Report No: AUS9147 Central America Migration and Women’s Agency in Agriculture - Women in Agriculture: The Impact of Male Out- Migration on Women’s Agency, Household Welfare and Agricultural Productivity May 2015 GFADR LATIN AMERICA AND CARIBBEAN Standard Disclaimer: This volume is a product of the staff of the International Bank for Reconstruction and Development/ The World Bank. The findings, interpretations, and conclusions expressed in this paper do not necessarily reflect the views of the Executive Directors of The World Bank or the governments they represent. The World Bank does not guarantee the accuracy of the data included in this work. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgment on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Copyright Statement: The material in this publication is copyrighted. Copying and/or transmitting portions or all of this work without permission may be a violation of applicable law. The International Bank for Reconstruction and Development/ The World Bank encourages dissemination of its work and will normally grant permission to reproduce portions of the work promptly. For permission to photocopy or reprint any part of this work, please send a request with complete information to the Copyright Clearance Center, Inc., 222 Rosewood Drive, Danvers, MA 01923, USA, telephone 978-750-8400, fax 978-750- 4470, http://www.copyright.com/. All other queries on rights and licenses, including subsidiary rights, should be addressed to the Office of the Publisher, The World Bank, 1818 H Street NW, Washington, DC 20433, USA, fax 202-522-2422, e-mail: pubrights@worldbank.org. 1 *Authors/Acknowledgements: This study was financed by a grant from the Umbrella Facility for Gender Equality. The study was led by Victoria Stanley with a research team composed of Maira Emy Reimão, Barbara Coello, Sophie Theis and Marc Smitz. Guidance and review was provided to the team by Holger Kray and Martin Henry Lenihan. The team is also thankful to Jason McMann for his support with editing and finalizing the paper. The team also wishes to thank Elizaveta Perova, Pablo Chacon and Khanti Consultants, Abla Safir, Gero Carletto, Maria Beatriz Orlando, Sanna Liisa Taivalmaa, and Katherine M. Scott. Comments received during the session of the Land and Poverty Conference 2015 and at the presentation organized by the Gender and Rural Development Thematic Group are gratefully acknowledged. The questionnaire constructed for this study was built upon the Women’s Empowerment in Agriculture Index Questionnaire (WEIA) developed by IFPRI. 2 Table of Contents 1. Study Overview ................................................................................................................................... 5 Five Findings of Interest ...................................................................................................................... 5 2. Introduction ........................................................................................................................................ 6 Methodology ....................................................................................................................................... 7 Background on Guatemala .................................................................................................................. 8 Migration and Women’s Agency ....................................................................................................... 10 3. Access to and Use of Endowments ................................................................................................... 10 Access to and Use of Endowments: Land ......................................................................................... 10 Access to and Use of Endowments: Labor ........................................................................................ 12 Access to and Use of Endowments: Knowledge ............................................................................... 13 4. Impacts on Women’s Agency ............................................................................................................ 13 Agricultural Agency Index ................................................................................................................. 16 Non-Agricultural Agency ................................................................................................................... 16 Soft Agency........................................................................................................................................ 17 5. Impacts on Household Welfare......................................................................................................... 19 Income: Amounts and Sources ......................................................................................................... 19 Household Food Security and Diversity ............................................................................................ 21 6. Conclusions ....................................................................................................................................... 22 References .................................................................................................................................................. 23 Appendix A: Data Tables. ............................................................................................................................ 25 Appendix B: Additional Agency Information and Results. .......................................................................... 27 Measuring Agency ............................................................................................................................. 27 Further Agency Results ..................................................................................................................... 28 Appendix C: Explanation of Variables. ........................................................................................................ 36 Agency Variables ............................................................................................................................... 36 Soft Agency Variables ................................................................................................................... 36 Non-Agricultural Agency Variables .............................................................................................. 36 Agricultural Agency Variables ...................................................................................................... 37 Agency Indices .............................................................................................................................. 38 3 Other Variables of Interest ........................................................................................................... 38 Appendix D: Regressions of Interest. .......................................................................................................... 40 Tables Table 1. Analytical Framework...................................................................................................................... 7 Table 2. Agency Index Summary. ................................................................................................................ 14 Figures Figure 1. Year of Departure for Current Migration. ...................................................................................... 9 Figure 2. Average Agricultural Earnings per Hectare, by Household Type (US$ per Ha)............................ 11 Figure 3. Distribution of Agency Indexes by Household Type. ................................................................... 15 Figure 4. Soft Agency Index......................................................................................................................... 18 Figure 5. Distribution of Autonomy Self-Rating (Left Panel) and Difference in Rating (Right Panel). ........ 19 Figure 6. Income Source Distribution by Household Type, Yearly USD. ..................................................... 20 Figure 7. Distribution of Food Insecurity (Left Panel) and Food Diversity (Right Panel). ........................... 21 Maps Map 1. Departments of Guatemala. ............................................................................................................. 8 Boxes Box 1. What is Agency? ............................................................................................................................... 15 4 WOMEN IN AGRICULTURE The Impact of Male Out-Migration on Women’s Agency, Household Welfare and Agricultural Productivity1 1. Study Overview Migration is transforming rural economies, landscapes, and, potentially, gender relations. Migration is one of the drivers of the so-called “feminization of agriculture� in Latin America (Deere and León 2001). This feminization has relevance for everyone given agriculture’s role in regional food security, national shared prosperity and household resilience to shocks. However, this phenomenon is little studied and not well understood and new evidence is needed. To fill this gap, financing was secured from the Umbrella Facility for Gender Equality and household surveys were conducted in two departments in Southeastern Guatemala. The objective of this study is to investigate the “feminization� of agriculture as well as its implications for women’s agency, household welfare and agricultural productivity. We are fundamentally interested in what forms of engagement a woman wants to take and is able to take in agriculture, the household, and the community in the absence of her male partner. In particular, this analysis seeks to understand how male out-migration is influencing women’s agency in agriculture; to understand if, when women are in control of the farms, it changes the types of decisions they make and thus the results that they obtain; and finally, to get a better sense of how these differences in agency (if any) lead to better or worse outcomes for the farm household. Five Findings of Interest Contrary to the popular belief held by local officials, policymakers, and researchers alike, the vast majority of households remain in agriculture even as the male head of household migrates. The continuation of agriculture as a household livelihood strategy is characterized by the transformation and expansion of the role of married women in agricultural production. As men in southeastern Guatemala now migrate for years at a time, their partners face greater responsibilities in agricultural production, both in decision-making and in production itself. These households are more likely than the other types of households to employ non-household members or paid workers for agricultural labor. This gap persists even when controlling for the dependency ratio2 and household size. 1 Unless otherwise noted, all tables and figures appearing in this study derive from the original survey research described herein. 