Understanding Household Resilience of the Poor in Burkina Faso Interim Report

Transcription

1 Consultative Group to Assist the Poor (CGAP) Clients at the Center Financial Inclusion Research Working Paper PICTURE from BURKINA FASO Understanding Household Resilience of the Poor in Burkina Faso Interim Report Megan Gash and Bobbi Gray January 30, 2015 Understanding Household Resilience of the Poor in Burkina Faso i

2 Executive Summary The key theory of change for the microfinance industry has historically been one in which access to and use of formal financial services reduces household poverty; however, research conducted around the world has yet to successfully validate this theory. Newer theories of change suggest the primary benefit of financial services for poor households is that they help build household resilience by helping households anticipate, adapt to and/or recover from the effects of shocks in a manner that protects their livelihoods, reduces chronic vulnerability and facilitates growth. This interim research report aims to fill some of the gaps in knowledge about how different financial services, particularly those designed to anticipate and respond to common household shocks, contribute to household resilience in Burkina Faso and is guided by two primary research questions: 1) What strategies do poor households employ to manage economic, environmental and health shocks that disrupt their financial lives? 2) What roles do formal, nonformal and informal financial products play in improving household resiliency and building assets? Two Freedom from Hunger partners le Reseau des Caisses Populaires du Burkina Faso (RCPB), a credit union that provides formal financial services, and Office de Développement des Eglises Evangéliques (ODE), a non-governmental organization (NGO) that facilitates savings groups (SGs) participated in this research project. The research has focused on members from both institutions who live in the Samba department of the Passoré province or the Gourcy department of Zondoma province where both organizations have programs. Forty-six women are being followed for seven months. Twenty-five of the 46 women are RCPB members; the remaining 21 are from ODE. All participants are women. The research agenda was designed to answer the above two questions and has four main datacollection activities: formative research, resilience diaries, economic games and a wrap-up qualitative assessment. This interim report will summarize findings from the formative research and the first five of ten resilience diaries. It will focus primarily on answering the first research question but will attempt to share some initial impressions on the second research question. What strategies do poor households employ to manage economic, environmental and health shocks that disrupt their financial lives? Resilience has primarily been explained as an attitude towards life. A resilient household is made up of people who get up early, go to bed late and work hard to get ahead. Yet resilience is not an easy capacity to develop. The results from the first five surveys of the resilience diaries reveal significant levels of vulnerability across the group; 64 percent of the women live below the $1.25/day line at 2005 purchase-power parity (henceforth referred to as the $1.25/day international poverty line ) and a greater majority are chronically food insecure. Most describe the good life as having food, heath and the ability to take care of their children; despite these being basic needs, only 74 percent of them believe it s possible to achieve the good life. In addition, poverty was labeled as a shock Understanding Household Resilience of the Poor in Burkina Faso ii

3 resulting from the long-term lack of resources to cover daily life expenses versus a definition of their income status. The most common, or idiosyncratic, shocks are illness and injury, loss of livestock, death of family members, and poor harvest. They also face semi-regular covariate shocks, such as droughts every two to four years, as well as the less common but still relevant famine, political crisis, and threats such as Ebola from bordering countries. There are also stressors that are more specific to the study participants, by their nature as women. While women play a substantial role in the household economy, they are limited both by gender norms that inhibit their ability to pursue more profitable income-generating activities (IGAs), as well as lack of time and resources to focus on building successful enterprises. Almost three-quarters reported that being a woman negatively influenced their ability to be successful at generating income, mostly because they have to prioritize family needs before they can devote time to their IGAs. Managing food resources figure prominently among the strategies used to cope with shocks. In general, the most common coping mechanisms are first using savings at home, then reducing food consumption, selling grain, selling small livestock, purchasing on credit and lastly, borrowing from an SG. Borrowing from financial institutions, family and friends is less preferred. As resources become available to them, the women reprioritize the way they manage any particular shock. For example, after harvest, more sell grain and fewer reduce food consumption, make purchases on credit or borrow from friends and family. What roles do formal, nonformal, and informal financial products play in improving household resiliency and building assets? The women s overall reported use of formal financial products is very limited and the demand is widely unmet. The most common formal products or services used are RCPB loans and remittance services, although one-half of remittance services are through informal means. The formal services used seem to somewhat help improve cash flow, in terms of increasing access to cash to cover the costs incurred from shocks, but it is not clear the extent to which they help with building assets and resilience. It is clear that the use of nonformal and informal financial services greatly exceeds the use of formal financial services. SGs appear to be one of the more commonly used nonformal services among women claiming to use SGs to save money for purchasing livestock, paying health expenses, school fees and for food and IGA expenses. For informal services, the women borrow from friends and family, and make purchases on credit from local merchants and, as mentioned earlier, receive remittances often by hand-to-hand transporters. It seems that the nonformal and informal services play a role in improving cash flow or consumption smoothing as well, but it is difficult to gauge the extent to which they are helpful in building resiliency, particularly since the most common shockcoping strategies include selling assets and managing food resources. Even though they are not strictly speaking financial services, the frequent selling of livestock almost makes it a financial service in and of itself, acting both as a form of currency in the marketplace as well as an interest-bearing savings account (when fattened over time and sold for a higher price than Understanding Household Resilience of the Poor in Burkina Faso iii

4 bought). Even though the savings account on four legs is available to cash in, it is likely not often as profitable as it could be or is desired to be. Livestock appears to be a risky investment, but is extremely common given that financial services are generally lacking. Also not of central focus, financial services are related to the other main coping strategies: reducing food consumption and selling grain. Savings and microcredit are used to buy food, but almost 50 percent of the women have also gone hungry to make a loan payment or a savings contribution. Microcredit is used with grain speculation investments, although its use seems limited among this group of women. Reducing food consumption means they are buying less food and can divert money they would have spent on food towards other expenses. Selling grain means that they are selling food stocks that they would have eaten during the lean season. However, both of these strategies are likely detrimental to their health and play a role in increasing chronic food insecurity rates. Reducing food consumption and selling grain are the last strategies they should use, not the first. In relative terms, these women can define which households are resilient, and which are not. However, in absolute terms, using these coping strategies, it is difficult to consider them as resilient. They find ways to respond to shocks and stressors in the short term, but these mechanisms do not seem to help them recover or make them more resilient in the long term. Even without multi-year data in this study, one can guess that eating less and eating lower-quality foods will have long-term effects on their health and their productivity. It is also difficult to tell how much they are increasing livestock assets over time with the constant acquisition and sale of animals again, selling assets seems to be a response behavior and not part of recovery. Their capacities to respond are not preventing long-lasting adverse development consequences, which is the foundation of the definition of resilience itself. Considering this reality, how can resiliency be built or increased for this population? These coping mechanisms have likely been used for a long time and are not easy concepts to address. Nevertheless, acknowledgement of these adverse coping mechanisms must be part of the ongoing conversations related to financial inclusion. Financial services need to be designed, at a minimum, to provide vulnerable households with more options to assist them in not only responding to shocks, but also in recovering from these shocks. Although this study is only at the midpoint, the following are ideas for improving financial product and service design for this population: Financial products and services that help this population get more and better food should be prioritized. Any financial service cannot come at the cost of them not eating; any new financial product or service should take into account the opportunity costs of a family s food resources. Food security should be considered as part of the design of a financial service, not seen simply seen as an outcome. Terms (including grace periods, number of payments, frequency and timing of payments) must be more flexible to account for the unpredictability in their lives. This would require a balancing act for financial service providers seeking to manage risk while producing profits. Understanding Household Resilience of the Poor in Burkina Faso iv

5 Creativity is required. Standard microcredit and savings products seem to be somewhat helpful for this population, but more specialized products that speak to their reality are in demand. Devices such as commitment savings products might help prepare households better for future expenses. Products and services that target productivity beyond offering financial services are in demand. Coupling financial services with consumption support and technical-skills training would greatly improve their use. Graduation programs and similar multi-component models seem appropriate for this highly vulnerable population. Financial products and services can and should play a role in helping these women and their households build resilience, but therein lies the challenge to find better solutions to meet the needs of their complex lives. This research will continue to explore these issues, identifying and dissecting more parts of the problem, with the hope to making some of these solutions more within our reach. Understanding Household Resilience of the Poor in Burkina Faso v

