, Brazil. Core activities took place in 2011–2013. Vitória da Conquista was headed by a Workers’ Party mayor from 1997 to 2017, making the municipality an apt site for the study of 21st-century welfare policy.7

Fieldwork was primarily based in two neighboring villages, here called ‘‘Maracujá” and ‘‘Rio Branco.” The villages sit along dirt roads, about 100 km from the urban center. Both villages are com- posed of small farmers who rely on rainfall to raise livestock and cultivate dryland crops like beans, coffee, and manioc, with few opportunities for regular waged employment. Rio Branco’s 103 inhabitants,8 living in 35 households, mostly belong to an extended family that has farmed the area since the 19th century. Maracujá, by contrast, has 205 inhabitants in 62 households. Maracujá village was founded in 1996, when landless farmers occupied a plantation by organizing themselves through Brazil’s Movement of Landless Rural Workers, the MST.9 The occupation was successful: the federal land reform agency expropriated the plantation, compensated its owner, and redistributed the land. Today, at Rio Branco and Maracujá, almost all households have access to a plot of farmland.

Research was grounded in participant observation, including interviews (Epstein, 1967; Briggs, 1986; Smith, 2014; Hardesty, 2015). I resided in the villages, living mostly in two families’ house- holds. I accompanied people as they carried out their everyday activities: work in the home, work in the fields, hospitable visits to friends, parties on the front porch, and appointments at the wel- fare office.10

Along with participant observation, I carried out a standardized census with every household (96 total households)11 between late 2011 and early 2012. Modeled on the PNAD,12 the census asked about each household’s earnings and assets for 2011; it also inquired into social program utilization and posed open-ended questions

7 For details on the sweeping changes that the Workers’ Party has brought to Brazil’s welfare state since the presidential elections of 2002, see (Rego & Pinzani, 2013; Ansell, 2014).

8 Unless otherwise specified, all demographics for the villages are for October 2011. 9 For more on Brazil’s landless movement, see (Morissawa, 2001; Loera, 2010;

Wolford, 2010). 10 All participants provided informed consent. The consent process was approved as IRB protocol H07130, University of Chicago. 11 In one household, the respondent declined to participate. All other households participated. One household reported a large negative income for the year because of the purchase of a number of cattle; I exclude this household when reporting information on income in the villages. The excluded household did not receive either the Maternity Wage or Bolsa Família. For details on the exclusion, see (Morton, 2015a, p. 454). In reporting currency from the survey, I use the exchange rate of 1.86 Brazilian reais to 1 US dollar, the market rate for January 1st, 2012. 12 The Pesquisa Nacional por Amostra de Domicílios (PNAD) is an annual household survey conducted by IBGE, Brazil’s statistical service.

G.D. Morton / World Development 113 (2019) 352–367 355

about work history, migration, and visions of the future.13 The sur- vey’s asset module included a checklist of moveable assets, which was written with the advice of village leaders.14 Houses and land were not included in the survey, since they were rarely bought and sold in the villages. To assign a market value to the assets on the checklist, I spoke with knowledgeable merchants selling compa- rable items in the nearby city.

Based on the results of the census, I selected nine households to participate in ‘‘focus family interviews.” I returned to these same households each week and asked about income and expenditures for the week.15

Interviews and field notes provided the base for analysis. I made use of the recursive process characteristic of ethnography, review- ing results and refining hypotheses, then having subsequent con- versations in the field to check conclusions (Thorne, 2000; Smith, 2014, p. 419). This checking confirmed that I needed more infor- mation about the Maternity Wage. Thus, I carried out a new research stage in which I interviewed every woman who had ever been likely eligible for the Maternity Wage.16 After iterative rounds of checking, I presented results to local leaders, municipal officials, and social movement organizers. I also consulted with specialists on social assistance in the US and Brazil. These overlapping sources made it possible to search for alternative interpretations, thereby helping to test conclusions.

3. Ownership, assets, and gender in the villages

Small-farming families have to contend with a dry landscape at Maracujá and Rio Branco. Families reside in small houses, usually whitewashed, and in order to facilitate water delivery these homes are clustered together near the dirt soccer fields and the Protestant and Catholic chapels that form the center of each village. With rare

13 The survey was conducted by me directly. I lived in the villages and had personal relationships with village residents, which might have influenced responses. In some cases, respondents might have attempted to deceive me for personal reasons. In other cases, they may have been more honest about their asset ownership, knowing that I could potentially see assets for myself. Residence in the villages also made it easier for me to determine how long each person had been living in the countryside, which was an important element in determining eligibility for the Maternity Wage. I became familiar with the reasons behind the gap between actual residence and proof of residence, a gap that lies at the root of many problems in obtaining the Maternity Wage. 14 The module also asked respondents to identify any other objects of value that they owned. On the standard checklist, respondents were asked about the number of the following items that they owned: cows, pigs, chickens, bee hives, guinea fowl, horses, donkeys, mules, sheep, ducks, goats, turkey, other animals, tables, chairs, stoves (wood or gas), computers, refrigerators, horse-drawn carts, cars, pots and pans, bicycles, motorcycles, beds, televisions, radios, CD players, DVD players, telephones (land line or cell), phone antennas, water filters, water tanks, parabolic TV antennas, freezers, clothes washing machines, sofas, kitchen shelves, clothes wardrobes, living- room shelves, television shelves, chests of drawers, fans, sewing machines, electric shower heads, clothes irons, rugs, video games, cameras, water pumps, and guitars. The checklist did not include clothes, jewelry, or shoes, because of the difficulty of enumerating separate small items and also because informants reported that these items were not significant stores of wealth in the local area. 15 The focus family interviews continued for a period ranging from two to six months, depending on the household. For details on the methods used in all of these surveys and the survey results, see Appendix 1 from (Morton, 2015a). 16 The new stage identified 24 women who had applied for the Maternity Wage, of whom 12 received the benefit. This is a small group on which to base an analysis. Fortunately, however, the Maternity Wage was a major topic of conversation among many people living in the villages at the field site, so I was able to compare the opinions and practices of actual Maternity Wage recipients with the viewpoints expressed by a larger number of village residents. In total, I interviewed 49 women who were likely eligible for the benefit (see Table 2), and I had informal conversations about the Maternity Wage with a much more diverse group of women and men. I found the views of non-recipients to be highly consistent with the practices of recipients. There was widespread agreement that, if a woman received the Maternity Wage, she should spend it on a productive agricultural asset. This consistency provided some reassurance that the behavior observed among the small group of recipients was behavior that corresponded to a widespread norm at the field site.

exceptions, no irrigation is available. Farmers count on the region’s twice-annual rains to grow coffee, manioc, and pineapples for mar- ket sale or home consumption, with beans, corn, dryland sugar cane and garden vegetables grown for home consumption only. Farmers also raise livestock, particularly cattle, pigs, and chickens. Typical family farms range in size from ten to twenty hectares.

Given the arid climate and the distance from an urban market, most families cannot subsist on farming alone. They combine income from a variety of sources, including retirement pensions, employment in rural schools and health clinics, day labor on nearby plantations, and cyclical migration to cities or more distant plantations. In the 2011 survey conducted as part of this research, median annual household income (excluding Bolsa Família and Materntiy Wage) for the two villages was R$2732 (US$1469) per capita.

People at Maracuá and Rio Branco tend to live in households anchored by a male-female couple, often with their children, grandchildren, in-laws, siblings, or friends residing in the home as well.17 Each adult will usually contribute to the household’s sus- tenance through several forms of work.18 In the dominant local model, women specialize in housework, child care, and the home production of food for sale, while men devote themselves to work in fields. But gender stereotypes about work are not unbreakable. Women toil in rows of plantation coffee, team up with friends to plant their own bean fields, and travel to the city so they can labor in factories. Men watch children and teach classes at school.

Because of the arid climate, farmers cannot count on crops every year. It becomes especially important to hold assets, partic- ularly livestock, that can be sold in a time of need. Beyond animals, households own a range of other moveable assets, including furni- ture, appliances, and motorcycles. In 2011–2, the average house- hold in Maracujá owned moveable assets (including livestock) valued at 2.69 years’ worth of the household’s annual income; in Rio Branco, 1.55 years’ worth of annual income. In each village, livestock accounted for nearly half of the value of these assets. (Insert Figs. 1 and 2 here.)

Livestock, however, do not usually belong to a household. They belong to a particular person inside the household. In everyday social interaction between villagers, intra-household ownership becomes perhaps most salient through the practice of gift-giving. It is common for adults to give livestock as a gift to children. The act of gifting requires an adult to declare that he or she owns an animal and then transfer ownership publicly to the child. These declarations become the topic of neighborly conversation, with farmers spreading the news of a gift. Thus, which animal belongs to which household member is common knowledge. Neighbors remember that the black spotted cow belongs to the oldest daugh- ter in the family next door, or that the duck wandering through the backyard is the property of the younger son.

Beyond the case of livestock gifts, however, it is frequently con- sidered contrary to the ethos of cooperation for a person to declare that certain objects in the household belong to herself or himself individually. Respondents explained to me that naming individual owners of assets is a sign of family discord. This leads to difficulties in interpreting survey responses. Interviewees often expressed ambivalence about which person inside the household was the owner of which assets. Some respondents identified more than

17 More than half of households have children living in them, and more than 75% of these households with children have a male-female couple living in them (not always the child’s parents). For details, see (Morton, 2013) footnote 4 and (Morton, 2015a), Appendix 1. 18 Child labor is now rare in the villages, beyond chores at home and occasional help in the family fields. However, villagers report that child labor was common on plantations and in small fields as recently as ten years ago. Villagers say that child labor has stopped because of rigorous government enforcement of laws, along with the recent appearance of schools and social programs in the countryside.

Fig. 1. Household income and moveable assets, per capita, Rio Branco.

Fig. 2. Household income and moveable assets, per capita, Maracujá.

356 G.D. Morton / World Development 113 (2019) 352–367

one owner for an object—as in the case of a male farmer who told me that the cattle belonged to him, although he considered that they also belonged to his wife.

But while respondents express reticence about claiming objects for themselves, there exist patterns in the responses. These pat- terns emerge when one considers the person whom a respondent

first mentions in connection with an object. Men are overwhelm- ingly, but not exclusively, described first as the owners of cattle. Villagers speak of senior men as the owners of the houses, since it is believed to be the man’s duty to build a house. Although it is not habitual to describe any individual as the owner of crops in the field, men are closely associated with cash crops, because

Table 1 Access to benefits, Maracujá and Rio Branco, 2012.