2 The dependency ratio refers to the ratio of the number of household members under age 15 and over age 64 to the number of working-age household members. 5 As agriculture is still seen as a traditionally male endeavor, women reported having to not only take on, but also learn how to farm once their husbands migrated. Extension services and technical assistance generally fail to reach women in rural areas. However, households with a male partner having migrated have the highest levels of food security and food diversity relative to other groups. Given the higher level of remittances received by these households and the fact that they tend to go directly to women, this result is in line with literature showing that money controlled by women is allocated at greater rates towards family nutrition than money controlled by men (e.g. Thomas 1990). 2. Introduction This section provides some introduction to women in agriculture, lays out the study methodology and provides background information on migration, women and agriculture in Guatemala. Women have a central role at the nexus of rural development, food security, and agriculture.3 According to the Food and Agriculture Organization (FAO), if rural women in developing countries had the same access to productive resources as men, they could increase yields on their farms by 20 –30%. This could raise total agricultural output in developing countries by 2.5–4%, which could in turn reduce the number of hungry people in the world by 12–17% (FAO2011b, IFPRI 2003). According to a recent World Bank (2014) study, this access to inputs has to include access to labor, technology and knowledge and may need to be tailored for women farmers. Women’s role in agriculture is even more crucial in Guatemala, which suffers from the double burden of chronic malnutrition and obesity. The country has a competitive agro-food sector, while at the same time the rate of chronic malnutrition in rural areas is one of the highest in the world. The agriculture sector represent 11% of GDP, with food exports representing more than 44% of total exports.4 While, according to the World Food Programme, Guatemala’s chronic under-nutrition rate is currently at 49.8% among children under five.5 Very little data has been collected at the microeconomic level to analyze the impact on women left alone on farms after their partners’ migration (FAO 2011, World Bank 2012a, etc., among many others). Women seem to be large but statistically invisible contributors to rural life through paid and unpaid employment. According to FAOSTAT (2013), women in Guatemala represent almost 10% of the labor force in agriculture. On the other hand the International Labor Organization reports that 12.6 % of female employment in Guatemala was in the agriculture sector.6 3 FAO (2011B) and World Bank (2011). 4 World Bank (2015). 5 World Food Programme (2015). 6 Data come from the World Bank’s (2015) World Development Indicators (“ Key Indicators of the Labour Market Database�). 6 Methodology This analysis seeks to investigate the impact of male migration on agriculture, but also its implications for women’s agency and agricultural productivity, as mediated by factors such as land tenure and access to agricultural extension services. In particular, this analysis seeks to better understand how male out-migration is influencing women’s agency in agriculture; to understand if, when women are in control of their farms, it changes the types of decisions they make and thus the results that they obtain; and finally, to get a better sense of how these differences in agency (if any) lead to better or worse livelihood outcomes for the farm household. Table 1 below outlines the framework used for the analysis, based on the analytical framework used for the 2012 World Development Report (WDR) on Gender and Development. The WDR recognized the importance of access to, and use of, endowments, such as land, labor and knowledge; but also raised the profile of women’s agency as a key to economic development. Finally, we add an analysis of the data around household income and food security to understand the impact of migration and women’s agency on household well-being. Table 1. Analytical Framework. Access to/Use of Endowments Women’s Agency Livelihood Outcomes – Land – Soft Agency – Food Security – Labor – Agricultural – Income – Knowledge – Non-Agricultural This study is based on a quantitative field survey conducted in August 2014, as well as qualitative focus groups and interviews conducted in May 2014 to test the questionnaire. The work was performed in two southeastern departments of Guatemala, Jutiapa and Chiquimula (see Map 1), on a sample of 572 agricultural households.7 The sampling process ensures that the results presented here are representative of these two departments near the border with Honduras and El Salvador. 7 See Appendix A for additional details. 7 Map 1. Departments of Guatemala. The households interviewed are classified into three groups:  Type 1: Women whose male partners are currently migrants.  Type 2: Women in households where both the male and female heads are present.  Type 3: Single women-headed households. We compare three groups of rural female heads of households: those with migrant partners, those with partners present (independent of possible migration history), and single female heads of households. The use of three groups (one treatment and two control groups) allows us to parse out the effects of having a migrant partner from being a single head of household and gives us the opportunity to see general social norms across households in a given community. For simplicity, from this point forward, we refer to women/households by their “type,� as classified above. Background on Guatemala In Central America and particularly in Guatemala, male out-migration is accelerating; more than 70% of migrants are young males and almost 90% of these migrants are in the United States. (Cohn and al. 2013). This report draws on data collected in Chiquimula and Jutiapa. As this region did not suffer as much displacement due to the civil war, we see little evidence of migration in previous generations (only six of the women interviewed said that either one of their grandparents or their spouse’s grandparents had ever lived abroad). Nonetheless, as many as 10% of women said that their father had lived in the United States, and those women are more likely to be currently married to migrant husbands. 8 These days, migration episodes tend to occur only a few times in a person’s life, though they are relatively long. While only 16% of partners in dual-headed households ever lived in the US, those who did spent on average 50 months away. Our sample also confirms that out-migration is largely a male phenomenon in rural eastern Guatemala, as over the last ten years only 15% of women with migrant husbands have lived outside the locality in which they were interviewed, with most living elsewhere in the country rather than abroad. Figure 1. Year of Departure for Current Migration. Notes: Figure based on data from Type 1 households only. The decision to migrate appears to fall mostly within men’s domain. In speaking of their partner’s most recent or current migration episode, 81% of women with migrant partners and 77% of women currently in dual-headed household said that the decision to migrate was made by their partner alone. Only 15% and 18%, respectively, said that their partner’s migration was a joint decision. In southeastern Guatemala, agriculture is a traditionally male endeavor: though women participate in several areas of the production process, men are the primary decision-makers. In 85% of Type 2 households in our sample, for instance, women do not participate in the decision of what to plant. Similarly, 88% do not take part in deciding what inputs to use. Nevertheless, about half of Type 2 women participate in some part of the agricultural production process, with 27% purchasing inputs and 30% taking part of the crop harvest. Further, qualitative interviews revealed that women play a critical supporting role on a daily basis as well: as some of the land used is hours away from the house by foot, men may spend the day there, while women walk back and forth to bring food and supplies as needed. The vast majority of households remain in agriculture even as the male head of household migrates, contrary to the popular belief held by local officials, policymakers, and researchers alike. The household’s persistence in agriculture has been defined by the transformation and expansion of the role of married women in agricultural production. As men in southeastern Guatemala now migrate for years at a time, their partners face greater responsibilities in agricultural production, both in decision-making and in production itself. In contrast to Type 2 households, half of women in Type 1 households participate in the decision of what to plant and what inputs to use (and the majority of these make the 9 decision alone). Even more dramatically, 73% of women in Type 2 households actually participate in some part of agricultural production. 