6 Table of Contents Introduction... 1 Background... 2 Burkina Faso... 4 Methods... 7 Research Partners... 7 Research Activities... 8 Research Methods for the Resilience Diaries... 9 Resilience Diary Participants... 9 Design of the Resilience Diaries... 9 Data Processing Findings Formative Research Resilience Diaries Findings for Surveys Demographics The Good Life Livestock Food Security Education Resilience Income Generation Financial Services Attitudes Discussion Conclusion Annex Resilience Diary Composition by Survey Number... 52

7 Introduction The key theory of change for the microfinance industry has historically been one in which access to and use of formal financial services reduces household poverty. Rigorous evaluation results from research conducted around the world have not successfully validated this theory. 1 However, a new theory 2 is emerging in which the primary benefit of financial services for poor households is that they help build household resilience; financial services play a role in helping households anticipate, adapt to and/or recover from the effects of shocks in a manner that protects their livelihoods, reduces chronic vulnerability and facilitates growth. 3 Thus, financial services contribute to poverty alleviation, and not necessarily poverty reduction. Helping households manage risk and develop appropriate coping strategies is, however, a necessary steppingstone to reducing poverty and improving economic growth. Households that can avoid depleting assets can have a positive impact on current and intergenerational poverty. A review of results from existing qualitative and quantitative studies suggests that access to and use of financial services do indeed build resilience; 4 however, further insight is needed to understand the decision-making processes about how households use financial services to manage shocks to continue improving financial products and services offered. As an attempt to further understand these decision-making processes, Freedom from Hunger is conducting research in the study Chutes and Ladders: Understanding How the Poor in Burkina Faso Anticipate and Recover from Shocks and the Role of Financial Services in Building Household Resilience (henceforth referred to as Understanding Resilience in Burkina Faso ). This project falls under the CGAP Clients at the Center Financial Inclusion Research Fund, funded by The MasterCard Foundation. It aims to help fill the gaps in knowledge about how different financial services, particularly those designed to anticipate and respond to common household shocks, contribute to household resilience. The research is guided by the following two research questions: 1) What strategies do poor households employ to manage economic, environmental and health shocks that disrupt their financial lives? 2) What roles do formal, nonformal and informal financial products play in improving household resiliency and building assets? It is anticipated that this research will promote financial inclusion by contributing to the understanding of the needs, preferences and behaviors of poor households in Burkina Faso, as well 1 (Accessed June 25, 2013) 2 Dunford C Evidence Project Blog. (Accessed June 25, 2013) 3 Adapted definition from USAID. (Accessed June 25, 2013) 4 Dunford C Evidence Project Blog. (Accessed June 25, 2013) Understanding Household Resilience of the Poor in Burkina Faso 1

8 as provide information for improving financial product and service design to better meet the needs of the poor. This interim report summarizes findings from the formative research as well as the first five of ten planned resilience diaries and demonstrates how this data contributes to answering the two key research questions listed above. As this research is not related to any specific program or product, its primary purpose is to illustrate how families anticipate and cope with shocks and relates household behaviors and preferences to the opportunities for improving financial services offered to them for building their resilience. It is organized by first reviewing some of the recent literature on resilience and providing contextual information on Burkina Faso. Then it describes research partners and research methods used for the various activities included in the study. The findings section describes resilience diary outcomes in detail. Last follows the discussion and conclusion, with these two sections summarizing key findings related to the two research questions and examining possible implications for the design of financial services whose purpose is to build household resilience in the face of continual covariate and idiosyncratic shocks. Overall, the paper focuses on presenting outcomes to date rather than drawing overarching conclusions from the study as a whole. Background Resilience has become an increasingly common focus in international development work in the past few years. It is considered a new focus for development practitioners, although perhaps some see it as a new lens or framework applied to relief and humanitarian development work already in progress. To initiate this work, organizations are attempting to define the concept of resilience, envision what it means to build it, and evaluate how successful programs are in building it. This has led to various organizations putting forth definitions, causal frameworks and prescriptions for measurement. Because resilience is so multifaceted, finding consensus on these concepts has been difficult, therefore, making it difficult to make comparisons across projects. The conversation on resilience is a moving target; how it is defined and measured today may not be how it is defined and measured tomorrow. An excellent illustration of these complexities exists in Catherine Fitzgibbon s Technical Brief Progress to Date with Measuring Resilience in the Horn of Africa, 5 which describes various obstacles currently present in the conversation around understanding resilience. She first points out that not only do the various institutional definitions prohibit a consensus on the concept, but that the definitions are so general that nearly any intervention can now be re-labeled as resiliencebuilding. 6 Some organizations put an emphasis on responding to a disaster, some look at resilience as disaster plus development, and others look at it as a process, with none of them describing the 5 Fitzgibbon, C. March Technical Brief: Progress to Date with Measuring Resilience in the Horn of Africa. Drylands Learning and Capacity Building Initiative for Improved Policy and Practice in the Horn of Africa: Kenya. Resilience_FINAL_March% pdf (Accessed January 13, 2015) 6 Ibid, pg. 1. Understanding Household Resilience of the Poor in Burkina Faso 2

9 end state of resilience. 7 Measurement indicators can help clarify what it means to build resilience, but Fitzgibbon observes that when organizations try to select them, they are paralyzed by choice and end up with exhaustive lists of indicators attempting to cover all facets of the concept. 8 These lists are not only impractical to collect, but also introduce a complexity for comparisons; since it is not clear whether resilience is a universal or locally contextual concept, how does one compare these indicators? Fitzgibbons points out that in a study in East Africa, a household in Uganda is considered resilient if it has more than 10 cattle, but in Kenya, a household is considered resilient if it has 50 cattle (or camels), plus 200 goats. 9 Acknowledging locally relevant definitions of resilience can help within one study, but comparisons across multiple studies are difficult without a universal definition. These issues can make initiating a research agenda on resilience seem somewhat daunting. To guide its thoughts on the topic, Freedom from Hunger sought insight from an emerging leader in the field of resilience to suggest a framework. In 2013, Freedom from Hunger adapted a resilience framework from TANGO International ( who had created a framework built on previous disaster and livelihood frameworks (see bottom notation in Figure 1 for reference). Staff at TANGO International have been active in industry conversations on resilience and are an active and publishing member of the Resilience Measurement Technical Working Group 10 (RM-TWG) of the World Food Programme. It is worth noting that, in 2014, RM- TWG published a consensus definition of resilience, 11 as well as a set of 10 resilience measurement principles 12 and a model for measurement. 13 Freedom from Hunger has continued to follow the work of both TANGO International and RM-TWG to incorporate key resilience concepts into its work. The Freedom from Hunger Resilience Framework (see Figure 1) focuses on adaptive capacity, or the ability to learn from experience and adjust responses to changing external conditions, yet continue operating. 14 In much of the resilience literature, the concept of resilience is discussed as a capacity with which to respond to shocks. There are three types of resilience capacities to consider: absorptive, adaptive and transformative. 15 Adaptive capacity is where financial services falls best in 7 Ibid, pgs Ibid, pg Ibid, pg. 4, referencing a Community-Based Resilience Analysis Assessment Report by the United Nations Development Programme Drylands Development Centre. 10 The Resilience Measurement Technical Working Group, co-sponsored by the European Union and USAID, is a group of 20 leading experts in resilience measurement from government and NGOs, including FAO, IFAD, Cornell University, TANGO International, Mercy Corps and others. For more information, see 11 Resilience is the capacity that ensures adverse stressors and shocks do not have long-lasting adverse development consequences. Constas, M., T. Frankenberger and J. Hoddinott. January Resilience Measurement Principles: Toward an Agenda for Measurement Design. Resilience Measurement Technical Working Group, Food Security Information Network. Technical Series No. 1. Rome, Italy: World Food Programme. p 6. 12Ibid. 13 Constas, M., et al. November A Common Analytical Model for Resilience Measurement: Causal Framework and Methodological Options. Resilience Measurement Technical Working Group, Food Security Information Network. Technical Series No. 2. Rome, Italy: World Food Programme. 14 Von Grebner, K., et al Global Hunger Index: The Challenge of Hunger: Building Resilience to Achieve Food and Nutrition Security. Bonn, Washington, D.C. and Dublin: Welthungerlife, International Food Policy Research Institute and Concern Worldwide. p Ibid. Understanding Household Resilience of the Poor in Burkina Faso 3