Number of households. . . Maracujá 62 households, 205 people

Rio Branco 35 households, 103 people

Total for both villages

Receiving BF 31 20 51 Not receiving, but likely

eligible 8 2 10

% of likely-eligible households that receive BF

79.5 90.9 83.6

Have received Maternity Wage

6 6 12

Did not receive, but likely eligible

25 12 37

% of likely-eligible households that receive Maternity Wage

19.4 33.3 24.5

Data from census conducted by author in two villages, 2011–2012.

G.D. Morton / World Development 113 (2019) 352–367 357

men often work in the fields, sell the crops, and pocket the money.19 On the other hand, in interviews, senior women tend to be mentioned first as owners of domestic objects—stoves, beds, linens, and plates, some of which may have come in the bride’s trousseau. This distinction between male objects and female objects corresponds to the (loosely-enforced) local model in which men in the villages work outside the house and women work inside of it. As two women explained it to me in conversation one day, the man owns the house, but the woman owns everything inside it.

In the household, then, both men and women possess assets that hold value over the long term. These assets hold value for dif- ferent reasons, however. Women’s objects hold value because they are durable, and these objects become more influential as they become more durable. A better stove will last longer and hence extend the influence of the woman who owns it. By contrast, men’s objects, in at least some cases, hold value because they are repro- ducible. A cow gives birth to more cows, stretching value forward across generations. Crops are harvested and the seed is sown again. As men’s assets reproduce, they also produce income. Some calves can be slaughtered; some of the harvest can be sold. While women’s objects slowly lose value over time, men’s objects may hold value steady or, through the reinvestment of income in the field or herd, even increase in value (see Weiner, 1976, p. 236).

This is the pattern of asset gendering that Bolsa Família tends to reinforce in the villages. The Maternity Wage, however, can disrupt the pattern. The following sections describe the two programs’ impact on asset ownership inside households.

4. Bolsa Família and asset purchases

In 2011–12, Bolsa Família was widespread in the villages; 84% of likely-eligible households were receiving the benefit.20 In the 51 receiving households, Bolsa Família payments ranged from R$ 38 (US$ 20) to R$ 226 (US$ 122) per household per month, with a mean payment of R$ 117 (US$ 63) (SD = 42.7, median = 102). (Insert Table 1 here.)

Bolsa Família benefits are delivered on a monthly basis, and the money is often spent on fast-cycling items like food and school supplies. However, women report in interviews that they also strive to set aside at least a portion of the Bolsa Família cash each month so they can turn it into a durable asset.

They can achieve this thanks to roving peddlers called mas- cates.21 The peddlers play a major role in the use of Bolsa Família in the countryside. Mascates ply the back roads in heavily-laden cars and small trucks, passing through each village once a month. They sell furniture, appliances, and other household items. Villagers report that Bolsa Família has allowed mascates to extend their reach into rural areas. Aware that his22 customers now have a source of monthly income, a mascate provides credit on a personal basis. He delivers a stove, couch, or similar item to a family as soon as the fam- ily makes the first payment, then he returns each month to collect installments until the debt is paid. Although the poorest women have difficulty devoting any Bolsa Família money to assets (Morton, 2013), mascates facilitate purchase by offering flexible and renegotiable terms of credit. In exchange for credit and conve- nient transportation, mascates charge prices much higher than in

19 Women do sometimes plant crops by themselves or in conjunction with other women, but these are typically subsistence rather than cash crops. 20 I determined likely eligibility for Bolsa Família and the Maternity Wage by taking the information that each household reported to me on the census survey and comparing that information to the government’s eligibility requirements for the programs. When necessary, I also considered further information from the household, such as length of residence in the countryside. 21 The name mascate comes from the city of Muscat, in Oman, which since ancient times has served as a commercial emporium. 22 At Maracujá and Rio Branco, mascates are overwhelmingly men.

the city—typically double, according to a merchant I interviewed in 2012. Despite prices, the mascate system has led to a rural expansion in durable goods, from metal pots and blenders to couches and televisions.

Women plan these purchases well in advance. They often speak about the objects they aim to acquire with Bolsa Família: perhaps a bed for a child, then, once it is paid off, a couch, then a television. Dona Marlene recalled the history of the items she had bought, one after the other, in a chain of installment payments stretching over years.

23

‘‘a no m ac m

Right now just recently I bought this stove with the Bolsa Família money. [. . .] I bought a sieve [. . .] I bought the ceramic tiles for this house. I made a monthly credit agreement for fifty reais. So I would pay the fifty reais for the credit payment on the tiles. Then, after I finished the tiles of this kitchen – [. . .] Then I bought – that kitchen cabinet there, look at it.

From an anthropological perspective, ccounting models” rather than ‘‘mental t, in fact, the minds of my interlocutor odels) of their action and the reasons beh counting” is well established in the lite ental accounting, see Thaler, 1990 and K

Agora mesmo esses dias eu comprei esse fogão com o dinheiro da Bolsa Família. [. . .] Eu comprei peneira. [. . .] Eu comprei a cerâmica dessa casa. Aí eu fiz uma prestação de cinquenta reais. Aí eu pagava os cinquenta reais da prestação da cerâmica. Aí, depois que eu terminei a cerâmica dessa cozinha—[. . .] Aí eu comprei-- esse armário aí, Ó.

In the villages, some Bosla Família beneficiaries describe a charac- teristic ‘‘mental accounting” practice that they follow in order to budget the benefit money.23 First they decide on an amount they can devote to assets each month. Then they commit to credit with a mascate (or sometimes a store) for this amount. Finally, after the monthly asset money has been spent, the remaining money is avail- able to buy transitory items, like food.

Martina used this practice to spend fifty reais per month on household assets. Martina received a little more than R$100 in Bolsa Família each month. She recounted the reasoning that she used to allocate this money.

it might be preferable to speak about accounting,” since what I observe here is s, but rather their own descriptions (or ind it. However, because the term ‘‘mental rature, I follow common usage here. On och & Nafziger, 2016.

2

Fa fro ot Bu els th in sp du fie th ho co m ad fo ab re sit to ex an – hu

358 G.D. Morton / World Developm

Bolsa Família isn’t enough for you to buy things, only if you put together money from several months, you know? [. . .] I myself, I always buy something like that, something that costs around fifty reais. Because then you’ve got—I’m going to make the monthly credit payment, and I’ll have another fifty left over. So I always buy like that, on credit, and I pay and pay. Every month that I get Bolsa Família, I go right there and make the credit payment.

4 It is especially striking that women ma mília on durables rather than food, sinc m cash transfer programs in other nation her nations, see Adato et al., 2000, p. xi dgeting practices in northeastern Braz ewhere because of specific expectation at men should take on the responsibility o the poorest households at Maracujá and end all of the benefit money on food, rables; see a longer discussion in (Morto ld site mentioned the following worry, i at men might contribute less money to usehold received Bolsa Família or the Ma uld conceivably ‘‘crowd out” men’s spend embers of the household. This concern m amant views about the importance of no od, which was the man’s paradigmatic r out the changes in men’s spending patte ceive benefits. However, it is worth notin e and elsewhere, suggests that Bolsa Fam devote increased resources to self-empl ample, in one household that I accompan d a husband both told me with great enth how Bolsa Família covered some basic sband to devote time to improving the f

A Bolsa Família não dá para você comprar, só se você for juntando várias meses, né? [. . .] Eu mesma, eu sempre compro uma coisa assim, no valor de cinquenta reais. Que aí vai-- eu vou pagar a prestação, vai me sobrar mais de cinquenta. Então, sempre eu compro assim, a prestação, e vou pagando. Todo mês que eu recebo Bolsa Família, eu já vou lá e pago a prestação.

As beneficiaries determine the amount of Bolsa Família they will devote to household assets, conflict can emerge between men and women. Providing food for the household is widely described as a key masculine duty; once a man has managed to ‘‘put food inside the house” (botar comida dentro de casa), a woman can prepare it for consumption (Morton, 2013). Men are accustomed to worrying about food. When a woman spends Bolsa Família on food, she con- tributes toward the main masculine task. On the other hand, when she buys assets, she obligates men to find another way to feed the household. This food-versus-assets conflict – rather than disputes over men’s purported luxury spending – is the most visible Bolsa Família argument inside households at the field site.24 A woman interviewed by Suárez and Libardoni, elsewhere in Brazil, described the conflict:

My husband doesn’t give anything for inside the house. According to him, the Bolsa money is just to buy food, but I don’t think so. I buy other things. I invest in the house. I decide. (2007, p. 145).

Meu marido não dá nada para dentro de casa. Para ele o dinheiro do Bolsa é só para comprar alimentos, mas eu não acho. Compro outras coisas. Invisto na casa. Eu decido. (2007, p. 145).

ke an effort to prioritize spending Bolsa e this stands in contrast to observations s (for examples of such observations from ii and van den Bold et al., 2013, p. 14). il may differ from budgeting practices s, inside northeastern Brazilian families, f providing food. (It should be noted that Rio Branco, it is sometimes necessary to with none left over for the purchase of n, 2013).) Although no respondent at the t is possible that women were concerned wards shared household expenses if the ternity Wage. In other words, benefit cash ing on goods that have importance other ight explain why women often expressed t spending the entire benefit amount on esponsibility. I lack detailed information rns that actually occur once households g that ethnographic evidence, at the field ília can encourage both men and women oyment ventures (Morton, in press). For ied every week, at Maracujá village, a wife usiasm – and separately from each other household expenses and freed up the

amily’s shared field.

At Maracujá, Francisca spoke similarly.

ent 113 (2019) 352–367

2

to Th ve alt of fo sh ca of

Men don’t know how to manage money, like this one here [. . .] He wants to go shopping right away, buy beans and rice [. . .] The woman thinks about buying some sandals.

5 At Maracujá and Rio Branco, it is rare for buy seeds, livestock, tools, or other agricu eir farms are too small, too unprofitable ndors to extend them credit. Government hough often accompanied by complicated using credit, farmers tend to apply the pr r the next year. They also engage in a varie arecropping (in various forms) and gado ttle to another farmer to be raised, with fspring.

O homem não sabe administrar o dinheiro, tipo esse [. . .] Ele quer logo fazer compras, comprar feijão e arroz [. . .] A mulher já pensa de comprar umas sandálias.

Another time, Nalta, at Rio Branco, referred to this tension.

Ronaldo [her husband] always tells me that. ‘‘How does this happen? You’ve got Bolsa Família—they, they say it’s for food. You keep putting it towards installment payments.”

Ronaldo [esposo] sempre me fala isso. ‘‘Como é que você tem o Bolsa Família, já, já fala que é para alimentar. Você fica colocando em prestação.”