60% purchase inputs, 50% harvest, 42% said they participate in planting, and 44% participate in cleaning the land. Migration and Women’s Agency The literature on the effects of male migration on women’s agency and empowerment reveals a mixed picture (Menjivar and Agadjanian 2007). First, agency is variable depending on the domain; undoubtedly, migration of a male partner does not increase agency across all domains – nor does it decrease agency across the board, either. In some cases, women see an expansion of their traditional roles. Some studies found that out-migration increases women’s participation in the labor force, often even in traditionally masculine activities (Mummert 1988). Often the increase in responsibility is not by choice but out of necessity, when remittances are insufficient or erratic (Pessar, 2005). These new roles may represent an excessive time burden with the loss of male labor, or represent obligations that are not always accompanied by social approval or access to the same economic support systems. Other researchers have found that traditional gender divisions of labor can be reinforced by male migration. Given few labor opportunities outside of agriculture, they found it was rare for women to join the labor force in agriculture or otherwise. Pessar (2005) also notes that there are instances in which “women (commonly from more economically secure households) are forbidden by migrant husbands to work outside the home.� However, migration also has the potential to change social norms within a community that prescribe how women participate in agriculture, community groups, household decision-making, and so on. The experience of heading the farm and household in the absence of her partner may earn a woman more trust and authority from her partner, peers, and community – and possibly increase her own sense of self-efficacy, or the internal component of agency. Certainly, bargaining within the household is affected by structural conditions and institutions in which the household is embedded (Agarwal). Furthermore, women in communities with high levels of out-migration, even if they themselves do not have a partner that has migrated, may all experience changes in gender roles over time. 3. Access to and Use of Endowments In the first section we look at the access to, and use of, endowments such as land, labor and knowledge and the differences between household types. Here we find that women’s access to both labor and knowledge impact their ability to farm all of the land they own or have access to. Interestingly, women’s access to land does not appear as the primary constraint and women are just as likely to have documents to their land as men. Access to and Use of Endowments: Land Land productivity is similar across all three groups. Overall, Type 2 households have farming incomes that are almost twice as large as those for households in the other two types: US$ 776 for type 2 10 households, in contrast to US$ 415 and US$ 365 for Type 1 and Type 3 households, respectively. However, the former also use more land than the other two groups. When considering agricultural income per hectare, most of this difference disappears. Type 2 households produce on average US$ 1,527 per year per hectare (including household consumption) but Type 1 households are not far behind, at an annual rate of US$ 1,435 per hectare. Type 2 households are 16% more likely to sell at least some agricultural production. The lack of economies of scale for Type 2 households may be a signal of labor and input constraints in rural Guatemala.8 Figure 2. Average Agricultural Earnings per Hectare, by Household Type (US$ per Ha). 1800 1600 1400 1200 1000 800 600 400 200 0 Type 1 Type 2 Type 3 Farming Farming + Animal Husbandry Notes: Income reported here represents annual income. For Type 1 households, with one fewer male member and a shift of the purpose of agriculture towards subsistence or consumption smoothing during periods of lower remittances, agricultural production is lower than for Type 2 households. This is a function of land use, not productivity. Type 2 households tend to use more agricultural land than the other two groups. Nonetheless, there is no statistically significant difference in the likelihood of owning agricultural land across groups (see Appendix A, Table A2). Two in five households in the sample own agricultural land (three in four own some land, including their household plot), though in all three groups it is common to use both land owned and land from others. Very few plots of land (less than 10%) are rented for money or used as share-cropping. Though women participate in agricultural production, their levels of land ownership are relatively low and hold evidence of male-preference in inheritance. Over half of the land owned by households in the sample was acquired through inheritance, and it is much more likely to come through the male lineage 8 Note that land quality is not taken into account. 11 than the female. Furthermore, in only 58% of cases in which a plot was inherited through the woman’s side of the family (from her parents or relatives) is she listed as an owner of that plot.9 There are other important differences in women’s ownership across groups, with more than 30% of women from Type 1 households owning at least one plot (jointly or as sole owner), in contrast to 21% of women from Type 2 households, and the difference is statistically significant. Furthermore, 20% of women from Type 1 households are the sole owner of at least one plot of land, in contrast to 13% of women in Type 2 households. Importantly, 80% of agricultural land owned by households in the sample has documentation (51% have a deed; 32% are also registered). No differences were found in the likelihood of documentation between female-owned and non-female-owned land, indicating that in this context the documentation process is no more inclusive of women than men. There are also no differences in the likelihood of documentation across household groups. Access to and Use of Endowments: Labor Type 2 households are less likely than the other two household types to employ non-household members or paid workers for agricultural labor. This gap persists even when controlling for the dependency ratio and household size, suggesting that in Type 2 households the male head of household may be undertaking a significant portion of the agricultural tasks that cannot be easily replaced by female work. Instead, Type 1 and Type 3 households rely on outside help (paid or unpaid) to replace this source of labor. During the qualitative interviews, women explained that one of the reasons they cannot cultivate all their land was the lack of available labor. During the qualitative work, women also explained other difficulties with hiring laborers – weak negotiating power, inability to monitor the quality of work, and women not being considered as a “real� farmer. Most women cope with these constraints by asking for help from another male in the household or community to manage the hiring and supervision of the workers. Households that can employ outside workers have a significantly higher agricultural income. This factor, after the total amount of land is incorporated, is the second most important factor in the explanation of agricultural income. This seems to reinforce the idea that households in rural areas will have a higher income – between US$ 160-200 – when they are able to hire an external worker to help them accomplish some of the agricultural tasks. The possibility to hire outside workers was consistently mentioned by women across all types during the qualitative interviews, and seems to be a very important constraint for women in Type 1 households. Women in Type 1 households have one less adult than in Type 2 households, but also a 9 In the survey, women were asked to list the owners of each plot of land used or owned by the household. As such, it favors “perceived� ownership over legal ownership. Nonetheles s, the results indicate that in 42% of cases in which the plot of land was inherited from the woman’s side of the family, the respondent does not consider the land to be hers. 12 higher dependency ratio. The latter also explains why while Type 2 households tend to use more agricultural land than the other two household groups, there is no statistically significant difference in the likelihood of owning agricultural land across groups. Access to and Use of Endowments: Knowledge As agriculture is a traditionally male endeavor, women reported having to not only take on but also learn how to farm once their husbands migrated. In focus groups, several women said they did not know how to farm when their husbands decided to migrate, learning just before they left or, once their partners left, from male relatives or from their partners over the phone. Male relatives are an important source of information and advice for women on agriculture. In our quantitative study, we find that the majority of women who know to farm first learned from their fathers (70%). Partners are also a principal teaching source, especially for women with migrant partners: 24% of Type 1 women first learned how to farm from their partners, in contrast to 18% of Type 2 women. Extension services and technical assistance, however, generally fail to reach women in rural areas. Only 13 women in the entire dataset said they received technical assistance in the last 12 months. Two- thirds of women noted that they do not currently learn about agriculture from anyone, including extension services as well as neighbors, parents, etc. About 25% of women learn from family members or neighbors. The scarcity of extension services is corroborated by the fact that there are only six extension agents serving the entirety of the two departments in our study – with only three agents per department – two generally serving male groups and one serving women. The focus groups and consultations revealed that, with a few exceptions, the extension services offered to women focus on nutrition and food preparation. These efforts stand in stark contrast with women's preferences and the role they play in agriculture, as seven in ten women in the sample stated that they would like to receive extension services or training in agricultural production. The highest demand is for training on selecting seeds (42% of all women in the sample), animal immunizations (41%), and pest control (36%). The lack of technical assistance for women in agriculture is alarming, as households in which women reported that they do not learn about agriculture from anyone have lower agricultural and total incomes relative to other groups. Specifically, households in which women reported that they learned how to farm alone have agricultural incomes that are US$ 371 lower than those of other households; controlling for other variables such as household type and size, this difference decreases to US$ 240- 260, but is still significant. Similarly, total household income is US$ 1,084-1,216 lower for those where women learned how to farm on their own, even when controlling for covariates. This result highlights the high cost of the lack of extension services, borne not only by women and their households but also by the agricultural sector as a whole. 