10 Exposure Sustainable-livelihood Strategies for Women Asset-building (Human, Social, Financial) Access to Financial Services (Credit, Savings, Insurance) Social-service Networks (health, nutrition, livelihood) that access to financial services is a leverage point, or where an individual or household can make an incremental change in the response to a current shock or in anticipation of a future shock. That leverage point, or change, could help lead a household down a path of resiliency instead of one of vulnerability. Ultimately, a resilient path can lead to better food security, adequate nutrition and health status, and disaster risk-reduction for a household. Figure1. Reference Resilience Framework Freedom from Hunger Resilience Framework Shocks & Stresses e.g., natural hazard, food shortage, health, economic Adaptive Capacity e.g., ability to cope with Shocks & Stresses Freedom from Hunger Leverage Points Reaction to Shocks & Stresses e.g., survive, cope, recover, learn, transform Household Outcomes Shocks (-) ( + ) ( + ) ( + ) ( + ) Resilience pathway Bounce back better Food Security Adequate nutrition & health status Bounce back Disaster riskreduction Stresses Vulnerability pathway Recover but worse than before Food Insecurity Malnutrition & poor health status ( + ) (-) (-) (-) (-) Collapse Disaster losses Gender Lens TANGO Adapted from DFID Disaster Resilience Framework (2011), TANGO Livelihoods Framework (2007), DFID Sustainable Livelihoods Framework (1999) and CARE Household Livelihood Security Framework (2002) This framework serves as the basis for the design of the Understanding Resilience in Burkina Faso study. The study will capture information on the participants about their use of and access to each of the four leverage points, exploring how they influence which path households head down (or where they are already positioned). This study also acts as grounds for testing the framework; will this description of adaptive capacity prove to be a useful lens for examining resilience? Are there other aspects to resilience to consider? The outcomes will undoubtedly illuminate the complexity of resiliency for a population in a challenging environment. Burkina Faso Burkina Faso is one of the poorest countries in Africa, with 44.6 percent of the population living on less than $1.25/day international poverty line, and ranking 181 out of 187 countries on a global scale Understanding Household Resilience of the Poor in Burkina Faso 4

11 of development, qualifying it as a low human development country. 16 Drought in Burkina Faso, as with most of the countries in the Sahel, is often depicted and experienced as a crisis, but it is actually more the norm. 17 Due to chronic challenges of drought and consequential food crises, food and nutrition security are persistent challenges faced by vulnerable populations. 18 Rainfall is the key determinant in how households make a living: it determines how land is put to use, the degree to which households depend on livestock, and other nonagricultural sources of income. In the northeastern zones of Burkina Faso, where the resilience diaries research takes place, households allocate most of their land to millet, sorghum and cowpeas. Livestock herding assumes quite a bit of importance, as does gold mining and labor migration for households too poor to own substantial herds. There are also strong cultural and familial ties among households in this part of Burkina Faso and the Ivory Coast, resulting in remittances being important for livelihoods. Very little reliable rain falls in the north; consequently, households that rely on agriculture alone are financially vulnerable. 19 The seasonal and agricultural calendar highly influences the lives of Burkinabe households. Table 1 combines two calendars. Market research data from a prior Freedom from Hunger survey 20 in Burkina Faso from the same region where the resilience diaries are being conducted depicts the seasonality of illnesses as well as fluctuations in use of credit and savings and flows of expenditures and income. This data is provided below, with the columns corresponding to the months of the year. The high and low periods are depicted by an average range of numbers on a scale of 1 to 5, with 1 referencing a low level and 5 a high level. These high and low periods are also depicted to make it easier to view by using darker colors to reference high levels (dark red) and lighter colors for low levels (yellow). The Burkina Faso FEWSNET report also shares how specific seasons influence agricultural livelihoods. These influential periods are depicted in the row above the months of the year. This data is provided as a useful reference when interpreting the results shared in this report. 16 Human Development Report (Accessed January 20, 2014). 17 Hesse, C., et al Building Climate Resilience in the Sahel. International Institute for Environment and Development. 18http:// 19 Dixon, S., and J. Holt Livelihood Zoning and Profiling Report: Burkina Faso. United States Agency for International Development Famine Early Warning Systems Network (FEWSNET). Washington, D.C. 20 Traore B., et al Staying healthy means fighting poverty: The RCPB and health protection options for the poor in the north region. Unpublished market research report. Davis, CA: Freedom from Hunger. Understanding Household Resilience of the Poor in Burkina Faso 5

12 Understanding Resilience in Burkina Faso 6 Table 1. Seasonality Calendar for Burkina Faso Season Dry Season Rainy Season Dry Season Periods Cold Hot Winter Mild Cold Off-season harvest Transhumant migration Off-season hunger season Land preparation Cotton payment Hunger season Grain harvest Off-season land preparation Cotton market Transhumant migration Month Jan Feb Mar Apr May June July Aug Sept Oct Nov Dec Malaria Meningitis Diarrhea Stomach pain Cough Expenditures Income Credit Saving Codes 1 low high

13 Very few households in Burkina Faso have access to formal financial services (Table 2). The World Bank Findex study in estimated that no more than 13 percent of the total adult population held an account at a formal financial institution; only 6 percent of the poorest were likely to have an account. Eight percent were estimated to participate in community-based savings groups (SGs). Table 2. World Bank Global Findex Figures for Burkina Faso FINDEX Indicators Measures in 2011 Percentage of adults with an account at a formal financial institution 13% Poorest quintile of adults with an account at a formal financial institution 6% Percentage of women with an account at a formal financial institution 11% Percentage of adults using mobile money 1% Percentage of adults with a formal savings account 8% Percentage of adults participating in community-based SGs 8% Percentage of adults who took a loan from a formal financial institution 3% Methods Research Partners Two Freedom from Hunger partners le Reseau des Caisses Populaires du Burkina Faso (RCPB), a credit union that provides formal financial services, and Office de Développement des Eglises Evangéliques (ODE), a non-governmental organization (NGO) that facilitates SGs participated in this research project. RCPB has been a Freedom from Hunger partner for over 20 years, implementing Freedom from Hunger s flagship product Credit with Education, which is based on village-banking financial services integrated with nonfinancial services such as health, business or financial education. Also in 2006, RCPB partnered with Freedom from Hunger to further test health products and services as well as design and implement a health savings and loan product. ODE recently became a Freedom from Hunger partner and implements Saving for Change, a methodology for self-managed savings and lending groups integrated with simple trainings in health, business and money management. Saving for Change brings basic financial services to areas that are typically beyond the reach of microfinance institutions and, in doing so, creates sustainable, cohesive groups that tackle social issues facing their members and their communities. Saving for Change was jointly developed by Freedom from Hunger, Oxfam America and Strømme Foundation. The resilience diaries research has focused on members from both RCPB and ODE who live in the Samba department of the Passoré province or the Gourcy department of Zondoma province where both organizations have programs. A map of these provinces is provided in Figure Demirguc-Kunt, A., and L. Klapper Measuring financial inclusion: the Global Findex Database, Volume 1. World Bank. (Accessed January 22, 2015) Understanding Resilience in Burkina Faso 7

14 Figure 2. Location of Research Project: Passoré and Gourcy Provinces, Burkina Faso 22 Location of Resilience Diaries research (provinces 31 and 44) It is also important to note that while RCPB is a credit union and therefore offers a range of financial services, including individual credit and savings accounts, the sample of RCPB clients for this study are Credit with Education clients with an emphasis being placed on groups of clients that either have a health savings and loan account or have access to it if desired. Research Activities The project has four main data-collection activities: formative research, resilience diaries, economic games and a wrap-up qualitative assessment. After completion of data-collection and analysis, results will be presented to in-country partners and disseminated. Details on the activities are below: Formative research was conducted between May and June 2014 to gather community-level data that would better help us understand the types of shocks households have faced during the last five to ten years, how households anticipate and respond to shocks, and the gender dynamics behind their decisions. This phase was used as market research to help develop the financial diary survey instruments. A total of 15 focus group discussions (FGDs) with 116 members (28 men and 88 women) across 10 villages served by RCPB and ODE participated in the formative research. A formative research report has already been published. 23 Economic games were conducted in collaboration with the Agricultural and Resource Economics Department and Professor Michael Carter from the University of California, Davis. The economic games were designed to test women s responses to health shocks in light of 22 Provinces of Burkina Faso. Wikipedia. (Accessed January 15, 2015) 23 Gray, B. and M. Gash Understanding Resilience among Households in Rural Burkina Faso: Formative Research Summary. Davis, CA: Freedom from Hunger. Understanding Resilience in Burkina Faso 8