These debates between spouses point towards a larger dilemma. At Maracujá and Rio Branco, where small-farming families earn mod- est incomes at best, household members face needs that make it difficult to save. Bolsa Família arrives every month, and thus it frames the saving challenge in a particular way. It turns the prob- lem of saving into the problem of how to put money together. Bene- ficiaries must figure out how to convert regular payments into a usefully large sum (Collins, Morduch, Rutherford, & Ruthven, 2009, Ch. 4) with which to buy an asset.

Mascates offer a solution to this problem. Through instant delivery of a consumer good and follow-up installment payments, mascates allow women to pre-commit to a saving plan. This com- mitment is manifested in a characteristic mental accounting prac- tice, with women first setting aside a portion of their Bolsa Família for the asset and then spending the rest on food and other transi- tory objects. Women use such practices to plan asset purchases years in advance.

It is important to note that the mascate system has a gendered orientation. Mascates do not sell livestock or seeds for crops;25

they offer items that conform to expectations about what women should own. Thus, saving practices associated with Bolsa Família are savings practices that correspond, as Francisca puts it, to ‘‘how a woman thinks,” or, in other words, to pre-existing gender roles. Bolsa Família helps women at Maracujá and Rio Branco to plan for and acquire the durable assets that family members, neighbors, and merchants all expect women to strive toward.

5. The Maternity Wage and disruptive assets

Since 1923, Brazil’s legal system has included provisions allow- ing certain formal-sector workers to receive maternity payments (Neiverth and Mandalozzo, 2009; for overviews of maternity ben- efits in developing countries, see Fallon, Mazar, & Swiss, 2017;

any of the small farmers, men or women, ltural products on credit from merchants. , and too risk-prone for most commercial credit programs are sometimes available, and time-consuming paperwork. Instead

oceeds from one harvest to improvements ty of profit-sharing arrangements, such as de sociedade, in which one farmer ‘‘lends” the second farmer keeping some of the

G.D. Morton / World Development 113 (2019) 352–367 359

Nandiet al., 2016). The aim is to enable women to remain out of work in the months following pregnancy. In the 1988 Constitution, rural workers were granted this right (Article 7, Section 18). Because small family farmers are self-employed, Law 8861 of March 25, 1994 established that the federal government would take the employer’s place and provide these farmers with the Maternity Wage (Salário Maternidade), four months of minimum wage delivered in a lump sum.

In order to receive this money, applicants must prove they are small family farmers. A local administrator explained to me in 2012 that during the program’s first years, federal officials accepted many forms of proof but that, faced with overwhelming demand, officials subsequently became more demanding. A rural advocate showed me a list of 21 documents that a woman might be asked to provide for the Maternity Wage application, including vaccination records and receipts demonstrating the purchase of farm tools or seeds at least nine months before the baby’s birth. The process becomes especially onerous for women living far from the city. At Maracujá and Rio Branco, where mail services are nonexistent and phones unreliable, applicants sometimes spend days traveling to appointments in the city to follow their applica- tions. Fig. 3 shows the situation at a national level: after the initial expansion of the program, access was restricted. Locally, in 2011– 12 at Maracujá and Rio Branco, 24.5% of the women who were likely eligible for the Maternity Wage actually received it (See Table 2). In general, residents of the villages say that they consider the program to be ‘‘like a lottery” (no sorteio). (Insert Fig. 3 here.)26

The Maternity Wage resembles a lottery not only because its outcome seems uncertain, but also because it might deliver a large sum of money all at once. In 2011, the standard Maternity Wage value was R$2180 (US$1172), equal to nearly ten months of the vil- lages’ per capita median income.27 The 2011–12 survey measured the value of assets owned by both (a) households that had received the Maternity Wage and (b) households that had not received the Maternity Wage, but were likely eligible. As Table 2 shows, Mater- nity Wage beneficiaries had, on average, higher per-capita incomes and more assets. The mean value of livestock owned was 66% higher for Maternity Wage beneficiaries than for non-beneficiaries who were likely eligible. The mean value of household goods was 25% higher. This difference may indicate that the Maternity Wage had an effect on asset accumulation, especially livestock. However, the data are suggestive rather than conclusive: most differences are not significant in a two-tailed t-test.28 More importantly, since the data are cross-sectional, the direction of causation is unclear. It might be the case that owning more assets makes a household more likely to obtain the Maternity Wage, rather than the reverse.29

26 Data for Fig. 5 from Ministério da Previdência Social (n.d.a) and Ministério da Previdência Social (n.d.b). ‘‘INSS” is the Instituto Nacional de Seguro Social, Brazil’s social security system. Most formal-sector employees can request the Maternity Wage from their employer, but small family farmers receive it through INSS. 27 The Maternity Wage value is set at four months’ worth of the government- mandated minimum wage. Because most farmers at Maracujá and Rio Branco earn less than the minimum wage per capita, the Maternity Wage is typically worth more than four months of income for them. 28 Furthermore, the results are driven by households at Maracujá; among house- holds at Rio Branco, the difference between assets owned by Maternity Wage recipients and non-recipients is negligible. 29 Since households with more assets tend to be more influential in the villages and have better connections with bureaucratic systems, it seems likely that these households would be more likely to apply for the Maternity Wage. However, households that made failed applications for the Maternity Wage turn out to own assets similar to the assets owned by households that did not apply (to see this, compare columns B and C). This offers some reassurance that the difference between applicants and non-applicants is not the only factor driving the increase in asset values associated with the Maternity Wage. Nonetheless, the causation problem is not solved: households that start with more assets might still be more likely to succeed in their applications for the Maternity Wage.

Although the research approach employed here does not iden- tify causation, it does provide insight into the process through which the Maternity Wage is used inside households. When I inter- viewed women who received the Maternity Wage, they often told me that their first step was to spend money on short-term expenses like food. Kara, for example, described her response to the Maternity Wage as follows.

3

re fam m 31

re ad an re in (h fam 32

M eq sm 33

be

Man, when I got the money, I was really in need! So I bought some things, I bought some clothes for her [Kara’s baby daughter] – she was in need, too. [. . .] I did some of that food shopping.

0 Lui and Molina (2013, pp. 144–6) offer sidents in a rural Amazonian community ilies become pregnant in order to obt

oney in house construction and other ven One of these women was employed by

ceived the Maternity Wage benefit throu ministered differently from the Maternity d her household were not included in Ta ported spending the Maternity Wage on come, the money was used as follows: (ho ousehold 2) clothes, diapers, and blanke ily in the rural workers union, and (hou During a return visit in 2016, I learn

aracujá had obtained the Maternity Wag uipment so that she and her husband c all business selling snacks. In two cases, it was a male responden

en purchased with the money.

Moço, quando eu recebi o dinheiro, eu estava bem precisando! Aí eu comprei umas coisas, comprei umas roupas para ela—também estava precisando. [. . .] Fiz umas feiras aí.

But food shopping did not exhaust the Maternity Wage money. With the rest of the cash, Kara turned towards assets: ‘‘I bought the little pig for her [the baby daughter]. And—the money ran out.”

The Maternity Wage creates a disruptive positive shock to household budgets. Women, faced with the imperative of handling this lump sum, turn to a familiar strategy. They do what male farmers do with the earnings from an abundant cash crop harvest. They acquire assets, like livestock and new crops, that will gener- ate new income in the future.30 Of the eleven women in the villages who described to me the purchases they made with the Maternity Wage, eight bought a source of future income.31 Six of these eight women spent the money on livestock (including cattle, pigs, and donkeys), while two paid for help preparing a field with crops.32 I also conducted interviews with families who did not receive the Maternity Wage. In twelve of these families, a respondent described to me what would have happened to the money if the woman had obtained the benefit.33 Nine of the twelve reported that the money would have gone to purchase livestock. By contrast, in interviews about Bolsa Família, it was very uncommon to hear Bolsa Família recipients talk about using the monthly cash transfer to purchase an income-generating asset. The Maternity Wage, not Bolsa Família, is the funding source that allows women to acquire the reproducible assets that usually belong to men.

Women were explicit about the hope that these assets would turn into long-term sources of income. Francesca, who used the Maternity Wage to buy both a television and a calf, explained her second purchase in financial terms. ‘‘You’ve got to invest,” she counseled, ‘‘in an object that gives profit.” (Tem que investir em um objeto que dá lucro.) Marisol did not obtain the Maternity Wage, but if she had, she said, she would have spent the money on an income-generating asset.

a qualitative account of debates among . The residents argue over whether or not ain the Maternity Wage and invest the tures. the municipality as a health worker and gh her employer. Because this benefit is Wage given to small family farmers, she

bles 1 and 2. In the three households that something other than a source of future usehold 1) medical expenses for the child, ts for the child and membership for the sehold 3) a house. ed that after 2013, another woman at e. She used the money to buy restaurant ould convert an abandoned house into a

t who described to me what would have

0

100000

200000

300000

400000

500000

600000

1990 1995 2000 2005 2010 2015

Rural

Fig. 3. Number of rural Maternity Wage benefits granted per year through INSS, 1990–2014, Brazil.

Table 2 Assets in households eligible for the Maternity Wage. Maracujá and Rio Branco combined, 2012.

A: Households that received Maternity Wage

B: Households that did not receive Maternity Wage but were likely eligible

C: Households that did not receive Maternity Wage but were likely eligible and applied*

Number of households 12 37 12 Total moveable assets owned by household, mean per-

capita value including livestock, household goods, and other goods (in Brazilian reais; SD in parentheses)

6497 (5301)

3920 (4140)

3737 (2585)

Livestock owned by household, mean per-capita value (in Brazilian reais; SD in parentheses)

3820 (3286)

2068 (3377)

1573 (1631)

Household goods owned by household, mean per-capita value (in Brazilian reais; SD in parentheses)

1795 (568)y

1441 (700)y

1561 (727)

Annual income per capita (in Brazilian reais; SD in parentheses)

3865 (2710)

2409 (1956)

2245 (1137)

Data from census conducted by author in two villages, 2011–2012. * Column C is a subset of Column B. All households in column C are also included in Column B.

y For household goods, the difference between column A and column B is significant at the 10% level.

360 G.D. Morton / World Development 113 (2019) 352–367

Using part of it, I would have bought something for her [Marisol’s daughter] [. . .] And with the other part I would have bought a thing that would give an income [. . .] A calf [. . .] It’ll grow, produce.

Usando uma parte, comprava uma coisa para ela [. . .] e outra parte comprava uma coisa que ia dar renda [. . .] Bezerro [. . .] Vai crescer, produzir.

Nalta described her plans for the Maternity Wage similarly.

For his [her son’s] future, like that. Like that. For example, I would have bought, like, a—a female calf, something like, to—leave it there growing to yield something, you know? [. . .] To give income, like that, you know? [. . .] [Leave the calf] producing there so that afterwards it will give a good little bit of money.