4. Impacts on Women’s Agency The second realm of impact of male out-migration in Guatemala studied here is women’s agency. When men leave their farms to migrate internationally, to what extent, and in what way, are women 13 able to exercise their agency? Women living in the context of male out-migration represent an opportunity to empirically test two important research questions that remain to be answered to better understand agency: i) how women’s self-evaluated agency and agency outcomes are associated; and ii) how, potentially, agency in different domains relate to one another. As seen above, male out-migration has transformed agricultural roles, increasing women’s participation in agricultural production. The large influx of remittances also changes the distribution of money over which women have control, or are at least responsible for managing. Together, these shifts may influence women’s choices, self-perceptions, sense of empowerment, and ability to act. In this section we specifically look at women’s agency in agricultural decision-making, non-agricultural decision-making, and �soft agency� (see Box 1 below). What emerges from this section is that women in Type 1 households tend to have more agency – both agricultural and non- agricultural – thank Type 2 households, meaning that they are more involved in decision-making for both the farm and the household. However when looking at soft agency measures, women in Type 1 households do not necessarily see themselves as freer, or more autonomous, than other household types. Table 2. Agency Index Summary. Agency Index Component Variables Self-Determination/ Soft Autonomy Self-esteem/Aspiration Self-perception Decision in the Participation in the Non-Agricultural Access to financial services household community Agriculture Agricultural decisions Agriculture actions Agriculture ownership The distribution of agency measurements varies by household type and by the agency measurement used. Nonetheless, in all four agency indices, women in Type 3 households have, on average, a higher level of agency, as shown in the distributions below. Women in Type 1 households have a higher agency level in agriculture and non-agricultural dimensions of agency relative to those in Type 2 households, but the distribution of the soft agency measurement is similar for all three groups. With respect to the overall agency measurement, Type 2 women have the lowest level of agency, followed by Type 1 and then Type 3 women. 14 Figure 3. Distribution of Agency Indexes by Household Type. Box 1. What is Agency? This study draws upon the concept of agency to enrich our understanding of female empowerment in agriculture. The questionnaire used in this study built upon the Women’s Empowerment in Agriculture Index (WEAI) (Alkire et al. 2013), and was designed specifically to focus on trends in the feminization of agriculture due to male out-migration. Agency, then, is a quality or capacity exercised when a person is able to capitalize on endowments and economic opportunities to lead to desired actions. Amartya Sen (1989) defines agency as an individual’s ability to act on behalf of what the individual values and has reason to value. Agency is not “global� but rather multidimensional in the sense that an individual can e xercise different levels of agency in pursuit of multiple aims (Alkire 2005, 2007). These aims can be very diverse and someone may have variation in her level of agency with respect to different aims (Alkire 2008). A woman may have significant say in decision-making over what kinds of food to buy, but not have control over the amount of income she is allocated by her husband out of their earnings. Consequently, the measure of agency used captures three different dimensions of rural women’s agency related to agricultural and non-agricultural variables: (i) soft agency, (ii) non-agricultural agency, and (iii) agricultural agency. Each of these measurements is built using three components, as described in the table below. A description of each variable is included in Appendix C, along with a brief explanation of the methodology used to construct the agency measurements (principal components score). This study also includes a broader agency index, which combines all nine variables into a single index. One important part of survey instrument was the deeper exploration of “soft agency.� Based on 15 qualitative work, the survey questionnaire was developed with several modified psychosocial scales to measure women’s self-determination and self-esteem. This section included questions on women’s self- perception on specific qualities of agency in contrast to other women like her and perception of cultural norms to contextualize what kinds of choice and behavior is perceived as possible in a community. This soft agency section serves to test some of the psychosocial scales in a new context amongst rural women and to test links between soft agency and other factors of agency, like decision-making and access to endowments and economic opportunities. Agricultural Agency Index Women in Type 1 and Type 3 households have significantly more decision-making agency in agriculture. Women in Type 1 households are also more likely to participate in agricultural decision- making, with 64% of women in Type 1 relative to 20% of women in Type 2 households participating in the decision of what crops to plant. Only two percent of women in Type 2 households report being the sole decision-maker on that issue, while 50% of women in the other two groups report being the sole decision-maker. Even when several dimensions are combined, including how women participate in the decision on what to plant, the decision on inputs, and more generally on agricultural production, the results show a similar trend of women in Type 2 households participating less in these decisions. Among households with small animals, the woman tends to be responsible for them, though at a lower rate among Type 3 households. The latter might be due to her responsibilities for everything else. The survey included questions about large animals; however, very few households own them (16%), and when they do, women are usually not responsible for them. Thus, the index does not cover this dimension of animal ownership. Non-Agricultural Agency Our non-agricultural agency index is comprised of three dimensions: the distribution of household decision-making, participation in the community, and access to financial services. Household decision- making is comprised of various realms, some of which may be traditionally within women’s domains (e.g., food) and others that are not (e.g., the household’s overall budget). We also consider the extent to which women participate in local groups, both as a member and in leadership positions, as well as their access to banking. The roles played by individuals in decisions vary by household type, and married women are less likely to make decisions regarding their own time and employment. Among women who stated they were not employed, for example, 45% of those in Type 1 households and 35% of those in Type 2 households stated that their partners were the ones who decided that the women would not work outside the home. In contrast, 61% of single/widowed women who do not work made that decision themselves. The latter group is also much more likely to decide alone on any other activities they do outside the household: 89% of them decide on which activities to participate in, in contrast to 47% of women from Type 1 households and 34% of women from Type 2 households. Notably, however, over half of the 16 women who said they played no role in deciding on their activities outside the house also stated that they do not wish they had more decision-making power either. Women in Type 1 and Type 3 households have a greater say in the household budget than do women in Type 2 households. As many as 57% of women in Type 1 households and 77% of women in Type 3 households say they decide and manage the household budget alone, while only 13% of women in Type 2 households do so. Another 36% percent of women in Type 1 households share this responsibility, but 30% of women in Type 2 households have no say in the household budget. This pattern also holds true for household food decisions: in 39% of Type 2 households the male partner decides alone on how much to spend on food, while 75% and 83% of women are the sole decision-makers in Type 1 and Type 3 households, respectively. The participation of women in any type of productive group, other than church and sports activities, is very low, at around 22%. These results are somewhat surprising, particularly considering the extremely low stated participation in productive groups (less than 10%). It is possible, however, that the question on “belonging to a group� may not have been well understood or interpreted by the respondent as envisioned in the survey design, as the qualitative