15 available financial instruments. At the time of this report, the data collection had been completed and analysis was under way. Resilience diaries are a series of 10 surveys that examine shocks experienced by 46 households over a seven-month period, as well as their coping mechanisms. Each of the 10 diaries focuses on a different topic involving participants (1) demographics and hopes and dreams, (2) shockcoping strategies, (3) use of financial services, (4) food security, (5) income-generating activities, (6) social capital, (7) household decision-making dynamics, (8) health, (9) attitudes and perceptions, and (10) program participation and future outlook. A full description of these diaries is provided in the Annex. Wrap-up qualitative phase consists of FGDs and key informant interviews that are designed to help support interpretation of the data from the resilience diaries and the economic games. Partner presentations and dissemination will be conducted after the completion of the research activities. RCPB and ODE and other interested stakeholders will convene to discuss the implications of this research for their programs. This is one of the most important steps of this research phase as it will provide an immediate translation of the research findings into possible actionable steps by RCPB and ODE, as well as other actors. At the time of this report, the formative phase, the economic games, and seven of the 10 resilience diaries had been completed. This report will focus on the findings from the first four months of data collection consisting of the first five resilience diaries. The other diaries and economic games outcomes were still in analysis at the time of this report. Research Methods for the Resilience Diaries Resilience Diary Participants Forty-seven women were originally interviewed with the first diary; however, the remaining interviews included only 46 as one participant moved outside of the research area and was no longer included. Twenty-five out of the 46 women are RCPB members; the remaining 21 are from ODE. All participants are women. Design of the Resilience Diaries The resilience diaries consist of 10 individual surveys. Each of the 10 diaries has sections that were repeated for all of the diaries, including quantitative and qualitative questions related to a) the household members and any changes in household composition; b) recent positive surprises and shocks experienced in the prior three weeks and how the household coped with these events; c) household food security and whether food consumption changed during the prior three weeks; d) changes in household assets, with a particular focus on livestock; e) changes in household finances focusing on actual income received and its comparison to expected income, comparison of actual to expected expenses and use of financial instruments such as savings and loans; and f) changes in attitudes such as women s level of hope for the future, support from family and friends, overall happiness, and their perceived ability to meet their financial obligations. Understanding Resilience in Burkina Faso 9

16 In addition to the repeated information, each survey had a special focus and was meant to dive further into the topics being repeated at each interview point. Each of the special foci for each diary is described in the Annex. Again, please note, that this report will provide results for the first five surveys only. Data Processing Freedom from Hunger designed all survey instruments and provided these to Lessokon Sarl, a research firm located in Ouagadougou, Burkina Faso. Lessokon Sarl pre-tested all survey instruments with three or four different participants. While the interviews were written in French, Lessokon used local languages where necessary. Lessokon staff conducted the interviews for the first five surveys between August and November All diaries were transcribed into predesigned MSExcel databases. Responses to qualitative questions were summarized and inductive analysis was often used to describe key findings for each question. For quantitative questions, averages or distributions were tabulated in Excel. Findings Formative Research Resilience was primarily explained as an attitude towards life in the formative research. 24 A resilient household is made up of people who get up early, go to bed late and work hard to get ahead. Those who are not resilient are stuck in poverty and are generally lazy. Women were harsher, believing the non-resilient home was a parasite on the environment. Men, however, defined resilience in terms of their means an accounting of what they have. They thought nonresilient homes resulted from a lack of agreement within the household (on how to manage household resources) and that they live particularly stressful lives. Most members of RCPB and ODE placed themselves in the resilient category of people living in their communities. A drought and period of famine that started about 10 years ago was at the forefront of many people s minds. Most of the more current shocks either seem related to or are holdovers from the famine and most households still feel they are digging out of it. In the past year they have had better crops, resulting in a greater sense of hope for the future. Interestingly, poverty itself is considered a shock. Poverty was explained as the situation in which a household is pummeled by too many shocks to recover from as there are not enough resources to cover basic needs. Participants say it happens to a lot of people when they are weak and vulnerable, that the combination of shocks destroys one s ability to respond. As a result of these shocks, they described judiciously taking out credit and relying on remittances and help from family abroad or on government social programs meant to reduce malnutrition. They ve rationed food, taken on extra labor working in the fields, sold their assets and relied heavily 24 Ibid. Understanding Resilience in Burkina Faso 10

17 on their SGs. Reflecting their definition of resilience, the primary coping strategy was to simply work harder, whether this meant taking on additional income-generating activities, getting up earlier or staying at work later. While savings was mentioned as an important strategy for building resilience by many households, it did not seem enough. Since shocks hit regularly, the most current shock determined how savings were spent; there was not much earmarking of this resource to help manage specific shocks, such as health issues. Women were more likely to use credit than men, but both used it judiciously and sometimes sold assets before turning to it. While women had jurisdiction over decisions concerning the use of savings for health, food and education expenditures, as compared to other expenses, the husband typically made the final decision on responses to larger shocks. Generally, there did not appear to be much earmarking of their income or savings for shocks. Many of those interviewed indicated they had access to all the financial services they needed that could be useful in responding to shocks; however, the bigger issue was whether the implementation of these products facilitated quick access to their money when it was needed. Resilience Diaries Findings for Surveys 1 5 Demographics Survey 1 engaged 47 participants, 25 all of whom were women. About 53 percent of them were RCPB members (25 out of the 47) and the remaining were ODE members. Table 3 lists the Key Demographics. The average age was 46 years; the youngest participant was 23 years of age, the oldest was 65. The majority of women (66%) are married and living in a polygamous relationship; 26 percent of them are in a monogamous relationship and 8 percent are widows. Participants were asked about household size, using the definition of her husband and her direct family, which focuses on her direct dependents that would spend most of the next 12 months together, sleeping in the same place and sharing meals excluding temporary visitors and the extended households (usually brothers of the head of household and their wives and children). The average household size was 14 members, with some households reporting having as few as three members and as many as 42. There was an average of five children per household. Most participants reported also having a child or family member who no longer lives with the household; most of them are married or left to work in the gold mines or elsewhere within Burkina Faso or in the Ivory Coast. As these households were followed through Surveys 5, between four and seven households reported that the size of their household changed due to one of the following reasons: family members who returned from working in the Ivory Coast or elsewhere in Burkina Faso, births in the family, and families taking in other family members or orphans. In some households, as many as three people returned to the household at one time. 25 As referenced in the Methods section, after the Survey 1, which focused on demographics, one of the 47 original participants dropped out of the study; therefore, most of the remaining sections of the report will reference results as they relate to the 46 women who participated in Surveys 2 5. Understanding Resilience in Burkina Faso 11

18 Very few women could read and very few had ever attended school. For those who reported attending school at some point, only one person made it to the tenth level, and most reported the highest level attained as being the third level. Of their children, an estimated five children per household were of school-age, but on average, only two are attending school. Only 25 percent of the women who had school-aged children had all of their children attending school. Of those who are not in school, the most-mentioned reason for their children not attending was that they did not have the financial means to put them in school. Some mentioned that their husband, or the child s father, did not find it important for them to attend. Some of the young daughters are married and therefore are no longer in school. Almost half identified as Muslim; 32 percent as Christian and 19 percent as Animist. Most identify as being members of the Mossi ethnic group. A few identify with the Gourounsi and Bobo ethnic groups. The majority of the women identified their spouse as the head of household (81%); but 15 percent indicated they were head of household, often because the women were widows. If they said that someone else was head of household, it was as a widowed woman living under the roof of her father-in-law or deceased husband s brother. Table 3. Key Demographics Indicator Measure Percentage who are RCPB members 53% Percentage who are ODE members 47% Average age 46 years Percentage who are in a polygamous marriage 66% Average household size 14 Percentage who are illiterate 85% Percentage who ever attended school 11% Percentage who are Muslim 49% Percentage who are periodically food insecure 91% Percentage who are chronically food insecure 89% The majority of the women were classified as food insecure. Ninety-one percent were classified as periodically food insecure; likely facing food insecurity during the lean season whereas 89 percent of them were classified as chronically food insecure because they face this condition almost year-round. The high rates indicate that this population is generally chronically food insecure. It is important to note that the food security measure is subject to being influenced by the period to which it was applied. Since the first survey was conducted during the lean season, it is likely detecting food insecurity during a period when higher rates are expected more so than during any other period of the year. Due to the high levels of food insecurity in this population, it is not surprising to see that very few households reported eating meat; with the maximum being once a week for any of the households. More detail on food security outcomes are described in the food security section later in the report. Understanding Resilience in Burkina Faso 12