Futuro dele, assim. Assim. Por exemplo, comprava, assim, um—uma novilha, assim, uma coisa assim, para-- Deixar criando lá para render, né? [. . .] Dar renda assim, né? [. . .] Rendendo ali para depois dar um dinheirinho bom.

Income-generating assets tend to be more expensive than the household durable goods that are usually stereotyped as feminine assets in the villages. An average-size cow, for example, sold for R $1440 (US$783) in 2011, while a gas stove cost R$260 (US$141). Moreover, because income-generating assets are reproducible, they may hold value for longer. Lara drew a contrast between income- generating assets, like cattle, and a typical asset that women pur- chased with Bolsa Família: furniture.

L: If [the Maternity Wage] were a really good little bit of money, enough to buy something, I’d spend it on cows, like that [. . .] Because you see how it is. [You see] that sometimes people go running [to buy items], you know. They buy furniture, for inside the house. A sofa. Duff: Mm hm. L: And you see that that’s a useful thing. But in a short time, you can see that that [furniture] is there all

L: Se fosse um dinheirinho bom mesmo, que dasse para comprar, é aplicar em vacas, assim [. . .] Que você vê. Que às vezes eles correm, no caso. Eles compram moveis, para dentro de casa. Um sofá. D: Mm hm. L: Mas você vê que é uma coisa que serve. Mas com pouco tempo, você está vendo que está ali estragado. Agora, se Deus dá um-- uma boa sorte da gente comprar um bichinho, botar lá na manga—

3

M do 3

m eq co di tru th ha

G.D. Morton / World Development 113 (2019) 352–367 361

ruined. [. . .] Now, if God gives us good—good luck for us to buy a little animal, put it out there in the pasture— D: Mm hm— L: --tomorrow or the next day, in a time of need, you have something that you can—lay hold of, you know? [. . .] Yeah, an object inside the house, you can’t do that. D: Mm hm. L: You need something right here. ‘‘I’m going to sell this cow because I’m in need,” you know?

4 Fields are similarly risky: in one of the t aternity Wage on planting a field, respond ne well. 5 Because Maracujá is a village affiliat ovement, it is plausible that residents ar uality as a goal. The research design att mparing Maracujá with Rio Branco, a vill fferences in asset gendering were observe e that Rio Branco’s residents were clearly eir neighbors at Maracujá. Recipients’ a ve been impacted by the specific politica

D: Mm hm. L: --amanhã ou depois, numa precisão, você tem como você—pegar, né? [. . .] Mm— um objeto dentro de casa, você não pode fazer isso. D: Mm hm. L: Você está precisando aqui, ‘‘Eu vou vender essa vaca porque estou precisando,” né?

36 In every case where I recorded the final destination of livestock purchased with Maternity Wage money, the animal was ultimately given as a gift to a son or daughter of the woman who had received the payment. Note, however, that intra-household ownership is not a straightforward attribution. Donors seemed to continue to exercise considerable influence over the use of the gifted animals. To cite only one example, a woman at Maracujá received the Maternity Wage twice and purchased cattle for her children. Some of these cattle were later sold to pay for the woman’s brother’s trip to find work in a major city. The woman reported to me that eventually she hoped to sell the children’s remaining cattle in order to build a house. This would be, she said, ‘‘a house for them” (uma casa para eles.) In this case, as elsewhere, donors and their family members seem to play an important role in determining when to sell an animal that belongs to a child and how to use the resulting money. The money may be used for the children’s benefit, but the donor seems to have strong influence over how this takes place. 37 The authority that comes from having given away an asset (‘‘donor’s authority”) is clearly not the same as the authority that comes from owning that asset (‘‘owner’s authority.”) It might be argued that the latter is a stronger form of empowerment than the former. More broadly, there are good reasons to be suspicious of vague

Lara concluded that cattle were a better store of value—and hence a better use for the Maternity Wage. Cattle were not a better store of value because they were necessarily less risky. In the villages, cattle died frequently from disease, snakebite, predators, and drought.34

Rather, cattle stored value well because, if the risk paid off and the animals flourished, they could reproduce themselves and start a herd that would last longer than a chair or sofa.

Bolsa Família frames the saving challenge as the problem of how to put money together, but the Maternity Wage creates the opposite kind of saving challenge. It presents women with the problem of how to smooth money over time, stretching out a large sum so that it continues to generate income in the future. To solve this smooth- ing problem, women disturb the existing gendering of assets.35

Maternity Wage payments have an intrinsically disruptive effect because of their large size and their unpredictable nature. Bolsa Família, by virtue of its monthly regularity, makes it possible for mascates to develop a credit and saving system, a system based on the knowledge that women receive income every month. Bolsa Família inspires women to plan their purchases far in advance, and the regular nature of this system and these plans makes them con- ducive to reproducing the regular gendered asset order. The Mater- nity Wage, however, is not so regular. It comes not as the culmination of long-expected plans, but as the beginning of new ones.

6. Gift-giving and the Maternity Wage

Although the Maternity Wage allows women to become the owners of income-generating assets, oftentimes women do not remain owners for long. Commonly, at Maracujá and Rio Branco, women give livestock to their children. Thus, the Maternity Wage

wo households that reported spending the ents bitterly told me that the field had no

ed with the MST, a left-wing landless e unusually disposed to embrace gender empted to account for this possibility by age that is not connected to the MST. Few d between the two villages. However, it is influenced by the progressive example o ttitudes towards assets and gender may l context.

claims that a woman’s status is increased in the absence of concrete changes in power relations (see an important discussion in Molyneux, 2006, p. 439). It is worth noting, though, that at the field site, intra-household ownership is almost entirely non- juridical, since village residents rarely have recourse to the legal system to manage

t

f

frequently goes to purchase assets that end up in the hands not of a beneficiary, but of her daughters and sons.36

This gifting practice might seem to defeat any effort to use the Maternity Wage to build up women’s assets. After all, assets that a woman gives away do not have the same status as assets that a woman herself owns.37 Anthropological studies of gift exchange, however, have demonstrated how gift-givers can retain authority after the gift leaves their hands, thus ‘‘keeping while giving” (Mauss, 1967; Sahlins, 1972, Ch. 5; Weiner, 1976). At Maracujá and Rio Branco, women’s influence can persist even when women’s ownership has been transferred. By giving livestock to one or another child, a woman helps determine which child, in the future, will receive medical treatment or have post-secondary education paid for – both of these being commonly-discussed uses for a child’s ani- mal wealth.38

Women, as they grant such gifts, also secure their own reputa- tions.39 Years later, family members recall which donor gave which animal to whom. These recollections do not end with the animal’s slaughter or sale. In everyday conversation, children and relatives will trace out animal genealogies, remembering (as in one case I wit- nessed) that a particular cow is the granddaughter of a cow bought with a woman’s Maternity Wage money.

Bolsa Família money, like Maternity Wage money, can be spent on items destined for specific children. Women commonly use Bolsa Família to buy clothes for one child and school supplies for another. However, family members do not speak of or remember these items as gifts. They are simply necessities. It is the Maternity Wage, rather than Bolsa Família, that enables women to carry out acts of recognized generosity and thus to occupy an enduring place in the history of the household’s wealth.

The importance of this gift-giving network was driven home to me when I listened to Dona Eva, at Rio Branco, describe what she had done with the Maternity Wage that she received many years prior.

titles to livestock. Where legal institutions are not the relevant sources of authority, collective memory (such as the recollection of who gifted a particular animal) may be especially relevant in determining power relations. 38 Both girls and boys were recipients of livestock gifts, and I did not detect a pattern of preference for one gender in child ownership. 39 Mothers’ livestock gifts could be interpreted as a mechanism for retaining some authority over assets while avoiding the opprobrium that attaches to a violation of gender norms. By gifting a cow or a goat to her child, a mother insures that she is not herself the owner of an asset that is locally considered to be masculine. At the same time, she continues to wield long-term influence over the disposal of that asset. No- one at the field site described this logic to me explicitly, and I have no other direct indications that it is an important determinant of people’s actions, but it serves as a strong hypothesis; thanks are offered to an anonymous reviewer for suggesting it.

Woman receives Bolsa Família (monthly money)

Ins�tu�onal mechanism

Mascate ins�tu�on exists to offer credit for household assets

Predictability of money

Mental accoun�ng mechanism

Recipient considers money to be “reliable”

Result

Money spent on household assets

Woman receives Maternity Wage (lump-sum money)

Unpredictability of money

No ins�tu�on; woman uses saving prac�ce associated with men

Recipient considers money to be “risky”

Money spent on income- genera�ng assets

Small, regular payments

Large, one- �me payment

Fig. 4. Processes that explain different uses of monthly money and lump-sum money.

4

th in

362 G.D. Morton / World Development 113 (2019) 352–367

I took 200 reais, and I gave it to Andres [her husband] to buy a calf for Jeremias. [. . .] Still today, there are things from this money [. . .] It’s a cow today. She has already given birth about four times. It’s just that we sold the offspring [. . .] Andres sold them to take care of [Jeremias’s] teeth. It went for Jeremias. I gave it to him.

0 Fafchamps & Quisumbing, 2002 is an ex e value of gifts received inside Ethiopian heritance and gifts at the time of marriag

Peguei 200 reais, e dei a Andres para comprar um bezerro para Jeremias [. . .] Até hoje, tem desse dinheiro [. . .] É uma vaca hoje. Ela já pariu umas quatro vezes. Só que vendeu [. . .] Andres vendeu para cuidar dos dentes dele. Ficou para Jeremias. Eu dei para ele.

Dona Eva emphasized that the gift of the cow came from her: I gave it to him. She concluded by noting that in the future, she might make a similar gift to her husband. ‘‘If Andres is in need,” she told me, ‘‘I’ll give to him.” (Se Andres tiver precisão, eu dou para ele.) The Maternity Wage had helped Dona Eva make herself into the kind of person who could say that she gave gifts to her son, that she helped him get to a dentist, and that she envisioned herself as a generous donor to her husband.