work in several communities showed higher levels of women participation in groups organized by local NGOs. Very few women have a leadership position in their community. As expected, Type 3 women are on average slightly more likely to be leaders (23%) than Type 1 and Type 2 women, at 18% and 17%, respectively. The use of credit and insurance is low in this region of Guatemala. Less than 10% of households in the sample have any credit and fewer than seven percent have formal credit (with a bank or NGO). Around seven percent of households also carry some form of life insurance. However, 33% of women in Type 1 households have an independent bank account. This is significantly more than women in Type 2 households (11%) and women in Type 3 households (14%). Women in Type 1 households might enjoy a secondary effect due to their higher familiarity with financial institutions provided by the necessary management and receiving of remittances. In fact, receiving remittances increases the likelihood that a woman has a bank account by nine percentage points. Having a bank account is associated with higher incomes. Households in which women have a bank account alone have earnings that are US$ 898 more than households in which they don’t; this amount reaches US$ 1,023 for Type 1 households. Given the overall low rates of credit use in this context, only access to a bank account in used as a proxy of for access to financial services. We note that the level of agency measured for agricultural and non-agricultural decisions is higher for the women in Type 1 households than women in Type 2 households. Soft Agency The soft agency measurement designed through the survey is comprised of three variables. Specifically, it considers self-efficacy (sense of freedom and choice), aspirations (abilities and goals), and 17 autonomy. Figure 4 shows average scores in each of the three dimensions by household type. More detail on the questions used to elicit these psychosocial measurements is included in Appendix C. Figure 4. Soft Agency Index. 10 8 6 4 2 0 Type 1 Type 2 Type 3 Self-Efficacy Aspirations Autonomy A greater share of Type 3 women relative to women in other groups perceive themselves as very autonomous. In the “autonomy� question, women were asked to position themselves on a ladder with ten rungs, with the first rung representing someone without any freedom and the top rung (i.e. the tenth) representing someone who is completely free. Figure 5 shows that women in Type 3 households tend to perceive themselves as more free than others. It is also interesting to note that women in Type 1 households are more likely to give themselves the same rating of freedom as they assign to the rest of the women in their community. A follow-up to the autonomy question asked women to state the rung in which they thought most of the women in their community would be. Figure 6 shows the distribution of the difference between the woman’s own rung and the rung she assigned for women in her community, so that zero indicates she placed both of them on the same rung; a positive number indicates that the woman thinks she has more freedom than the rest of the women in her community while a negative number indicates less freedom. The survey finds that women in Type 3 households are more likely to not only to place themselves highly, but they also consider themselves to be freer relative to the rest of women in the community. The high concentration of Type 1 women at zero is inconsistent with higher levels of agency in agriculture and non-agricultural measures, and raises the possibility that their responses to the autonomy question are biased in an attempt to “fit in.� 18 Figure 5. Distribution of Autonomy Self-Rating (Left Panel) and Difference in Rating (Right Panel). Notes: The difference reported in the right panel represents respondents’ self -rating minus their rating of other women. 5. Impacts on Household Welfare In the context of the high levels of malnutrition found in Guatemala, two of the principal measurements of family welfare are household food security and diversity. This section explores the differences in income sources across the three groups, and the differences in food security and food diversity between them. Type 1 households have the highest levels of food security and food diversity compared to the other household types. Income: Amounts and Sources Contradicting the common belief in Guatemala, migrant households are not richer than the rest. Some of the new social programs that were being designed at the time of fieldwork excluded migrant households, assuming that they were always better off than other types of households, given that they had a supplementary income source in the form of remittances. Instead, one finds that households in the three groups have, on average, the same amount of total income (see Figure 6). The average annual income for Type 1 households is US$ 2,715; for Type 2 households, US$ 2,769; and for Type 3 households, US$ 2,437.10 10 Migrants earn more income than what they remit, and we do not take the total amount into account in computing “household income,� including only the amount received in remittances. (It was not possible to collect data on migrants’ total earnings, as interviews were carried out with their spouses, many of whom may not know or want to report their partner’s earnings abroad.) In this sense, we consider the “household� for economic purposes as the family members and other individuals living in the same house and sharing meals, with remittances as an additional source of income. 19 Figure 6. Income Source Distribution by Household Type, Yearly USD. Not surprisingly, women from Type 1 households have a higher share of income from remittances. On average, Type 1 households receive US$ 1,659 in remittances per year, in contrast to US$ 223 for Type 2 households and US$ 404 for Type 3 households. While total income across the three groups varies little, the composition differs, as Type 1 households use remittances to make up for losses in agricultural and wage income. Notably, however, among households in the latter two groups that do receive remittances, the transfers are also fairly large: on average, Type 2 remittance-recipient households receive US$ 1,023 per year; for Type 3 households, the average amount is US$ 1,158. Nonetheless, these are around half the amount received by Type 1 households that receive remittances (79%), for which the average annual amount is US$ 2,192. Interestingly, there is no difference across households in women's participation rate in deciding what to do with the remittances. Type 2 households that receive remittances are 21.6 percentage points less likely than Type 1 households to use any of the remittances for food. Type 3 households are nine percentage points less likely to do so. There is no difference in likelihood of spending remittances on education, even when accounting for the number of children. Type 1 and 2 households are just as likely to use remittances for agriculture (13-15%), but Type 3 households are less likely to do so. Type 2 households are also more likely to be engaged in wage/salaried work. 67% of Type 2 households have income from non-agricultural work, along with 55% of Type 3 households but only 26% of Type 1 households. Government transfers represent a very small amount of total income (i.e. other income) . 30% of households receive government cash transfers, but, as corroborated by the qualitative interviews conducted and the explicit exclusion of households with migrants from social programs, there are much 20 lower rates of transfers for Type 1 households (16%) compared to Type 2 households (32%) and Type 3 households (42%). The amount of the transfer is quite small, however, so Type 1 households receive on average US$ 13 per year, compared to US$ 30 among Type 2 households, and US$ 49 among Type 3 households. The most common type of in-kind transfer in rural Guatemala is fertilizer, on average 45% of Type 1 households receive it, compared to 65% of Type 2 households. Household Food Security and Diversity Type 1 households have the highest levels of food security and food diversity compared to the other groups, as indicated in Figure 7. Given the higher level of remittances received by these households and the fact that they tend to go directly to women, this result is in line with literature showing that money controlled by women is allocated at greater rates towards family nutrition than money controlled by men (e.g. Thomas 1990). A surprising and perhaps alarming result, however, is that Type 3 households (women-headed households) have the most precarious nutritional status, particularly with respect to their levels of food insecurity. Figure 7. Distribution of Food Insecurity (Left Panel) and Food Diversity (Right Panel). Notes: In the left panel, higher values on the x-axis indicate more food insecurity. In the right panel, higher values on the x-axis indicate more food diversity. Households with a higher share of agricultural income to total income are slightly less likely to be food insecure but also less likely to have food diversity. That is, while agricultural production stabilizes access to food, so that households are less likely to go days without eating or with little food, for instance, they are also less likely to experience diversity in their food, as they rely on their own production for food and that production is limited in diversity. Households that rely on remittances or other sources of income may buy a wider range of foods. As expected, higher income is correlated with lower food insecurity and higher food diversity. Households that receive remittances have higher levels of food diversity, though not necessarily food security. For households that receive remittances, the amount of remittances has a small but significantly positive effect on food security and diversity. Interestingly, and perhaps contrary to the literature on female allocation of resources, we do not find evidence that if women’s participation the 21 decision of how the remittances are allocated affects food security or diversity. This may be due to sample size limitations, or may be attributed to the fact that the majority of households (79%) allocate some of their remittances towards food anyway, regardless of whether the woman participates in the decision-making process. 