19 Using the Progress Out of Poverty Index (PPI), 26 Figure 3 shows that 49 percent are estimated to live below the national poverty line, estimated at XOF 226 per person per day based on year 2003 measures. Sixty-four percent live below the $1.25/day international poverty line (estimated at XOF 288 in 2003 measures); 27 percent live below the USAID extreme poverty line, which represents the median expenditure of people (not households) below the national poverty line. All poverty rates are slightly higher among our study population when compared to the national averages. Figure 3. Poverty Status and Relevant National Benchmarks 70% 60% 50% 40% 30% 20% Our sample National 10% 0% % living below the USAID Extreme poverty line % living below national poverty line % living below $1.25/day international poverty line While differences in food security and poverty levels may have been anticipated between the RCPB and ODE members, given perceived differences in poverty between microfinance groups and SG program strategies, there were no large differences in poverty or food security between the RCPB and ODE members. ODE members were slightly more likely to be below the national poverty line and the $1.25/day international poverty line were the same; very similar rates were found for both RCPB and ODE members for the USAID Extreme poverty line. Demographics Key Findings Study participants live in relatively large polygamous households whose membership tends to fluctuate over time from migration and return migration, birth, and orphans or other family members joining the household. Poverty levels are high with 64 percent of households living under the $1.25/day international poverty line. Illiteracy is high, with 85 percent of study participants indicating their inability to read. Almost equal amounts had never attended school. 26 This survey was developed using a national poverty survey conducted in Therefore, the benchmarks provided here are provided by Mark Schreiner in the documentation for the Burkina Faso PPI survey and may not relate to latest poverty measurements found by the World Bank or others. See the PPI documentation at Understanding Resilience in Burkina Faso 13

20 The Good Life When asked to describe the good life or well-being, the top five definitions reported matched descriptions of the lowest levels of Maslow s hierarchy of needs. These are physiological and safety needs and include the following: food, health, the ability take care of their children (meet their needs), money to cover basic needs and meeting schooling costs (Figure 4). Figure 4. Top 5 Definitions for The Good Life 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% Food Health Cover basic financial needs Children's education Ability to take care of children/see them as successful In addition to these top five responses, a few women mentioned having a successful income-generating activity (IGA), a good relationship with their spouse, cohesion of the family and ability to resolve their problems (primarily defined as financial problems). Only 74 percent feel they can achieve their definition of the good life mainly because they lack food and money. However, most feel their children will achieve the The good life is Good health, financial means and if you have children, having the necessary financial means to build my business so that I can support the education needs of my children and give them enough to eat. 40-year-old RCPB client living in Yako good life because of their children s future opportunities and because their hard work today contributes to the promise of their children being successful tomorrow. Some of their children are already married, which they see as providing them a promising future. In addition, in recent years, gold mining has become prevalent and they see their children s financial opportunities improving because of this new local economic opportunity. The Good Life Key Findings The top five definitions of the good life match descriptions of the lowest levels of Maslow s hierarchy of needs, which are physiological and safety needs: food, health, ability to take care of their children (meet their needs), money to cover basic needs, and meeting schooling costs. Only 74 percent feel they can achieve their definition of the good life mainly because they lack food and money. Understanding Resilience in Burkina Faso 14

21 Livestock Each survey asked participants for the number of livestock the household currently owned, whether there was a change in ownership since the prior survey, and the reason behind the change. This information, in addition to the questions related to shocks and corresponding coping mechanisms since the prior survey, revealed a frequent use of livestock as a means to pay for expenses related to shocks as well as general household expenses. Table 4 lists the average number of livestock owned by the 46 households, and tracks and the changes in the number of livestock owned during the five survey periods. On average, households owned 14 chickens, two cows, 11 goats, one donkey and one other type of livestock. As with many of the other dimensions studied in this report, comparing livestock ownership with household poverty level will be a useful analysis to facilitate a deeper understanding of the relationship between livestock ownership and poverty status. For example, it is not yet clear whether the majority of the livestock ownership is held with the less poor or not, as one might expect. This will be explored in the final report. The type of animal most mentioned is used for the category labels in Table 4. For example, the category chicken encompasses all poultry mentioned, such as guinea fowl. The category goats included sheep and pigs as well. The other category often included horses. Table 4. Average Household Ownership of Livestock Survey Periods Average Chickens (includes all poultry mentioned, such as guinea fowl) Goats (category includes sheep and pigs) Donkeys Cattle Other Table 5, however, shows that this average level of ownership per survey period hides a rather dynamic ownership of livestock with chickens and goats in particular. The percentages for changes come from comparing the difference in the number of livestock participant households owned at Survey 5 as compared to Survey 1. To explain further, about 72 percent of households had a change in the number of chickens they owned. This does not mean those who experienced no change did not gain or lose livestock during the five survey periods; this simply compares the number they started out with at Survey 1 and the number they accounted for in Survey 5. Thirty percent experienced a gain in the number of chickens owned. When there was a gain, the household gained on average nine chickens. However, 41 percent of the participant households experienced a loss, which includes both chickens sold and chickens that died. If they experienced a loss, they lost on average 11 chickens. Similar to chickens, goat ownership fluctuated quite a bit. Sixty-three percent experienced a change in goat ownership, with 41 percent experiencing loss of an average of five goats; 22 percent experienced a net gain by an average of six goats. By contrast, donkey, cattle and other (mainly horse) ownership changed very little. Understanding Resilience in Burkina Faso 15

22 Table 5. Change in Livestock Ownership Change Net gain Average gain Net loss Average loss Chickens 72% 30% 9 41% 11 Goats 63% 22% 6 41% 5 Donkeys 22% 11% 2 11% 1 Cattle 28% 17% 3 11% 2 Other 20% 7% 2 13% 2 Figure 5 further explains the reasons behind the net gain or net loss between Surveys 1 and 5. Compared to Table 5, the following data will account for all of the reasons reported behind the changes in livestock. In Figure 5, the left-hand vertical axis depicts the average number of participants that reported gaining or losing chickens, and the orange bars show the various reasons for the change. For example, 46 percent of participants had chickens die during at least one point over the course of the five surveys (resulting in a loss), and 37 percent had chickens that gave birth (resulting in a gain in chicken ownership). The blue blocks depict the average percent gain or loss and corresponds to numbers on the right-hand vertical axis. Therefore, 46 percent of participants had chickens die during any time during the five-survey period, with the average number of chickens lost at 15 (depicted as negative 15). Thirty-seven percent of participants increased the number of chickens they owned through the birth of new ones; the average gain was 12 chickens. It is striking to see how many participants lost their chickens due to death (some women mentioned an epidemic of disease killing their chickens, such as Newcastle disease), reiterating that livestock death can be a costly shock. One woman said she sold all of her chickens to prevent a loss from the disease outbreak. It is also notable to see the number of participants who sold their chickens to cover health expenses, education expenses, other household expenses, or to pay for food or grain. Participants sold an average of three to nine chickens to cover these expenses. Clearly, small livestock are used here as a form of currency or a savings account on four legs. In comparison to losing 15 chickens on average due to death, selling livestock for cash to pay expenses could represent a fairly significant financial loss if the livestock is sold earlier than expected or at a point when market prices are low. Understanding Resilience in Burkina Faso 16

23 Figure 5. Change in Chicken Ownership 50% 45% 40% 35% 30% 25% 20% 15% 10% 5% 0% Percentage of households that experienced change by reason given Average gain/loss The data on goats (including sheep and pigs) depict a similar story, as shown in Figure 6. Thirtyseven percent of participants reported goats dying at some point during the five surveys, with the average loss at four goats. Although there was no reported disease outbreak for goats, many get parasites and die from illnesses during the lean season (which is also the rainy season). Twenty-four percent of participants sold their goats to cover health care expenses; the average change was by four goats. There is also a link between the goats purchased and sold for the Tabaski holiday. 27 One family purchased 21 goats for fattening specifically for Tabaski and sold all 21. On average, one goat was killed by a household for the Tabaski celebration; however, the family that sold 21 brought the average up to eight for the entire sample. 27 Fête de Tabaski, also called Eid Al-Adha, is the most important Muslim holiday of the year, falling two months and 10 days after the end of Ramadan. During Tabaski, Muslims sacrifice a domestic animal, generally a goat (or other animals) and youth receive gifts. Understanding Resilience in Burkina Faso 17