In such cases, the gendering of assets comes to involve more than the ownership of an object. Women can use an asset to place themselves at the center of a web of exchanges. Through mecha- nisms like these, Maternity Wage recipients accrue honor and loy- alty from others, accumulating ‘‘wealth in people” (Guyer, 1993; see Sahlins 1972 on the ‘‘sociology of primitive exchange”). In gen- eral, empirical studies of asset gendering will be enriched by close attention to these dynamics that extend beyond ownership. This might, in the future, mean that researchers ask detailed questions about which assets are owned by children (and why) and that researchers chart out a history of recent gifts, with their givers and receivers.40

At Maracujá and Rio Branco, gifts are accompanied by a form of mental accounting that marks the Maternity Wage as a different kind of payment than Bolsa Família. Women tend to account for Bolsa Família (as discussed above) by setting aside a standard

ample of a study that catalogues in detail rural households, although the focus is on e.

sum of money each month to deveote to installments on a durable good, such as a mattress or a stove. In contrast, women account for their Maternity Wage purchases as intermediate stores of value. They say they do not know what the money will ultimately accom- plish. Kara, who used the Maternity Wage to buy a pig, gave the pig to her daughter. Kara described the pig’s usefulness in open-ended terms.

From that money afterwards we were going to buy something for her, you know?

Desse dinheiro depois a gente ia comprar uma coisa para ela, ne?

Saralinda spoke similarly about the Maternity Wage that she hoped to receive.

Then I plan to buy a cow, and, and, and buy for Venâncio’s [her son’s] future, you know? To be able to buy something for him in the future.

Aí eu planejo comprar vaca, e, e, e, comprar para o futuro de Venâncio, né? Para futuramente comprar alguma coisa para ele.

Saralinda employed a word – future – that plays a prominent role in women’s discussions of the Maternity Wage. While Bolsa Família leads to specific plans to obtain specific objects, the Maternity Wage opens the possibility of a more general, unanticipated future. Women are taking a risk, investing with the hope that tomorrow they will be able to purchase something better than what they could use the money to purchase today. The sense of openness was expressed by Marília as she described her potential goal for the Maternity Wage.

My plans, if I got the money, was to buy at least a little calf and leave it for him [her son], with his luck. Because he’s a child [. . .] So my plan was to buy something to leave in the

Meus planos, se eu recebesse o dinheiro, era eu comprar pelo menos uma bezerrinha e deixar na sorte dele. Porque ele está criança. [. . .] Então meu plano era de eu comprar alguma coisa para deixar no

G.D. Morton / World Development 113 (2019) 352–367 363

future. For him to buy what he needs. [. . .] Leave something, so that in the future, he won’t– won’t throw it in my face that I spent it uselessly, you know? Or that I didn’t leave anything for him. [. . .] So that when he grows up, he can say this, ‘‘Hey, my mom got the Maternity Wage, and she didn’t even know what she was accomplishing with it.”

futuro. Para ele comprar o que ele precisa. [. . .] Deixar alguma coisa, e para que no futuro, ele não—não jogue na minha cara que gastei à toa, né? Ou eu não deixei nada para ele [. . .] Então para quando ele crescer, ele dizer assim, ‘‘Ó, minha mãe pegou no Salário Maternidade, e não soube o que fez.”

41 See also Landsberger 1966. Thaler argues as follows: ‘‘[T]he MPC [marginal propensity to consume] out of windfall gains depends on the size of the gains. Small gains, relative to income, will be coded as current income, and spent. Larger gains will enter the assets account, where the MPC is lower (though still higher than the annuity value). The source of a change in wealth can also matter. Some windfalls, such as unrealized capital gains, are naturally treated as changes in the assets account. Others, such as the sale of a security, could be treated as income [. . .] Even cash receipts can enter the assets account if the inflow is in a large enough lump, and not considered regular income” (Thaler 1990, pp. 197–8).

In Marília’s accounting, she cannot know for sure what will happen to her Maternity Wage cash. This uncertainty is a sign of changed gender relations. Like a man who invests in cattle, she now faces the danger and the responsibility that comes with making a risky, large-sized purchase, one that others in the future may judge to be either useless or a great accomplishment. The money points towards the achievement of unforeseeable outcomes, ‘‘something to leave in the future.” Just as the Maternity Wage is disruptive in the present, it also holds out the possibility of disrupting the future, leading women to imagine a moment beyond their current capacity to predict.

7. Discussion: Processes that lead to different asset outcomes

Women at Maracujá and Rio Branco use monthly social assis- tance to buy durable household goods. These same women use lump-sum social assistance to buy agricultural assets that can gen- erate a flow of income in the future. What processes lead to these different results?

Both processes begin with the challenge of saving. Among households in poverty, a classic problem often arises (Collins et al., 2009, p. 96): how can the household shield a sum of money to save? A household member may solve this problem by pre- committing to a saving plan (Laibson, 1997). To pre-commit suc- cessfully, however, the member may need an institution that cred- ibly binds her/him/them to the saving plan.

For Bolsa Família money, mascates serve as such an institution. Mascates, as we have seen, know that Bolsa Família will arrive on a regular basis, and they deliver household goods immediately and collect installment payments later. Mascates allow women to make a credible commitment to turn over part of their future Bolsa Família money.

With the Maternity Wage, however, the situation is different. The Maternity Wage is a sizeable sum that must be spread out. Before the benefit arrives, women cannot make a pre- commitment with the mascate to take out credit and spend the sum in advance, because the Maternity Wage is so unpredictable; given the high rejection rates, the mascate does not have any assurance that the money will actually come. When women seek an instrument to save this sum, they turn towards the practice that men use to invest windfall profits: purchasing agricultural assets. Thus, one of the important differences between Bolsa Família and the Maternity Wage involves the different institutions that women use to commit to saving. Because of the unpredictable nature of the Maternity Wage, the existing institutions in rural Brazil do not allow women to pre-commit before the money arrives. Instead, women are encouraged to spend the benefit on income- generating assets.

But why income-generating assets? Why do women purchase furniture with Bolsa Família and cows with the Maternity Wage? To make sense of this distinction, it is helpful to turn towards the concept of ‘‘mental accounting,” as theorized by behavioral economists. Thaler (1990) argues that consumers do not act as if each dollar were fungible with every other. Instead, consumers tac- itly separate different streams of money into different mental accounts, such as a ‘‘current consumption account” for food and a ‘‘major assets account” for objects of value. Thus, for example, an increase in weekly wages may enter the consumption account and be spent entirely on better food, while a one-time inheritance may enter the asset account and get devoted fully to a new house, not affecting food consumption at all. Thaler observes that people tend to spend a small windfall differently from a large one, because people use different habits when accounting for differently-sized sums, placing these different sums into different mental accounts.41

At Maracujá and Rio Branco, women confirm the theory by clas- sifying Bolsa Família money into a ‘‘low-risk, small asset” account, while they place Maternity Wage money in a ‘‘high-risk, large asset” account. They offer explicit descriptions of these different accounting habits. Women say that they manage Bolsa Família money by getting into the habit of setting aside a consistent sum each month. They devote this sum to a visible saving goal: the household good that the mascate has already delivered. Beneficia- ries treat Bolsa Família, in other words, like reliable money. It is to be spent, reliably, on the sort of normal, low-risk assets, such as stoves and tables, that village residents expect women to purchase.

Women describe an altogether different practice for accounting for the Maternity Wage. This money is irregular and unpredictable. Women say that they use the money to purchase assets that they understand to be unpredictable as well. They pay for livestock that may fall sick and fields that may not bear fruit. They say that it is money meant to ‘‘produce,” to make an ‘‘income” or a ‘‘profit.” They hope that the profit will go to buy ‘‘something” in the future, but they do not specify that ‘‘something.” In other words, women treat the Maternity Wage like risky money. Just like gambling win- nings in other contexts (Thaler, 1990), the Maternity Wage is not seen as appropriate for mundane expenditures. It is good for spending on the sort of high-risk agricultural assets that break stereotypes and hold out the possibility of changing relations inside the household.

As the comparison between Bolsa Família and the Maternity Wage demonstrates, a benefit’s payment schedule can have an impact on at least two components of the asset accumulation pro- cess. The schedule can affect saving institutions and mental account- ing. First, differently timed payments promote different sorts of institutions for solving the saving problem. Second, by timing pay- ments, policymakers cue beneficiaries about how to use the money, whether for safe or risky endeavors, since different pay- ment schedules activate different kinds of mental accounts. For households at Maracujá and Rio Branco, thus, monthly cash and lump sums are simply not the same kind of money, and these dif- ferent kinds of money have divergent effects on the stock of assets owned by each member of the household. Social benefits can help women acquire (feminine-stereotyped) household durables or

43 For information on Jóvenes con Prospera, see ‘‘Acuerdo por el que se emiten las Reglas de Operación de PROSPERA Programa de Inclusión Social, para el ejercicio fiscal 2016,” in Diário Oficial de la Unión, México, December 30, 2015 (http://dof.gob.mx/ nota_detalle.php?codigo=5421756&fecha=30/12/2015). 44 Also note Bolivia’s Bono Juancito Pinto, an unusual example of a conditional cash transfer that provides the benefit through a single annual payment (Aguilar Pacajes, 2014). 45 This topic is also relevant to the microfinance literature, where scholars investigate the question of how microfinance clients decide between buying productive assets and making other purchases. Scholars have been attentive to the gendered dimension of the issue (Schuster, 2015; Kar & Schuster, 2016; Banerjee, Duflo, Glennerster, & Kinnan, 2015; Haile, Bock & Folmer, 2012). 46 Thanks for this point are due to Jonas Nahm. 47 Thanks are gratefully given to an anonymous reviewer for suggesting the notion of ‘‘serendipity.” 48 It may be especially important to evaluate existing programs that create savings accounts for women attached to CCTs. Bolsa Família includes such accounts, although women at the field site reported that they rarely used the accounts to store money. Also see similar programs in Peru (Meltzer, 2013; De Los Rios, & Trivelli, 2011) and Mexico (Chiapa & Prina, 2015). 49 Moser emphasizes (2008, p. 91) that asset-accumulation policy differs from social protection policy in that asset-accumulation policy involves promoting risk. Often, such a policy will encourage beneficiaries to seek out risk, in a sustainable and

364 G.D. Morton / World Development 113 (2019) 352–367

(masculine-stereotyped) agricultural assets, depending on the structure of the program.

Are these results particular to northeastern Brazil? Ethnography works with intrinsically local data, but the data may shed light on processes that take place in households beyond the area of focus. The article concludes by considering that possibility. (Insert Fig. 4 here.)

8. Conclusion: Lump sums as a tool for change

Inside a household, social assistance programs do more than increase income;42 they transform relationships and habits of mind. Ethnography traces such changes, which come embedded with the money. This article has described two mechanisms – institutions and mental accounting – that lead the recipients of social assistance to convert unpredictable lump sums into income-generating assets. These mechanisms may be relevant outside of northeastern Brazil. The mental-accounting result, in particular, is a good candidate for portability, since it conforms to the general prediction, made by behavioral economists, that agents think of one-time payments as resources for risk-taking (Thaler, 1990, pp. 197–8). Effects like these open up possibilities for policy. Because of their disruptive power, lump sums might be mobilized to reduce the gender asset gap.