6. Conclusions The research yielded important findings for policymakers, researchers and others interested in the impact of male out-migration on the agriculture sector and on the women and families they leave behind. Contrary to popular belief, the vast majority of households remain in agriculture after the migration of the male head of household. However, they tend to shift the purpose of agriculture towards subsistence and consumption smoothing during periods of lower remittances. When males out-migrate, women report having more agricultural agency and become more involved in agricultural and household decision-making. However, improved household welfare reported among migrant households arise primarily due to remittance flows and decisions about income allocation, rather than improvements in productivity. At the same time, these women may not see themselves as freer or may feel burdened by the need to make more decisions alone. While land productivity is similar across all three groups of households, farming income varied across households, with households in which a male head was present reporting the highest farm income. When considering agricultural income per hectare, most of this productivity difference disappears. The lack of economies of scale for migrant households may be a signal of labor, input and knowledge constraints in rural Guatemala. The lower farm income reported by these agricultural households appears to have less to do with decision-making and more to do with the high informational and labor barriers that women face. While women may wish to stay in agriculture, their lack of knowledge and access to labor and other inputs hampers them from becoming more productive. Diversifying risk in the household by diversifying agricultural production is an important factor of higher agricultural income. Remittances should not impact on households’ access to social transfers, as the remittances do not contribute to higher overall family income. Food security and food diversity could be achieved at a faster pace if women not only had more economic empowerment but also more ‘soft’ agency. 22 References Alkire, S. 2008. Concepts and Measures of Agency. Alkire, S., Meinzen-Dick R, Peterman A., Quisumbing A., Seymour G., and Vaz A.. 2013. The Women’s Empowerment in Agriculture Index. World Development. Agarwal 1997. Bargaining and Gender Relations within and beyond the Household. Feminist Economics 3(1), 1997 Cohn D., Gonzalez-Barrera A., and Cuddington D., 2013. Remittances to Latin America Recover—But Not to Mexico. Hispanic Trends Project, Pew Research. De Schutter,. 2012. Women’s Rights and the Right to Food. United Nations General Assembly A/HRC/22/5. Deere, Carmen Diana and Magdalena León de Leal, 2001, Empowering Women: Land and Property Rights in Latin America, University of Pittsburgh Press. Food and Agriculture Organization of the United Nations (FAO). 1999. El Acceso de la Mujer Latinoamericana a la Tierra. Rome. ———. 2011a. The Role of Women in Agriculture. Agricultural Development Economics Division. ESA Working Paper 11-02. Rome. ———. 2011b. The State of Food and Agriculture: Women in Agriculture: Closing the Gender Gap for Development. Rome. ———. 2013. FAOSTAT Database, http://faostat.fao.org/. International Food Policy Research Institute (IFPRI). 2003. Household Decisions, Gender, and Development. Washington, DC. Katz, E. 2003. “The Changing Role of Women in the Rural Economies of Latin America.� International Food Policy Research Institute (IFPRI) Women’s Empowerment in Agriculture Index, http://www.ifpri.org/publication/womens-empowerment-agriculture-index. Davis (ed.), Food, Agriculture and Rural Development: Current and Emerging Issues for Economic Analysis and Policy Research. Vol. I: Latin America and the Caribbean. Rome: FAO. Lastarria-Cornhiel, S. 2008. The Feminization of Agriculture: Trends and Driving Forces. Background paper for World Development Report (2008). Menjivar and Agadjanian 2007. Men’s Migration and Women’s Lives: Views from Rural Armenia and Guatemala. Social Science Quarterly 88(5) Mummert, G. 1998. "Mujeres de migrantes y mujeres migrantes de Michoacán: Nuevos papeles para las se quedan y para las que se van." In Movimientos de población en el occidente de México. Eds. T. Calvo and G. López. El Colegio de Michoacán/CEMCA. Pessar, P. R. (2005). Women, Gender, and International Migration Across and Beyond the Americas: Inequalities and Limited Empowerment. Expert Group Meeting on International Migration and Development in Latin America and the Caribbean. Mexico City, Department of Economic and Social Affairs, United Nations Secretariat Sen, Amartya (2001). Development as Freedom. Oxford New York: Oxford University Press. Thomas, D. 1990. “Intra-household allocation: An Inferential Approach.� The Journal of Human Resources, 24(4):635-664. ———. 2008. World Development Report 2008: Agriculture for Development. Washington, DC. ———. 2011. World Development Report 2012: Gender Equality and Development. Washington, DC. ———. 2012a. Women’s Economic Empowerment in Latin America and the Caribbean. Washington, DC. 23 ——— 2014. Migration & Remittances Data. Recent Trends and Outlook: 2013-2016. Washington, DC. World Bank (2015). World Development Indicators Database. World Bank and the ONE Campaign. 2014. Levelling the Field – Improving Opportunities for Women Farmers in Africa. Washington, DC. World Food Programme (2015). “Guatemala: Country Page (Overview),� https://www.wfp.org/countries/guatemala/overview. 24 Appendix A: Data Tables. Table A1. Descriptive Statistics by Household Type. Household Type Migrant Husband Dual-Headed Single-Female-Headed 11 Marital Status of Woman (interviewee) Single (%) 0 0 23.40 Married (%) 64.85 75.56 0 Common-law married (%) 35.15 24.44 0 Divorced (%) 0 0 1.42 Separated (%) 0 0 21.99 Widowed (%) 0 0 53.19 Woman’s Age (mean) 35.4 40.4 43.8 Partner’s Age (mean) 38.6 44.8 - Woman’s Literacy Can read and write (%) 68.48 60.15 39.01 Can read or write with difficulty (%) 15.76 13.16 17.73 Not literate 15.76 26.69 43.26 Partner’s Literacy Can read and write (%) 86.06 69.17 - Can read or write with difficulty (%) 3.03 8.27 - Not literate 10.91 22.56 - Woman’s Schooling12 None/Less than primary (%) 11.52 23.31 36.88 Some primary (%) 51.52 43.98 43.97 Completed primary (%) 28.48 24.44 13.48 Secondary or more (%) 7.87 6.77 5.67 Partner’s Schooling None/Less than primary (%) 10.30 24.44 - Some primary (%) 36.36 39.47 - Completed primary (%) 36.97 26.32 - Secondary or more (%) 8.48 9.03 - Household Size 4.5 5.6 5.0 Number of Kids (age<=12) in Household 1.6 1.8 1.5 Number of Woman’s Children13 2.9 4.1 3.7 Dependency Ratio14 1.18 0.75 0.86 11 Note that the survey did not make a distinction between de jure and de facto marital status, and instead simply asked women to select an option as they felt fit. 12 May not add up to 100% because of “do not know� answers. The same applies to “Partner’s Schooling� below. 13 Includes all living children of all ages, whether living in the household or elsewhere. 14 The dependency ratio refers to the ratio of the number of household members under age 15 and over age 64 to the number of working-age household members. 25 Table A2. Descriptive Statistics on Agricultural Land. Household Type Single-Female- Migrant Husband Dual-Headed Headed Number of plots used, managed, 2.29 2.42 ** 2.04 *** or rented out by the household Average plot size (per plot), m2 4,716 7,060 *** 4,702 Total land (across all plots), m2 9,403 15,316 *** 8,603 Share of plots used that are 57.14% 46.10% *** 50.00% * owned by household members Households that own at least one 76.36% 74.06% 76.60% plot Total land owned by household 6,507 7,087 4,664 (across all plots), m2 Share of plots owned by woman 19.15% 11.27% *** 32.17% *** Women that own at least one 30.30% 21.80% ** 49.65% *** plot Total land owned by woman 4,334 5,451 4,939 (across all plots), m2 26 Appendix B: Additional Agency Information and Results. Measuring Agency It is worth measuring agency since it is an intrinsically valuable expression of freedom and choice, and a pathway to gender equality. Because of the complex nature of agency, however, approaches for measuring it are varied. In the literature we find three general proxies for measuring agency, though each has shortcomings: 1. Endowments: The amount or share of goods owned by the woman. This traditional method considers proxies such as the amount of land owned by the woman or the land received by the couple as a dowry. But agency is not only a matter of a woman’s endowments and economic opportunities, despite the influence they have on her capacity to exercise agency. Two individuals with the same endowments and economic opportunities do not necessarily have the same goals or equal ability to advance the goals they value and have reason to value (Alkire 2008). 2. Actions. The woman’s behavior, with assumptions about what one’s behavior might be if free to choose. Here, the proxies used include participation in the labor force or a having a lower number of kids. The difficulty with this measurement, however, is that agency is not equivalent to action and should not be measured by a list of actions that a third party deems as expressing agency. A woman who does not participate in agricultural labor, for example, may be exercising her agency in the decision not to work and divide her labor strategically with her spouse. 3. Decision-making responsibilities. The share or number of decisions pertaining to the household in which the woman participates or is the sole decision-maker. Proxies in this approach include whether the woman participates in/is the sole decision-maker in the household’s expenditure decisions, schooling decisions, etc. or, more comprehensively, is based on her share of participation in various decisions. Nonetheless, sole decision-making alone is not a perfect measurement of agency, as female heads of households and other women may actually prefer to share decision-making duties with another person. 