24 Figure 6. Change in Goat Ownership 40% 35% 30% 25% 20% 15% 10% 5% 0% Percentage of households that experienced change by reason given Average gain/loss Generally, changes in cattle, donkeys and other were explained most by livestock giving birth or simply dying. Unfortunately, for cattle, some losses were not explained even though asked. In some cases, the numbers shared with the survey team seem to fluctuate slightly without reason; for example, they might report owning one cow at Survey 1, zero cows for the next several surveys and then one cow again at Survey 5, but without any explanation of changes. It is not clear why this happened, but selling large livestock such as cattle or horses can signal an economic hardship compared to the commonplace selling of chickens and goats. For example, several donkeys also died and only one family explained selling a donkey to cover education expenses. However, with chickens and goats, fluctuations were often explained. In the section later in the report on IGAs, it is noted that some women say that a feature of a non-resilient household is that it must sell large livestock to cover expenses. Perhaps one reason that some participants do not explain the reason for the change is because it is shameful, or even forbidden to discuss it, since it implies that she or her husband cannot adequately take care of their family. Understanding Resilience in Burkina Faso 18

25 Livestock Key Findings Households own very few cattle and other large animals such as horses and donkeys. On average, they own one to two, and these numbers do not fluctuate much over the five surveys. Birth and death of animals are reasons participants gave for changes in ownership of larger livestock. Households own several small (chickens and guinea fowl) and medium-sized animals (goats, sheep and pigs). Household ownership of these livestock fluctuates quite often even though average ownership (per survey period) stays fairly constant. Almost one-half of the households had chickens die during any time during the survey period, with 15 as an average number lost. Some women referred to a disease epidemic as the cause. Thirty-seven percent of households reported goats dying at some point during the five surveys; an average loss of four goats. Large livestock appear to be used mainly for labor and are usually only sold in times of economic hardship. Small and medium-sized livestock appear to function more as currency or savings accounts and are often sold to cover health, food, and household expenses. Food Security Survey 1, conducted the week of August 18, included two different food security instruments: a full food security survey consisting of 17 questions and a short, one-question food security scale that was applied at each subsequent survey. The full food security survey detected 91 percent of the study sample as food insecure (9 percent food secure), as did the short survey version, reinforcing that the short survey can detect food insecurity at similar rates as the full survey (for the purposes of this diary series). The data (in Figure 7) shows that over time, as lean season approaches, food insecurity increases and then begins to decrease during the months that align with harvest when households have more food to eat and money available from crop sales (October November). As of November, the rates of food insecurity were still high for this population, but improving. At all points during the five surveys, more than 70 percent of households were food insecure. These sustained high levels of food insecurity show a much higher level of vulnerability when compared to the poverty rates estimated earlier in the report where 64 percent were estimated to be living below the international $1.25/day international poverty line and half below the national poverty line. Additional analysis will help clarify whether those living below the international $1.25/day international poverty line and other two poverty lines are indeed those found to be the most food insecure. Related to the food security levels are the indicators of meat consumption in the prior three weeks as well as household perceptions of how their food security situation is changing. As food insecurity increases, more households report not eating meat during the prior three weeks; more also report that their food insecurity situation has worsened. As food insecurity lessens, fewer report not eating meat (so more are eating meat), and fewer say the food situation has worsened, with none reporting it had worsened by Survey 5. Understanding Resilience in Burkina Faso 19

26 Figure 7. Food Security over Time 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% Aug Sept 8-10 Sept 29-Oct 1 Oct Nov Food insecure in last three weeks Never ate meat during last three weeks Said food situation has gotten worse Food security is obviously an important concern and a significant hardship for these households. As has been previously shared (Table 1), these resilience diaries started during the lean season, which starts in July and runs through September. About 50 percent of the participants shared that they face a secondary hunger season that corresponds with the dry and hot season between March and May. This secondary hunger season is often due to harvests that were stocked during October and November being depleted, particularly if the previous harvest was not bountiful. Compounding the situation, it is during this period households also receive many more visitors than at other times of the year, thus further stressing their food stocks. To prepare for the hunger seasons, most mentioned using a variety of strategies to make the difficult time more manageable. Strategies included setting aside grain; diversifying income streams with activities such as increasing production of local beer or dolo, producing and selling shea butter, and gold mining; reducing their food consumption to stretch food stocks for as long as possible; and purchasing livestock to sell in the future to buy food for the family when needed. In the depths of the lean season, families respond by reducing the amount of food served at meals, and then later reducing the number of meals per day. They reward the younger children first with the majority of the food. Many mentioned gathering and consuming wild leaves from bean and sorrel plants. They will mix as many of these leaves as they can in the sauce that goes with boiled grain. They will also sell their livestock to purchase food, borrow money from SGs or others, and buy food and other items on credit from the merchants. While purchasing items using merchant credit is a common coping mechanism, a report by Trickle Up 28 reveals the risks of this strategy for all parties. The market is limited because its major characteristic is that transactions are carried out with acquaintances, resulting in sales on credit that then ends up compromising [the 28 Fofana, H The Experience of Women Living in Extreme Poverty in Burkina Faso: Passoré, Yatenga, Zondoma. New York: Trickle Up. Understanding Resilience in Burkina Faso 20

27 business] and eventually contributing to the failure of the business, which often can be the businesses of the women themselves as well as those they purchase items from on credit. They further indicate that to preserve their dignity in the village and the honor of their word, these women would otherwise bleed themselves dry to pay back the loan to a neighbor or retailer they regularly see. The participants were asked to indicate which coping mechanisms they used to enable them to purchase food when their food had run out. Figure 8 shows that all of them would resort to using their personal income before using their husband s income, followed by using the savings they have hidden at home, and then would sell small and medium-sized livestock. Note that half or less than half would use money held outside of the home, such as money with SGs (end-of-cycle savings or loans during the cycle), money in savings accounts or loans from financial institutions. However, it appears they will use loans and savings held outside of the home before borrowing from other people such as neighbors or other family members. Figure 8. Financial Coping Mechanisms Used to Buy Food Work day labor Remittances Delay payments on non-food items Borrow from other family Borrow money from neighbors, others Use savings held with a group Borrow from a financial institution Use savings held in an account Buy food on credit Use husband's income Sell assets Use savings held at home Use personal business money 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% What they do to buy food in reality, however, differs from what they say they would prefer to do. When asked to rank which two mechanisms they prefer or would prefer to use to purchase food, they prefer to use their husband s income first, then their own income, and then savings held at home. The only difference between the strategies they use versus strategies they prefer to use is the use of their husband s income. This suggests that while women will say that their husbands are responsible for covering food costs (like in the formative research findings), in reality they rely primarily on their personal income to pay for food. The next preferred methods, in descending order, are sell assets, purchase food on credit from merchants, use money from SGs and borrow from family. None indicated a preference for delaying payment on nonfood items. Only one mentioned borrowing from a financial institution as a preference. Understanding Resilience in Burkina Faso 21