Lump sum policies already create positive disruption in other policy areas. For example, the Earned Income Tax Credit (EITC) in the United States provides low-income workers with a single annual payment; just as with the Maternity Wage, recipients have difficulty predicting the payment in advance (Edin, Tach, & Halpern-Meekin, 2014; Congdon, Kling, & Mullainathan, 2009). Researchers find that recipients are satisfied with the lump-sum quality of the payment, using it as a savings device so they can pur- chase major assets, like cars, that facilitate upward mobility (Halpern-Meekin, Edin, Tach, & Sykes, 2015, p. 70). When the US government created an option for recipients to obtain the EITC ahead of time in monthly payments, the option proved unpopular (Halpern-Meekin et al., 2015, p. 71). Similar to the EITC, but set in low-income nations, ‘‘graduation programs” provide beneficiaries with a one-time infusion of assets or cash, along with regular income support for a two-year period. In an RCT, graduation pro- grams increased the value of the productive assets owned by households (Banerjee et al., 2015; Bandiera et al., 2013; see Blattman, Green, Jamison, Lehmann, & Annan, 2016 for a similar program in Uganda and Martinez, 2004 for a lump-sum benefit in Bolivia). Relatedly, an evaluation of GiveDirectly in Kenya found increases in asset accumulation (and declines in the stress hor- mone cortisol) when payments were disbursed as lump sums (Haushofer & Shapiro, 2016).

Conditional cash transfers are generally designed to provide payments on a regular, ongoing schedule, not as a lump sum. Nonetheless, some CCTs have included lump-sum features. Argen- tina’s Asignación Universal por Hijo automatically saves 20% of each family’s monthly benefit in a bank account. The family accesses the savings at year’s end by proving that children have attended school and received vaccines (Rabinovich & Diepeveen, 2015). Participants in Ecuador’s Bono de Desarrollo Humano can save multiple months of the benefit in an account (Rinehart & McGuire, 2017, p. 167). In Colombia, Más Familias en Acción pro- vides families with a bimonthly cash transfer, and in 2005 the city of Bogotá supplemented this transfer with two pilot programs that provided randomly-selected families with lump sums (Barrera- Osorio, Bertrand, Linden, & Perez-Calle, 2008). Mexico’s Prospera CCT offers Jóvenes con Prospera, a one-time cash benefit to stu-

42 Increasing income also matters. It may be the case that women’s power in the household would be even more powerfully strengthened if the Bolsa Família benefit became more generous. This article does not consider that possibility.

dents who graduate high school.43 Similarly, from 2007 to 2010, New York City’s short-lived Opportunity NYC program topped off its monthly cash transfers with one-time incentives for achieve- ments, such as passing high school exams (Peck & Theodore, 2015, p. 57).44 PANES, in Uruguay, provided new beneficiaries with back pay from their initial date of application, and this money often amounted to a sizable lump sum; Corboz found (in keeping with the present article) that women often spent the sum on an income-generating asset (2013, p. 70–77). The wide variety of these experiences suggests the viability of further research to determine not only the short-run outcomes of lump sums, but also the deeper mechanisms that these payments activate inside households.45

One unexpected conclusion of the present article is that unpre- dictable payments to women can help reduce the gender asset gap. Unpredictability is not often a feature of good social policy.46

Nonetheless, it may be possible for policymakers to generate benefi- cial, serendipitous forms of unpredictability.47 For example, a lump- sum payment to women could be conditioned on the fulfillment of an important but difficult (hence difficult-to-predict) task, like grad- uating school, receiving a job offer, completing house construction, becoming pregnant (Fallon, et al., 2017), or achieving a saving goal (Zimmerman & Moury, 2009; see also Chiapa and Prina, 2015).48

Such a benefit could be incorporated in an existing CCT program. By rendering the lump sum difficult to anticipate, it is hoped, policy- makers will prevent the money from becoming assimilated into existing gendered institutions, instead turning the payment into a true positive shock to women’s budgets.

The broader principle at stake, here, is that disruption can have salutary effects for household budgeting.49 CCTs have largely been grounded in the opposite principle, the model of regular income sup- port. This model is proven to achieve notable improvements in health, nutrition, education, and income inequality (Lustig, Lopez- Calva, & Ortiz-Juarez, 2013; de Brauw, Gilligan, Hoddinott, & Roy, 2015; Skoufias, 2005; Kabeer, Piza, & Taylor, 2012; Fizbein, Schady, & Ferreira, 2009; Saavedra & García, 2012),50 and consistent pay- ments like CCTs are doubtlessly crucial elements in a study safety net. Barrientos, summing up the evidence, concludes, ‘‘In order to maximize their effect on investment, social transfers need to be reg- ular and reliable” (2012, p. 18). But irregularity and unreliability may serve important purposes as well. Particularly when confronting deeply rooted forms of inequality, such as the gender asset gap,

manageable way. 50 See Piperata, McSweeny, & Murrieta, 2016 for a more complex view of outcomes. Using detailed anthropometric data, they demonstrate a possible connection between Bolsa Família and worsened nutritional results for beneficiary households in the Amazon.

G.D. Morton / World Development 113 (2019) 352–367 365

unsettling instruments have a role to play. The toolshed of social assistance should contain implements for stability and disruption alike.

Conflict of interest

None.

Acknowledgments

This article exists, in large part, due to the guidance, inspiration, and faith of Alicia Menendez and Harold Pollock. Marcos Rangel, Mauro Barbosa de Almeida, and Manuela Carneiro da Cunha pro- vided valuable support. Helpful commentary was offered by Bill Sites, Patrick Heller, Catherine Lutz, Julia Vich-Bertran, Lindsey Reynolds, Narges Bajoghli, and James Green.

Funding

This work was supported by the Inter-American Foundation (Grassroots Development Fellowship), the Social Science Research Council (International Dissertation Research Fellowship), and the Center for Latin American Studies at the University of Chicago (Tin- ker Field Research Grant). All are gratefully acknowledged.

References

Adato, M., De la Briere, B., Mindek, D., & Quisumbing, A. (2000). The impact of PROGRESA on women’s status and intrahousehold relations. Washington, DC: International Food Policy Research Institute. Retrieved from https://pdfs. semanticscholar.org/1a78/5e4625fd4c6fe08c2ef7c30d1a07eaa7fcad.pdf.

Adato, M., & Roopnaraine, T. (2010). Women’s status, gender relations, and conditional cash transfers. In M. Adato & J. Hoddinott (Eds.), Conditional cash transfers in Latin America. Baltimore: Johns Hopkins.

Agarwal, B. (1994). A field of one’s own: Gender and land rights in South Asia. Cambridge: Cambridge University Press.

Agarwal, B. (1997). ‘‘Bargaining” and gender relations: Within and beyond the household. Feminist Economics, 3(1), 1–51.

Aguilar Pacajes, H. (2014). Evaluación del impacto del ‘‘Bono Juancito Pinto” en Bolivia. Revista de Análisis del BCB, 21(2), 37–66.

Ahlert, M. (2013). A ‘‘precisão” e o ‘‘luxo”: usos do benefício do Programa Bolsa Família entre as quebradeiras de coco de Codó (MA). Política & Trabalho, 38, 69–86.

Anderson, S., & Eswaran, M. (2009). What determines female autonomy? Evidence from Bangladesh. Journal of Development Economics, 90(2), 179–191.

Angelucci, M. (2008). Love on the rocks: Domestic violence and alcohol abuse in rural Mexico. The BE Journal of Economic Analysis & Policy, 8(1). Retrieved from https://www.degruyter.com/view/j/bejeap.2008.8.1/bejeap.2008.8.1.1766/ bejeap.2008.8.1.1766.xml.

Ansell, A. M. (2014). Zero hunger: Political culture and antipoverty policy in Northeast Brazil. Chapel Hill: The University of North Carolina Press.

Antonopoulos, R., & Floro, M. S. (2005). Asset ownership along gender lines: Evidence from Thailand. Journal of Income Distribution, 13(3/4). http://levy.org/ pubs/wp418.pdf.

Attanasio, O., & Lechene, V. (2002). Tests of income pooling in household decisions. Review of Economic Dynamics, 5(4), 720–748.

Bailey, J. P. (2010). Rethinking poverty: Income, assets, and the Catholic social justice tradition. Notre Dame: University of Notre Dame Press.

Bandiera, O., Burgess, R., Das, N., Gulesci, S., Rasul, I., & Sulaiman, M. (2013). Can basic entrepreneurship transform the economic lives of the poor? LSE Working Paper. http://R4d.Dfid.Gov.Uk/Pdf/Outputs/PEDL/Bandiera.Pdf.

Banerjee, A., Duflo, E., Glennerster, R., & Kinnan, C. (2015). The miracle of microfinance? Evidence from a randomized evaluation. American Economic Journal: Applied Economics, 7(1), 22–53.

Banerjee, A., Duflo, E., Goldberg, N., Karlan, D., Osei, R., Parienté, W., ... Udry, C. (2015). A multifaceted program causes lasting progress for the very poor: Evidence from six countries. Science, 348(6236), 772.

Barrera-Osorio, F., Bertrand, M., Linden, L. L., & Perez-Calle, F. (2008). Conditional cash transfers in education design features, peer and sibling effects evidence from a randomized experiment in Colombia. National Bureau of Economic Research.

Barrientos, A. (2012). Social transfers and growth: What do we know? What do we need to find out? World Development, 40(1), 11–20.

Blattman, C., Green, E. P., Jamison, J., Lehmann, M. C., & Annan, J. (2016). The returns to microenterprise support among the ultrapoor: A field experiment in postwar uganda. American Economic Journal: Applied Economics, 8(2), 35–64. https://doi. org/10.1257/app.20150023.

Bohannan, L., & Bohannan, P. (1968). Tiv economy. Evanston: Northwestern University Press.

Briggs, C. L. (1986). Learning how to ask: A sociolinguistic appraisal of the role of the interview in social science research. Cambridge University Press.

Chiapa, C., & Prina, S. (2015). Delivering conditional cash transfers via savings accounts. One-Pager No. 292. International Policy Centre for Inclusive Growth. https://faculty.weatherhead.case.edu/prina/pdfs/CP_IPC_UNDP_2015.pdf.

Chiappori, P.-A. (1992). Collective labor supply and welfare. Journal of Political Economy, 437–467.

Collins, D., Morduch, J., Rutherford, S., & Ruthven, O. (2009). Portfolios of the poor: How the world’s poor live on $2 a day. Princeton: Princeton University Press.