4. Elicited psychosocial measurements. This method uses questions to elicit women’s perceptions of their own level of agency or a similar notion. A survey question asking women how they rate their level of freedom relative to other people in their household or village, for instance, can be a proxy for a woman’s degree of agency in this context. Similarly, women may be asked whether they think their opinions are heard or whether they feel capable to do what they set out to do. These newer measurements rely on the assumption that complex questions are adequately understood. However, individuals may have incorrect perceptions, such that a woman who says she has more freedom relative to another may not have so in reality; her low level of agency may have led her to expect and accept a lower level of freedom. 27 In this report, we take all four approaches into consideration, building indices that combine more than one of them. With this, we offer more holistic measurements of agency, and mitigate the shortcomings of each measurement by supplementing them with others. Reassuringly, however, we find positive correlations between all of the measurements used. Further Agency Results Using our calculated agency indices, we conducted some regressions to understand the relationship between agency and other outcomes of interest. Some of these results are included in Appendix D, but in this sub-section we highlight some interesting and/or policy-relevant findings. Perhaps not surprisingly – but reassuring for the validity of our measurements – higher levels in the soft agency index and in the non-agricultural agency index are associated with higher income. For Type 1 women, higher levels of agency and particularly aspiration levels are associated with higher incomes. When including each measurement of agency separately, the soft agency index is positively correlated with a better agricultural income. For Type 1 women, a higher autonomy rating is very strongly associated with higher agricultural earnings. In contrast, for Type 3 women, a higher autonomy rating may be associated with lower agricultural earnings. This might be explained by the fact that when they are more autonomous than the average, they may invest more effort in trying to increase other sources income, rather than just the agricultural income, and is corroborated by the fact that, among Type 3 women, greater levels of autonomy are associated with more time spent on productive activities other than agriculture. However a surprising result is the negative correlation existing between the agricultural agency index and total agricultural income. This might be a sign that women are experiencing significant less access to inputs, or that higher agricultural incomes are associated with larger production, with more labor and a lower participation level from women in both decisions and actions. Having a high level of soft agency is also correlated with higher level of food security and food diversity in the household. Higher autonomy seems to explain higher levels of food security, whereas higher self-determination is associated with higher levels of food diversity. This is again consistent with the literature on the greater tendency of women to allocate money towards food; women with higher levels of agency, who thus feel more capable to take control of and allocate resource, may be more successful at channeling income towards food expenditures for their household. 28 Figure B1. Soft Agency Index. 10 8 6 4 2 0 Type 1 Type 2 Type 3 Self-Efficacy Aspirations Autonomy Figure B2. Decide on Activities Outside the House. 100% 80% 60% 40% 20% 0% Type 1 Type 2 ** Type 3 *** Someone else Together Alone Figure B3. Decide on HH Budget. 100% 80% 60% 40% 20% 0% Type 1 Type 2 *** Type 3 ** Someone else Together Alone 29 Figure B4. Decide on Food. 100% 80% 60% 40% 20% 0% Type 1 Type 2 *** Type 3 Someone else Together Alone Figure B5. Decide on Own Medical Attention. 60% 40% 20% 0% Type 1 Type 2 *** Type 3 * Someone else Together Alone Figure B6. Belong to a Social Group (Excluding Church and Sports). 100% 80% 60% 40% 20% 0% Type 1 Type 2 Type 3 Yes No 30 Figure B7. Woman has a Leadership Role in a Social Group. 100% 80% 60% 40% 20% 0% Type 1 Type 2 Type 3 Yes No Figure B8. Woman has a Bank Account Alone. 100% 80% 60% 40% 20% 0% Type 1 Type 2 *** Type 3 *** Yes No 31 Agricultural Decision-Making Figure B9. Woman Participates in Decisions on Plantation (Sembrar). 100% 80% 60% 40% 20% 0% Type 1 Type 2 *** Type 3 Yes No Figure B10. Woman Participates in Decisions on Inputs (Insumos). 100% 80% 60% 40% 20% 0% Type 1 Type 2 *** Type 3 Yes No Figure B11. Woman Participates in Decisions Regarding Agricultural Production. 80% 60% 40% 20% 0% Type 1 Type 2 *** Type 3 Yes No 32 Figure B12. Woman Responsible for Small Animals. 100% 80% 60% 40% 20% 0% Type 1 Type 2 Type 3 *** Yes No HH does not have small animals 33 Agricultural Actions Figure B13. Woman Participates in Cultivations. 100% 80% 60% 40% 20% 0% Type 1 Type 2 *** Type 3 Yes No Figure B14. Woman Participates in Agricultural Production. 100% 80% 60% 40% 20% 0% Type 1 Type 2 *** Type 3 Yes No Figure B15. Woman Was the Respondent for the Agriculture Module. 100% 80% 60% 40% 20% 0% Type 1 Type 2 *** Type 3 Yes No 34 Figure B16. Woman Owns Agricultural Land. 100% 80% 60% 40% 20% 0% Type 1 Type 2 Type 3 * Yes No 35 Appendix C: Explanation of Variables. Agency Variables Soft Agency Variables Self-efficacy: A composite of answers to two questions regarding self-efficacy. The first asks the respondent to choose from four sentences the one that most describes her situation (e.g., On one extreme, “I always feel free to do whatever I decide to do� and, on the other, “Almost always what I do is not what I would have chosen to do�.). The second question offers an alternative set of four sentences (e.g., “I always choose the way in which I do things�, and, at the other end of the spectrum, “I never choose for myself the way in which I do things.�). The answer to each of the two questions is given a score from 0 to 3, and these are added, for a self-efficacy score ranging from 0 (least self-efficacy) to 6 (most). Aspirations/ Self-esteem: A composite of answers to four questions regarding aspirations, all of which ask whether the respondent completely disagrees, disagrees, agrees, or completely agrees with the description of themselves. The statements are: “Sometimes I think I am not good at anything,� “I am capable of doing things just as well as most people,� “I generally do not dare share my ideas,� “I think I am capable of fulfilling some of my dreams.� The answer to each of the questions is given a score from 0 to 3, and these are added together, for an aspirations score ranging from 0 (fewest aspirations) to 12 (most). Autonomy: This variable comes directly from a question asking respondents to imagine a ladder with ten rungs, “where people with the least amount of freedom are at the bottom rung and people with the most freedom are at the top rung�, and state which rung the respondent believes she is on. This question was aided by a visual representation of a ladder, and the answers range from 0 (lowest rung) to 10 (highest). Non-Agricultural Agency Variables Participation in budget: Measure of women’s participation in two facets of the household budget: deciding on the overall budget and managing the budget. It is coded as done alone (2), together with someone else (1), or by someone else/ no participation (2). Participation in food expenditures: Measure of women’s participation in three facets of food expenditures: deciding on the overall amount allocated towards food, deciding on what food to buy, and making the purchase. It is coded as done alone (2), together with someone else (1), or by someone else/ no participation (2). Participation in decisions on own activities outside the household: Measure of women’s participation in deciding on the activities she carries out outside the household. It is coded as done alone (2), together with someone else (1), or by someone else/ no participation (2). 36 Participation in decisions regarding own health care: Measure of women’s participation in two facets of one’s own health care: when feeling ill, whether to get care and where to get care. It is coded as done alone (2), together with someone else (1), or by someone else/ no participation (2). Participation in non-agricultural decisions: A sum of the previous four variables, ranging from 0 to 8. Social participation: A composite of two indicators: whether the woman participates in a group in her community (excluding church and sports groups, due to high participation in the latter), and whether she holds a leadership position in any group (including church and sports groups). This variable ranges from 0 (no participation or leadership) to 2 (participation and leadership in at least one group). Woman has a bank account alone: Single variable based on answers to whether anyone in the household has a bank account, and who owns that account. Coded as “no� (0) or “yes� (1). Agricultural Agency Variables Participation in decisions on what to plant: Single variable based on the listing of household members who participate in the decision of what to plant for agricultural production. Coded as “no� if the woman is not listed among the participants (0) or “yes� otherwise (1). Participation in the decisions on inputs: Single variable based on the listing of household members who participate in the decision of which inputs to use in agricultural production. Coded as “no� if the woman is not listed among the participants (0) or “yes� otherwise (1). Participation in decisions on agricultural production: Women were asked whether they participate in the household’s agricultural production decisions or not. Coded as “no� (0) or “yes� (1). Responsible for small animals: Single variable based on the listing of household members who are responsible for the small animals owned by the household, by type of animal (rabbits, chicken, roosters, turkeys, and ducks). Coded as “no� if the woman is not responsible for any of the sm all animals owned by the household (0) or “yes� if she is responsible for at least one type of small animal (1). Participation in agriculture decisions: A sum of the previous four variables, ranging from 0 to 4. Participation in cultivations: Single variable based on the listing of household members who participate in crop cultivation. Coded as “no� if the woman is not listed among the participants (0) or “yes� otherwise (1). Participation in agricultural production: Women were asked whether they participate in the household’s agricultural production or not. Coded as no (0) or yes (1). Woman answered agriculture module: In the survey implementation, the default protocol was for our selected interviewees to answer all of the modules in the questionnaire. However, for the agriculture module, they were first asked if they believed they could answer a module on the land use and agricultural production of the household. If not, they could indicate a different respondent for that 37 module. In this variable, we note whether the woman was the respondent for the agriculture module (1) or not (0). Participation in agriculture actions: A sum of the previous three variables, ranging from 0 to 3. Woman owns agricultural land: Single variable based on the listing of each plot of land used or owned by the household, and the listing of individuals who own each plot. Coded as “no� if the woman is not listed as an owner for any of the plots listed (0) or “yes� if she owns at least one of the plots listed (1). Agency Indices Soft agency: Principal-component factor using the three variables listed in the soft agency variables category above, each rescaled to range from 0 to 1. Factor analysis considers the correlation between the included variables and creates a composite of them, giving weights according to the correlation matrix. It has a mean of 0 and standard deviation of 1. Non-agriculture agency: Principal-component factor using the last three variables listed in the non- agriculture agency variables category above, each rescaled to range from 0 to 1. Factor analysis considers the correlation between the included variables and creates a composite of them, giving weights according to the correlation matrix. It has a mean of 0 and standard deviation of 1. Agriculture agency: Principal-component factor using the variables “participation in agriculture decisions,� “participation in agriculture actions,� and “woman owns agriculture land� listed in the agriculture agency variables category above, each rescaled to range from 0 to 1. Factor analysis considers the correlation between the included variables and creates a composite of them, giving weights according to the correlation matrix. It has a mean of 0 and standard deviation of 1. Agency: Principal-component factor using all nine variables included in the three indices above, each rescaled to range from 0 to 1. Factor analysis considers the correlation between the included variables and creates a composite of them, giving weights according to the correlation matrix. It has a mean of 0 and standard deviation of 1. Other Variables of Interest Dependency ratio: The dependency ratio refers to the ratio of the number of household members under age 15 and over age 64 to the number of working-age household members. Food insecurity: Variable based on a standard set of nine questions measuring household food insecurity (for an example of the questions, see: http://www.unscn.org/layout/modules/resources/files/Household_food_insecurity_Sp.pdf, p. 6). The scores range from 0 (least food insecurity) to 27 (most food insecurity). Food diversity: Households were asked whether, over the previous 24 hours, anyone in the household consumed vegetables, fruits, meat (beef, chicken, or pork), fish/sea food, eggs, and milk or milk products. One point was given for each “yes� and zero for “no ,� such that this composite score ranges from 0 to 6. 38 Time spent on agriculture: Number of hours the respondent spends working in agriculture “on an average working day�. Time spent on other income-generating activities: Number of hours the respondent spends “on an average working day� on income-generating activities other than agriculture. 39 Appendix D: Regressions of Interest. Table D1. Determinants of Household Agricultural Income from Farming and Animal Husbandry, Annual US$. (1) (2) (3) (4) (5) Type 2 Household 125.5* 109.9 70.17 46.38 97.84 (62.70) (65.43) (63.73) (67.39) (63.50) Type 3 Household -48.95 -21.90 -28.14 -64.43 -16.50 (60.40) (59.85) (58.94) (60.24) (59.67) Soft Agency 56.66* 67.00* (28.34) (29.52) Non-Agriculture Agency -19.09 -16.18 (35.13) (36.14) Agriculture Agency -70.67* -75.55* (30.91) (31.26) Agency Index -33.61 (33.02) Land size (1000m2) 52.52*** 53.28*** 51.59*** 51.00*** 52.80*** (9.443) (9.400) (9.330) (9.509) (9.295) Land size (1000m2) squared -0.282** -0.288** -0.275** -0.272** -0.284** (0.104) (0.102) (0.0993) (0.101) (0.101) Woman learned alone how to farm -232.6* -252.7** -233.4* -200.9* -250.6* (96.67) (96.72) (99.32) (99.07) (97.85) Woman does not know how to farm 4.765 -13.62 -65.45 -57.22 -30.22 (115.7) (115.7) (119.3) (118.7) (119.1) Number of crops cultivated 92.88 87.51 101.2* 105.5* 91.31 (48.03) (47.68) (47.46) (49.48) (46.64) Farm labor includes workers 186.7** 188.2** 196.6** 198.9** 189.9** from outside the household (70.16) (70.75) (70.32) (70.36) (70.92) Household owns a plot of land 58.02 53.94 57.22 57.15 53.70 (58.61) (59.06) (59.61) (58.39) (59.32) Household size -25.57* -27.36* -29.40* -27.79* -28.15* (11.80) (12.17) (12.36) (12.08) (12.31) N 572 572 572 572 572 Notes: Standard errors in parentheses; * p<0.05, ** p<0.01, *** p<0.001. These regressions also include control variables for distance to market, the dependency ratio, respondent age, and respondent literacy. Coefficients have been omitted here to conserve space. Full results are available upon request. 40 Table D2. Determinants of Total Household Income, Annual US$. (1) (2) (3) (4) (5) Type 2 Household -470.4 -163.2 -247.9 -113.6 -156.9 (249.4) (281.4) (277.7) (294.8) (284.2) Type 3 Household -563.3 -464.0 -455.8 -515.7 -511.9 (301.4) (298.0) (299.8) (297.2) (302.4) Soft Agency 201.6* 140.6 (102.4) (102.4) Non-Agriculture Agency 295.9 214.7 (150.9) (159.4) Agriculture Agency 229.7 143.7 (132.8) (136.1) Agency Index 303.7* (143.3) Household owns a plot of 46.49 93.20 53.99 77.36 72.42 land (239.3) (232.6) (239.1) (233.8) (237.3) Time spent in other 330.0*** 319.1*** 349.9*** 321.1*** 335.6*** productive activities (75.34) (73.80) (74.63) (74.27) (74.94) Household size 210.2*** 205.9*** 211.9*** 212.7*** 211.9*** (45.73) (45.28) (45.62) (45.25) (45.47) Dependency ratio -449.4*** -451.2*** -462.2*** -449.8*** -456.2*** (125.3) (122.7) (125.6) (123.9) (125.1) N 503 503 503 503 503 Notes: * p<0.05, ** p<0.01, *** p<0.001. These regressions also include control variables for distance to market, number of crops, size of agricultural land, size of agricultural land squared, respondent age, and respondent literacy. Coefficients have been omitted here to conserve space. Full results are available upon request. 41 Table D3. Determinants of Food Insecurity. (1) (2) (3) (4) (5) Type 2 Household 1.228 1.278 2.762** 2.839** 1.981* (0.875) (0.947) (0.950) (0.962) (0.958) Type 3 Household 2.465** 2.020* 2.256** 2.805** 1.957* (0.898) (0.874) (0.848) (0.869) (0.865) Soft Agency -1.016** -1.190*** (0.335) (0.326) Non-Agriculture Agency 0.0759 -0.177 (0.365) (0.372) Agriculture Agency 1.587*** 1.808*** (0.351) (0.368) Agency Index 0.773* (0.368) Number of crops cultivated 0.358 0.436 0.167 0.0329 0.380 (0.486) (0.488) (0.488) (0.490) (0.488) Time spent in other -0.462* -0.539** -0.488** -0.383* -0.556** productive activities (0.189) (0.191) (0.184) (0.189) (0.185) Household receives remittances -0.933 -0.927 -0.792 -0.766 -0.917 (0.696) (0.705) (0.677) (0.679) (0.692) Household owns a plot of land -0.336 -0.331 -0.364 -0.383 -0.292 (0.734) (0.747) (0.722) (0.707) (0.741) Household size 0.119 0.146 0.201 0.175 0.169 (0.139) (0.137) (0.135) (0.138) (0.136) N 555 555 555 555 555 Notes: * p<0.05, ** p<0.01, *** p<0.001. These regressions also include control variables for size of agricultural land, size of agricultural land squared, dependency ratio, respondent, and respondent literacy. Coefficients have been omitted here to conserve space. Full results are available upon request. Note that a higher dependent variable indicates higher food insecurity. 42 Table D4. Determinants of Food Diversity. (1) (2) (3) (4) (5) Type 2 Household -0.167 -0.130 -0.183 -0.243 -0.0560 (0.202) (0.216) (0.225) (0.226) (0.225) Type 3 Household -0.365 -0.230 -0.236 -0.380 -0.240 (0.210) (0.210) (0.211) (0.211) (0.210) Soft Agency 0.298*** 0.308*** (0.0706) (0.0711) Non-Agriculture Agency 0.0343 -0.0156 (0.0778) (0.0768) Agriculture Agency -0.0225 -0.0623 (0.0797) (0.0803) Agency Index 0.105 (0.0855) Number of crops cultivated 0.0968 0.0785 0.0796 0.107 0.0691 (0.105) (0.108) (0.108) (0.106) (0.107) Time spent in other 0.0307 0.0482 0.0506 0.0296 0.0481 productive activities (0.0430) (0.0421) (0.0419) (0.0432) (0.0420) Household receives remittances 0.601*** 0.588*** 0.593*** 0.600*** 0.594*** (0.157) (0.161) (0.160) (0.159) (0.159) Household owns a plot of land 0.0244 0.0287 0.0255 0.0238 0.0315 (0.162) (0.166) (0.166) (0.162) (0.166) Household size -0.0935** -0.101*** -0.102*** -0.0956*** -0.0983*** (0.0285) (0.0290) (0.0290) (0.0285) (0.0289) N 555 555 555 555 555 Notes: * p<0.05, ** p<0.01, *** p<0.001. These regressions also include control variables for size of agricultural land, size of agricultural land squared, dependency ratio, respondent age, and respondent literacy. Coefficients have been omitted here to conserve space. Full results are available upon request. 43