28 When asked to explain their preferences, they said they prefer to use their husband's income first because he is in charge of the household, and then use her income when his does not cover all of the costs for food. Some women rely primarily on their income because they must cover costs when their husband is absent. When it comes to using cash versus selling assets, several indicated they prefer to sell an asset for food before using cash or savings because other needs, such as health care expenses, can come along unexpectedly and cash is needed to cover those needs. Therefore, they will sell assets to cover food needs before using savings. When they do not have enough money for food, they were asked to explain the nonfinancial coping mechanisms that their household applies. Figure 9 ranks the nonfinancial coping mechanisms that they use to enable them to find enough food to eat. Figure 9. Nonfinancial Coping Mechanisms Used to Access Food Take children out of school Feed working family members first Beg for food Send family to eat elsewhere Go entire day without eating Borrow food from family Consume conserved stocks of seed Eat cheaper foods Limit consumption of adults to permit children to eat Limit portion sizes Reduce number of meals per day Eat wild leaves 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Almost all report consuming wild leaves from bean and sorrel plants since these can be freely gathered locally, and reducing the number of meals per day. They mix the maximum number of leaves they can with their boiled grain. The third mechanism used is to limit portion sizes. As one woman shared, We think it is normal and easier to first reduce the number of meals consumed. If the crisis continues, we now will reduce the portion size (of the remaining meals). A few households (not represented in the graph) will also work as day laborers to enable them to bring something home to eat at night. It appears, given the importance of education of their children, which will be examined in depth later, all other coping mechanisms will be exhausted before removing their children from school. They will also reward their children with the majority of the food (bigger portions) when prioritizing food at meal times. Their preferences almost matched the strategies they reported using, as they preferred to consume wild leaves; however, they would prefer to reduce portion sizes prior to limiting the number of meals per day. The third and fourth preferred strategies were borrowing from family and buying cheaper food. Some also indicated they would borrow money before selling their grain in order to eat. Understanding Resilience in Burkina Faso 22

29 Seventy-two percent indicated they had never been turned down by someone when they asked for food. For those who had, some had asked for food or money from family and been refused because the family member was also suffering from hunger and could not help. Some asked for loans/money/credit and found that the lender was skeptical of their ability to repay and therefore did not provide support. The majority (70 percent) reported not receiving government assistance or support for food. For those who did, some indicated their children receive rice at school that is provided by the government. Some receive vitamin A-enriched flour for their children, and some have received subsidized grain for the past several years. Food Security Key Findings As lean season approaches, food insecurity increases and then begins to decrease during the months that align with harvest (October November). At all points during the five surveys, more than 70 percent of households were food insecure. About 50 percent of this population shared that they face a secondary hunger season that corresponds with the dry and hot season between March and May. Households use various financial coping mechanisms to buy food during the lean season with their personal income, husband s income, savings at home and livestock as the most common mechanisms used. The most common nonfinancial coping mechanisms to deal with food shortages included consuming wild leaves from plants, reducing the number of meals per day, limiting portion sizes, giving preference to feeding children and limiting consumption by adults, and eating cheaper foods. Education All women interviewed see education as an investment that provides their children with the skills and knowledge needed to succeed in life. However, education is also seen as a sort of a quid pro quo insurance policy. They invest in their children today so that their children can help them later in life. They foresee their children building them homes, paying health expenses and giving them food. As one respondent put it, "Nowadays, it is very difficult to succeed in ignorance, so the education of our children is important for their future and ours as well." Although the women talk highly about the importance of education, they often do not make the decision whether their children attend school. When asked who decides whether children attend school, slightly more than half (52 percent) said their husband makes (or made) the decision, 30 percent said it was a joint decision, and 17 percent said they made the decision themselves. Figure 10 shows the most commonly mentioned financial mechanisms used by the household to cover education expenses. Almost all women mentioned using savings at home, followed by their income, selling assets, then using the husband s income. However, as with food, they prefer to use their husband s income first, then household savings and then her income. Most households indicated that the husband s income is primarily used to pay tuition and the women s income covers other supplies and materials. When the husband s income does not cover the tuition expenses, they use her income to cover the difference. Several women mention they prefer to use income and savings over other financial options outside of the household because this affords them privacy of their household s internal management of money. Understanding Resilience in Burkina Faso 23

30 Figure 10. Financial Coping Mechanisms Used to Cover Education Expenses Use remittances Delay noneducation expenses Borrow from financial institution Borrow from family members Use savings account Borrow from others Use group savings Husband's income Sell assets Use IGAs Use savings at home 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% The participant households work hard to ensure that those children in school can remain in school. They are willing to sell assets, take out loans and help their husband make payments. They make sure their children eat before going to school and after returning home as well as encourage them to attend school. All participants mentioned they also felt their husband sacrificed for their children to go to school from paying school fees, selling assets, worrying about fees and going without food to save money for school expenses. It should be noted, however, that in the demographics section of this report, on average only two school-aged children were attending school, even though there were on average five school-aged children per household. A little less than half (46%) indicated that they or another member of their household have suffered from hunger to cover education expenses. For those who said yes, many of them mentioned selling grain, beans and millet that they were saving for consumption to cover education costs. We prefer to suffer from hunger than to have our children expelled from school, one woman shared. Education Key Findings All participants see education as an investment that provides their children with the skills and knowledge needed to succeed in life, and that later their children will invest in taking care of their parents. Commonly used financial mechanisms to pay for education expenses include, in descending order, savings at home, followed by using her income, selling assets (livestock), then using the husband s income. Although school fees are relatively high for most Burkinabe families, they work hard to make sure the children who attend school can remain there. On average, only two school-aged children are attending school out of an average of five school-aged children per household. Resilience The concept of resilience in this section could also be called shock-coping. This section covers positive surprises and negative shocks that households experienced, how women coped with the shocks either emotionally or financially, and how decisions were made with other household members regarding how to cope. Survey 1 asked whether in the prior year they experienced a Understanding Resilience in Burkina Faso 24

31 positive surprise and a shock, with 100 percent having experienced both; therefore, comparisons between points of time will only be made between Surveys 2 and 5 where the recall period for experiencing surprises and shocks was for the three weeks prior to the survey. Positive surprises Part of understanding resources, social networks and people s perception of unexpected events can be explored by asking about positive surprises events that have benefitted them or at least increased their happiness. In Survey 1, all women but one shared a positive event that occurred in the prior year. Figure 11 shares the most commonly reported gifts received. These include receiving money in the form of a remittance from family living abroad or elsewhere within Burkina Faso, gifts of money from family members living close by, higher income than expected due to improved business outcomes and more abundant harvests than expected. A few mentioned receiving household gifts such as food and sugar or receiving mobile phones as gifts. Positive surprises mentioned but not shown in Figure 11 include ones that may have been somewhat expected, but are still seen as a positive event: a woman giving birth to her child, the return migration of a husband, a wedding in the family, the inclusion of a son s new wife into the family, receiving a loan from a financial institution, having more in savings than anticipated and having a person repay a loan to them. If amounts of money were mentioned, the range of money received or earned ranged greatly, from a little under US$3 29 to $180, with an average of $43. The $180 received was related to the sale of a bean and peanut crop by an ODE member. Figure 11. Examples of Positive Surprises from the Prior Year Remittances Money Business money Good harvests Household gifts Mobile phone As a result of the positive surprise, most were happy to receive it and used the money to cover general expenses or business expenses, purchase food and invest in livestock. Some were simply grateful for the gift as it came at a time when they did not know how they would make ends meet. 29 Using an exchange rate of FCFA500=US$1. Understanding Resilience in Burkina Faso 25

32 In the subsequent interviews, when participants were asked to share whether they experienced a positive surprise in the prior three weeks, many fewer answered yes. Figure 12 shows, as one might expect, that the number who experienced positive surprises fluctuates over time. The types of surprises experienced were similar to those mentioned before people received remittances, greater income or harvests than expected (this particularly relates to the interviews from the November period), additionally some said none of their children being sick during the prior three weeks was a positive surprise. In addition to feeling happy because of the surprise, there is a greater sense of people feeling relief because of the struggle entailed with the deepening hunger over the first surveys. One respondent shared that the gift of rice by the survey team (this was a planned incentive used to encourage ongoing participation of the women in the 10 resilience diary surveys) was her positive surprise; as a result, they had a celebration in her family since the rice came at a time when they were particularly suffering from hunger. In the October and November surveys (Surveys 4 and 5), which were during and just after the time school fees were due, the women mentioned receiving remittances or gifts of money to help cover school costs. One woman mentioned that her brother took her children in to keep them enrolled in school as her positive surprise. Figure 12. Experienced Positive Surprise in Prior Three Weeks 45% 40% 35% 30% 25% 20% 15% 10% 5% 0% Sept 8-10 Sept 29-Oct 1 Oct Nov Shocks As with positive surprises, all of the participants reported experiencing a shock that occurred in their household in the past year. Figure 13 highlights the shocks reported, with illness and injury as the most common shocks, followed by loss of livestock (through death or theft), death of family members, and poor harvest. The category other comprised several singular reasons grouped together, including a divorce in the family, unexpected expenses due to the sale of a plot of land and a respondent s roof collapsing due to intense rains. Understanding Resilience in Burkina Faso 26