Congdon, W. J., Kling, J. R., & Mullainathan, S. (2009). Behavioral economics and tax policy. National Tax Journal, 375–386.

Corboz, J. (2013). Third-way neoliberalism and conditional cash transfers: The paradoxes of empowerment, participation and self-help among poor Uruguayan women. The Australian Journal of Anthropology, 24(1), 64–80.

Cramer, R., & Shanks, T. R. W. (2014). The assets perspective: The rise of asset building and its impact on social policy. Palgrave Macmillan.

Davis, B., Handa, S., Stampini, M., & Winters, P. (2002). The lure of tequila and the bestowing of motherly love: Does it matter whether public cash transfers are given to women or men? Evidence from the PROGRESA and PROCAMPO programs in rural Mexico. Zaragosa (Spain): Paper No. 24836 presented at the Xth EAAE Congress, Exploring Diversity in the European Agri-Food System. Retrieved from https:// www.researchgate.net/profile/Marco_Stampini/publication/23509173_The_ Lure_of_Tequila_and_the_Bestowing_of_Motherly_Love_Does_it_Matter_ Whether_Public_Cash_Transfers_are_Given_to_Women_or_Men_ Evidence_ from_the_PROGRESA_and_PROCAMPO_Programs_in_Rural_Mexico/links/ 0deec5253169ef222b000000.pdf.

de Brauw, A., Gilligan, D. O., Hoddinott, J., & Roy, S. (2014). The impact of Bolsa Família on women’s decision-making power. World Development, 59, 487–504.

de Brauw, A., Gilligan, D. O., Hoddinott, J., & Roy, S. (2015). The impact of Bolsa Família on schooling. World Development, 70(Suppl. C), 303–316. https://doi. org/10.1016/j.worlddev.2015.02.001.

De Los Rios, J., & Trivelli, C. (2011). Savings mobilization in conditional cash transfer programs: Seeking mid-term impacts. Instituto de Estudios Peruanos. Proyecto Capital. https://www.ids.ac.uk/files/dmfile/ DeLosRiosTrivelli2011SavingsmobilisationinCCTinPeruCSPconferencedraft.pdf.

Deere, C. D. (2010). Household wealth and women’s poverty: Conceptual and methodological issues in assessing gender inequality in asset ownership. In S. Chant (Ed.), The international handbook of gender and poverty: Concepts, research, policy (pp. 347–352). Cheltenham: Edward Elgar.

Deere, C. D., & Doss, C. R. (2006). The gender asset gap: What do we know and why does it matter? Feminist Economics, 12(1–2), 1–50.

Deere, C. D., & León, M. (2003). The gender asset gap: Land in Latin America. World Development, 31(6), 925–947.

Doss, C. R., Deere, C. D., Oduro, A. D., & Swaminathan, H. (2014). The gender asset and wealth gaps. Development, 57(3–4), 400–409.

Edin, K., Tach, L., & Halpern-Meekin, S. (2014). Tax code knowledge and behavioral responses among EITC recipients: Policy insights from qualitative data. Journal of Policy Analysis and Management, 33(2), 413–439.

Epstein, A. L. (Ed.). (1967). The craft of social anthropology. London: Tavistock. Fafchamps, M., & Quisumbing, A. R. (2002). Control and ownership of assets within

rural Ethiopian households. Journal of Development Studies, 38(6), 47–82. Fallon, K. M., Mazar, A., & Swiss, L. (2017). The development benefits of maternity

leave. World Development, 96, 102–118. Fizbein, A., Schady, N. R., & Ferreira, F. H. (2009). Conditional cash transfers: Reducing

present and future poverty. Washington: World Bank. Gertler, P. J., Martinez, S. W., & Rubio-Codina, M. (2012). Investing cash transfers to

raise long-term living standards. American Economic Journal: Applied Economics, 4(1), 164–192.

Gil-García, Ó. F. (2016). Gender equality, community divisions, and autonomy: The Prospera conditional cash transfer program in Chiapas, Mexico. Current Sociology, 64(3), 447–469.

Gomes, S. (2011). Notas preliminares de uma crítica feminista aos programas de transferência direta de renda– o caso do Bolsa Família no Brasil. Textos & Contextos (Porto Alegre), 10(1), 69–81.

Guyer, J. I. (1993). Wealth in people and self-realization in equatorial Africa. Man, 28 (2), 243–265. https://doi.org/10.2307/2803412.

Haile, H. B., Bock, B., & Folmer, H. (2012). Microfinance and female empowerment: Do institutions matter? Women’s Studies International Forum, 35(4), 256–265. https://doi.org/10.1016/j.wsif.2012.04.001.

Halpern-Meekin, S., Edin, K., Tach, L., & Sykes, J. (2015). It’s not like I’m poor: How working families make ends meet in a post-welfare world. Berkeley: University of California Press.

Handa, S., Peterman, A., Davis, B., & Stampini, M. (2009). Opening up Pandora’s box: The effect of gender targeting and conditionality on household spending behavior in Mexico’s PROGRESA program. World Development, 37(6), 1129–1142.

Hardesty, M. (2015). Epistemological binds and ethical dilemmas in frontline child welfare practice. Social Service Review, 89(3), 455–498.

Haushofer, J., & Shapiro, J. (2016). The short-term impact of unconditional cash transfers to the poor: Experimental evidence from Kenya. The Quarterly Journal of Economics, 131(4), 1973–2042. https://doi.org/10.1093/qje/qjw025.

Hoddinott, J., & Haddad, L. (1995). Does female income share influence household expenditures? Evidence from Côte d’Ivoire. Oxford Bulletin of Economics and Statistics, 57(1), 77–96.

366 G.D. Morton / World Development 113 (2019) 352–367

Holmes, R., & Jones, N. (2013). Gender and social protection in the developing world: Beyond mothers and safety nets. London: Zed Books Ltd.

Kabeer, N., Piza, C., & Taylor, L. (2012). What are the economic impacts of conditional cash transfer programmes? A systematic review of the evidence. London: EPPI- Centre, Social Science Research Unit, Institute of Education, University of London. Retrieved from https://assets.publishing.service.gov.uk/media/ 57a08a6840f0b649740005a4/CCTprogrammes2012Kabeer.pdf.

Kar, S., & Schuster, C. (2016). Comparative projects and the limits of choice: Ethnography and microfinance in India and Paraguay. Journal of Cultural Economy, 9(4), 347–363. https://doi.org/10.1080/17530350.2016.1180632.

Koch, A. K., & Nafziger, J. (2016). Goals and bracketing under mental accounting. Journal of Economic Theory, 162, 305–351.

Lagarde, M., Haines, A., & Palmer, N. (2007). Conditional cash transfers for improving uptake of health interventions in low-and middle-income countries: A systematic review. JAMA, 298(16), 1900–1910.

Laibson, D. (1997). Golden eggs and hyperbolic discounting. The Quarterly Journal of Economics, 443–477.

Landsberger, M. (1966). Windfall income and consumption: Comment. The American Economic Review, 56(3), 534–540.

Lavinas, L., & Nicoll, M. (2006). Pobreza, transferências de renda e desigualdades de gênero: Conexões diversas. Parcerias Estratégicas, 11(22), 39–76.

Lindert, K., Linder, A., Hobbs, J., & De la Brière, B. (2007). The nuts and bolts of Brazil’s Bolsa Família program: Implementing conditional cash transfers in a decentralized context. Washington, DC: Social Protection Discussion Paper 0709. World Bank. Consultative Group to Assist the Poor.

Loera, N. R. (2010). ‘Encampment time’: An anthropological analysis of the land occupations in Brazil. Journal of Peasant Studies, 37(2), 285–318.

Lui, G. H., & Molina, S. M. G. (2013). Benefícios sociais e transição de modos de vida rurais: uma análise do Bolsa Família e da aposentadoria rural entre pequenos produtores na Amazônia. Politica & Trabalho, 38, 137–155.

Lundberg, S., & Pollak, R. A. (1993). Separate spheres bargaining and the marriage market. Journal of Political Economy, 988–1010.

Lustig, N., Lopez-Calva, L. F., & Ortiz-Juarez, E. (2013). Declining inequality in Latin America in the 2000s: The cases of Argentina, Brazil, and Mexico. World Development, 44(Supplement C), 129–141. https://doi.org/10.1016/ j.worlddev.2012.09.013.

Maldonado, I., Nájera, M., & Segovia, A. (2006). Efectos del Programa Oportunidades en las relaciones de pareja y familiares. In M. de la P. López & V. Salles (Eds.), El Programa Oportunidades examinado desde el género (pp. 95–130). México, D. F.: Programa de Desarrollo Humano Oportunidades. Retrieved from http:// ogp.colmex.mx/libreria/OportunidadesOGP.pdf.

Maluccio, J. A. (2010). The impact of conditional cash transfers on consumption and investment in Nicaragua. Journal of Development Studies, 46(1), 14–38.

Mariana, S. A., & Carloto, C. M. (2009). Gênero e combate à pobreza: Programa Bolsa Família. Estudos Feministas (Florianópolis), 17(3), 901–908.

Martinez, S. (2004). Pensions, poverty and household investments in Bolivia https:// eml.berkeley.edu/webfac/bardhan/e271_f04/martinez.pdf. Berkeley: University of California.

Mauss, M. (1967). The gift: Forms and functions of exchange in archaic societies. (I. Cunnison Trans.). New York: W. W. Norton & Company. Original work published 1925.

McElroy, M. B., & Horney, M. J. (1981). Nash-bargained household decisions: Toward a generalization of the theory of demand. International Economic Review, 333–349.

Meltzer, J. (2013). ‘Good citizenship’ and the promotion of personal savings accounts in Peru. Citizenship Studies, 17(5), 641–652.

Ministério da Previdência Social. (n.d.a). Anuário estatístico da Previdência Social 2014. Brasília: Secretaria de Políticas de Previdência Social. Retrieved from http://www.mtps.gov.br/dados-abertos/dados-da-previdencia/previdencia- social-e-inss/anuario-estatistico-da-previdencia-social-aeps.

Ministério da Previdência Social. (n.d.b). Anuário estatístico da Previdência Social. Suplemento histórico 2011. Brasília: Secretaria de Políticas de Previdência Social. Retrieved from http://www.mtps.gov.br/dados-abertos/dados-da- previdencia/previdencia-social-e-inss/anuario-estatistico-da-previdencia- social-aeps.

Molyneux, M. (2006). Mothers at the service of the new poverty agenda: Progresa/ Oportunidades, Mexico’s conditional transfer programme. Social Policy & Administration, 40(4), 425–449.

Molyneux, M. (2009). Conditional cash transfers: A pathway to women’s empowerment? Pathways Brief, 5.

Morissawa, M. (2001). A história da luta pela terra e o MST. São Paulo: Expressão Popular.

Morton, G. D. (2013). Acesso à permanência: Diferenças econômicas e práticas de gênero em domicílios que recebem Bolsa Família no sertão baiano. Política e Trabalho, 1(38), 43–67.

Morton, G. D. (2015a). Leaving labor: Reverse migration, welfare cash, and the specter of the commodity in Northeastern Brazil. Chicago: Department of Anthropology and School of Social Service Administration, University of Chicago.

Morton, G. D. (2015b). Managing transience: Bolsa Família and its subjects in an MST landless settlement. The Journal of Peasant Studies, 42(6), 1283–1305.

Morton, G. D. (forthcoming). Saying no: Bolsa Família, self-employment, and the rejection of jobs in northeastern Brazil. In M.E. Balen & M. Fotta, Money from the Government in Latin America: Conditional Cash Transfer Programs and Rural Lives. London: Routledge.

Moser, C. (2008). Reducing global poverty: The case for asset accumulation. Washington: Brookings Institution Press.

Moser, C. (2010a). Ordinary families, extraordinary lives: Assets and poverty reduction in Guayaquil, 1978–2004. Washington: Brookings Institution Press.

Moser, C. (2010b). Moving beyond gender and poverty to asset accumulation: Evidence from low-income households in Guayaquil, Ecuador. In S. Chant (Ed.), The international handbook of gender and poverty: Concepts, research, policy (pp. 391–398). Cheltenham: Edward Elgar.

Nandi, A., Hajizadeh, M., Harper, S., Koski, A., Strumpf, E. C., & Heymann, J. (2016). Increased duration of paid maternity leave lowers infant mortality in low-and middle-income countries: A quasi-experimental study. PLoS Medicine, 13(3) e1001985.

Natali, L., Handa, S., Peterman, A., Seidenfeld, D., & Tembo, G. (2016). Making money work: Unconditional cash transfers allow women to save and re-invest in rural Zambia (Office of Research – Innocenti Working Papers No. WP-2016-02). Florence (Italy): UNICEF. Retrieved from https://ideas.repec.org/p/ucf/inwopa/ inwopa827.html.

Neiverth, E. M. H. B., & Mandalozzo, S. S. N. (2009). A licença-maternidade e sua ampliação facultativa. Revista do Tribunal Regional do Trabalho da 9a Região, 34 (63), 175–188.

Parry, J. P., & Bloch, M. (Eds.). (1989). Money and the morality of exchange. Cambridge, England; New York: Cambridge University Press.

Peck, J., & Theodore, N. (2015). Fast policy: Experimental statecraft at the thresholds of neoliberalism. Minneapolis: University of Minnesota.

Pereira, M. C., & Ribeiro, F. B. (2013). No Areal das mulheres: um benefício em família. Política & Trabalho, 38, 87–104.

Piperata, B. A., McSweeney, K., & Murrieta, R. S. (2016). Conditional cash transfers, food security, and health: Biocultural insights for poverty-alleviation policy from the Brazilian Amazon. Current Anthropology: A World Journal of the Sciences of Man, 6, 806–826.

Pires, F. F. (2013). Comida de criança e o Programa Bolsa Família: Moralidade materna e consumo alimentar no semiárido. Política e Trabalho, 38, 123–135.

Pires, F. F. (2014). Child as family sponsor: An unforeseen effect of Programa Bolsa Familia in Northeastern Brazil. Childhood, 21(1), 134–147.

Pires, F. F., & da Silva Jardim, G. A. (2014). Geração Bolsa Família: Escolarização, trabalho infantil e consumo na casa sertaneja (Catingueira/PB). Revista Brasileira de Ciências Sociais, 29(85), 99–112.

Quisumbing, A. R., & Maluccio, J. A. (2000). Intrahousehold allocation and gender relations: New empirical evidence from four developing countries (FCMD Discussion Paper No. 84). Washington, DC: International Food Policy Research Institute. Retrieved from https://www.researchgate.net/profile/Agnes_Quisumbing/ publication/239548128_Intrahousehold_Allocation_and_Gender_Relations_ New_Empirical_Evidence_form_Four_Developing_Countries/links/ 0deec524076ce73f4b000000.pdf.

Quisumbing, A. R., & Maluccio, J. A. (2003). Resources at marriage and intrahousehold allocation: Evidence from Bangladesh, Ethiopia, Indonesia, and South Africa. Oxford Bulletin of Economics and Statistics, 65(3), 283–327.

Rabinovich, L., & Diepeveen, S. (2015). The design of conditional cash transfers: Experiences from Argentina’s Universal Child Allowance. Development Policy Review, 33(5), 637–652.

Ranganathan, M., & Lagarde, M. (2012). Promoting healthy behaviours and improving health outcomes in low and middle income countries: A review of the impact of conditional cash transfer programmes. Preventive Medicine, 55, S95–S105.

Rego, W. G. D. L., & Pinzani, A. (2013). Vozes do Bolsa Família: Autonomia, dinheiro e cidadania. São Paulo: Editora UNESP.

Rinehart, C. S., & McGuire, J. W. (2017). Obstacles to takeup: Ecuador’s conditional cash transfer program, The Bono de Desarrollo Humano. World Development, 97, 165–177.

Rivera, L., Hernández, B., & Castro, R. (2006). Asociación entre la violencia de pareja contra las mujeres de las zonas urbanas en pobreza extrema y la incorporación al Programa Oportunidades. In M. de la P. López & V. Salles (Eds.), El Programa Oportunidades examinado desde el género (pp. 69–94). México, D. F.: Programa de Desarrollo Humano Oportunidades. Retrieved from http://ogp.colmex.mx/ libreria/OportunidadesOGP.pdf.

Rubalcava, L., Teruel, G., & Thomas, D. (2009). Investments, time preferences, and public transfers paid to women. Economic Development and Cultural Change, 57 (3), 507–538.

Saavedra, J. E., & García, S. (2012). Impacts of conditional cash transfer programs on educational outcomes in developing countries: A meta-analysis. RAND Labor and Population Working Paper Series, WR-921-1. Retrieved from https://pdfs. semanticscholar.org/1a92/253ac56da92e8e5737e0f62784ac47f110b4.pdf.

Sahlins, M. (1972). Stone age economics. Chicago: Aldine-Atherton. Schuster, C. E. (2015). Social collateral: Women and microfinance in Paraguay’s

smuggling economy. Berkeley: University of California Press. Sherraden, M. (1988). Rethinking social welfare: Toward assets. Social Policy, 18(3),

37–43. Sherraden, M. (1991). Assets and the poor. Armonk: ME Sharpe. Sherraden, M. (2005). Introduction. In M. W. Sherraden (Ed.), Inclusion in the

American Dream: Assets, poverty, and public policy (pp. 3–19). Oxford: Oxford University Press.

Skoufias, E. (2005). PROGRESA and its impacts on the welfare of rural households in Mexico (Research Report 139). Washington, DC: International Food Policy Research Institute.

Smith, Y. (2014). Rethinking decision making: An ethnographic study of worker agency in crisis intervention. Social Service Review, 88(3), 407–442.

Soares, F. V., & Silva, E. (2010). Conditional cash transfer programmes and gender vulnerabilities in Latin America. Overseas Development Institute. Retrieved from https://www.odi.org/resources/docs/6260.pdf.

G.D. Morton / World Development 113 (2019) 352–367 367

Ssewamala, F. M., Sperber, E., Zimmerman, J. M., & Karimli, L. (2010). The potential of asset-based development strategies for poverty alleviation in Sub-Saharan Africa. International Journal of Social Welfare, 19(4), 433–443.

Streuli, N. (2012a). Child protection: A role for conditional cash transfer programmes? Development in Practice, 22(4), 588–599.

Streuli, N. (2012b). Children’s experiences of Juntos, a conditional cash transfer scheme in Peru (Working Paper No. 78). Young Lives, Oxford Department of International Development, University of Oxford. Retrieved from http://repositorio. minedu.gob.pe/handle/123456789/1570.

Suárez, M., & Libardoni, M. (2007). The impact of the Bolsa Família Program: changes and continuities in the social status of women. In J. Vaitsman & R. Paes- Sousa (Eds.), Evaluation of MDS Policies and Programs. Results 2 (Vol. 2– Bolsa Família e Assistência Social, pp. 117–60). Brasília/DF: Ministério do Desenvolvimento Social e Combate à Fome. Retrieved from http://www.ipc- undp.org/doc_africa_brazil/MDS_Vol2_eng.pdf.

Taussig, M. T. (1980). The Devil and commodity fetishism in South America. Chapel Hill: University of North Carolina Press.

Thaler, R. H. (1990). Anomalies: Saving, fungibility, and mental accounts. The Journal of Economic Perspectives, 4(1), 193–205.

Thomas, D. (1990). Intra-household resource allocation: An inferential approach. Journal of Human Resources, 25(4), 635–664.

Thorne, S. (2000). Data analysis in qualitative research. Evidence Based Nursing, 3(3), 68–70.

Valencia Lomelí, E. (2008). Conditional cash transfers as social policy in Latin America: An assessment of their contributions and limitations. Annual Review of Sociology, 34, 475–499.

Van den Bold, M., Quisumbing, A. R., & Gillespie, S. (2013). Women’s empowerment and nutrition: An evidence review (IFPRI Discussion Paper 01294). Retrieved from https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2343160

Villatoro, P. (2005). Conditional cash transfer programmes: Experiences from Latin America. CEPAL Review, 86, 83–96.

Wang, S.-Y. (2014). Property rights and intra-household bargaining. Journal of Development Economics, 107, 192–201.

Weiner, A. B. (1976). Women of value, men of renown: New perspectives in Trobriand exchange. Austin: University of Texas Press.

Wolford, W. (2010). This land is ours now: Social mobilization and the meanings of land in Brazil. Durham N.C.: Duke University Press.

Yablonski, J., & Peterman, A. (2016, May 13). Cash transfers: What’s gender got to do with it? https://blogs.unicef.org/blog/Cash-Transfers-Whats-Gender-Got-to- Do-with-It/.

Yoong, J., Rabinovich, L., & Diepeveen, S. (2012). The impact of economic resource transfers to women versus men: A systematic review. London: EPPI-Centre, Social Science Research Unit, Institute of Education, University of London. Retrieved from http://www.fsnnetwork.org/sites/default/files/economictransfer2012 yoong.pdf.

Zimmerman, J., & Moury, Y. (2009). Savings-linked conditional cash transfers: A new policy approach to global poverty reduction (Global Assets Project Policy Brief). Washington, DC: New America Foundation.