33 Figure 13. Shocks Experienced in Prior Year Poor harvest 4% Death in family 20% Other 6% Loss of livestock 22% Illness/injury 48% When looking at the trends in reported shocks over the period of the five surveys in Figure 14, including outcomes from Survey 1 (based on the past year), illness and injury represent the most common shock, by a significant margin. Illness and injury are experienced consistently, despite harvest occurring during Surveys 3 and 4, and the lean season ending. Deaths in the family appear to be a constant threat, consistently hovering at about 11 percent. Any one of these families is experiencing a death within each three-week period and therefore has increased costs from funeral expense contributions, as well as a possible loss of income or loss of other contributions from the deceased. Loss of livestock is also a consistent threat, with the averages ranging between 5 and 10 percent over the five surveys. Figure 14. Top Three Shocks Experienced over Time 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% Survey Number Loss of livestock Illness/injury Death in family Understanding Resilience in Burkina Faso 27

34 While not depicted in Figure 14, poor harvests and hunger are both mentioned in Survey 3 as shocks. Inability to cover school fees is mentioned as a shock starting in Survey 3 through Survey 5, at 2 percent, 17 percent and 11 percent, respectively, which aligns with children returning to school. When asked who made the decision as to how they would respond to the shock, Table 6 shows that on average, the respondent and her husband are almost equally making the decision on how to respond to the shock. Others are also playing a role in the decision-making, usually in reference to decisions being made about co-wives or other extended family members. Table 6. Decision-Makers for How the Household Responds to the Shock Average Percentage of respondents who makes decision 51% 48% 46% 35% 30% 42% Percentage of husbands who makes decision 45% 37% 35% 39% 46% 40% Percentage of others who makes decision 4% 4% 9% 9% 4% 6% Percentage of husbands and participants together who makes decision 0% 0% 0% 2% 0% 0.4% Given it was anticipated that husbands and wives might make decisions in some areas over others (for example, participants might make more decisions over health while husbands might make more decisions over livestock), an additional analysis was conducted to determine whether there were any consistencies to reports on how decisions were made. Table 7 shares the results of an analysis on decision-makers for coping with illness. This table demonstrates that the participant tends to make most decisions regarding her own health and illnesses of her children. However, her husband is obviously making many of the decisions as well. For others making decisions, as well as the husband making decisions about others illnesses, these are often referencing illness of extended family members, co-wives and their children. Interestingly, a few women mentioned that their own children, or even their neighbors, made decisions on how to respond to a shock regarding the participant s health. These were in cases where someone helped pay for transport to the hospital and to seek services (for example, when one woman injured her foot). Table 7. Decision-Makers for How the Household Responds to Illness Children's illness Participant's illness Other's illness Average Percentage of respondents who makes decision 52% 82% 30% 55% Percentage of husbands who makes decision 40% 9% 60% 36% Percentage of others who makes decision 6% 9% 10% 8% Percentage of husbands and participants together who makes decision 1% 1% A similar pattern emerges in Table 8 as it relates to coping with deaths in the immediate or extended family and others. Participants and husbands seem to equally make decisions (per the participant s perception) related to coping with death. While she makes slightly fewer decisions, she has almost equal responsibility as her husband. Often when she indicated making the decision, they were related Understanding Resilience in Burkina Faso 28

35 to the deaths of her parents and her extended family members (such as grandchildren, nephews, nieces, etc.). In one case, her making the decision was related to her own grief and not necessarily how the household would respond financially. It is important to note that when asked about who made the decision, the question was not specific to finances so it is not entirely clear whether she made a financial contribution decision, or made the decision about how the family would be supportive of the grieving families. Table 8. Decision-Makers for How the Household Responds to Deaths Death in immediate family Death in extended family/others Average Percentage of participants who makes decision 50% 45% 48% Percentage of husbands who makes decision 50% 50% 50% Percentage of others who makes decision 5% 5% Finally, decisions related to livestock loss were analyzed in a similar fashion (Figure 15). With livestock, husbands make the majority of the decisions, even when it was likely that the livestock were hers. As with issues of illness and death, it is not clear whether these were financial decisions or simply decisions on how to dispose of the animals or deal with theft. In either case, husbands made the majority of the decisions as to how the household would respond. In the few instances when participants made the decision in how to respond to livestock loss, it included decisions for both small and large livestock. Figure 15. Decision-Makers for How the Household Responds to Loss of Livestock Percentage of others making decision 11% Percentage of participants making decision 22% Percentage of husbands making decision 67% Figure 16 shows the top coping mechanisms to deal with any of the shocks mentioned earlier. The blue bar is the average number of times the coping mechanism was mentioned as having been used in the prior year to cope with shocks. This is in reference to the question asked in Survey 1 about shocks that occurred in the prior year. The red bar represents the overall average number of times Understanding Resilience in Burkina Faso 29

36 this coping mechanism was mentioned for shocks experienced in Surveys 2 5. Overall, the rating (numbers 1 4 in the figure) of each coping mechanism in terms of most frequently used is based on the past year average. This means that for the past year, savings held at home was the top coping mechanism for managing shocks, followed by reducing consumption, selling grain, selling small livestock, purchasing on credit and borrowing from an SG. However, one can see that for Surveys 2 5, saving at home and reducing food consumption are still the most common coping mechanisms. Households are using merchant credit and selling small livestock more often than selling grain, but this likely has more to do with the timing of Surveys 2 5 as the availability of grain to sell only became available towards Survey 5. Figure 16. Top Four Coping Mechanisms 4. Borrowed from savings group 4. Purchased on credit 3. Sold small livestock 2. Sold grain Average Surveys 2 5 Past year 2. Reduced food consumption 1. Saved at home 0% 20% 40% 60% 80% 100% Note that usng credit is not one of the top four choices, it actually is ranked number 8, following borrowing from family and friends. While not highlighted in the formative research summary earlier in this report, it was found that people don t typically like to take out credit unless it really is related to a business expense or a very large crisis for which they ve exhausted other means. To understand what might be behind the average use of the coping mechanisms mentioned above, it s important to look at the use of these coping mechanisms over time. Figure 17 demonstrates the use of the top four coping mechanisms. From the trends, while use of savings at home remains the primary coping mechanism, it decreases over time as we approach harvest in Surveys 4 and 5. Fewer are reducing their consumption, purchasing on credit, and selling small livestock. Understanding Resilience in Burkina Faso 30

37 Figure 17. Use over Time of Top Four Coping Mechanisms 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% 2 3 Survey Number 4 5 Saved at home Reduced food consumption Purchased on credit Sold small livestock When the sale of grains is added to the figure (Figure 18), there is a jump in use of grain sales as well as self-reporting of working harder as coping mechanisms for Surveys 4 and 5, corresponding to people being in the fields in Survey 4 for harvest, and post-field activities related to harvest in Survey 5. Figure 18. Use over Time of Top Six Coping Mechanisms 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% Survey Number Saved at home Reduced food consumption Purchased on credit Sold small livestock Sold grain Worked harder Finally, when just assessing the use of formal and informal financial services, Figure 19 demonstrates that as purchases on credit decrease over time, so does borrowing from friends and family. As harvest approaches, however, borrowing from financial institutions and from SGs increases at Survey 4, likely corresponding to the need for capital to pay field labor as harvest activities are in full swing. By Survey 5, use of any financial services falls below 30 percent at its maximum. Interestingly, Understanding Resilience in Burkina Faso 31

38 use of formal financial services never rises above 26 percent, even though half of the sample are RCPB members who are participating in credit-led village-banking services. Figure 19. Use over Time of Financial Services Coping Mechanisms 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% Survey Number Purchased on credit Borrowed from savings group Borrowed from friends and family Borrowed from financial institution These results suggest that proximity and convenience of resources greatly influence the use of the coping mechanisms available to them since the use of SGs and microcredit are not the top mechanisms used at any point of time. In fact, those who indicated they borrowed from a financial institution were all RCPB members, which suggests that the beginning of a new loan period may have begun. Also, the way credit was used was not always related to a shock. In some instances, the respondent did not report a shock, but still reported borrowing from a financial institution in the shock section of the survey, suggesting it is likely that a new loan cycle had begun (during Surveys 3 and 4). However, this was not the case with the report of borrowing from SGs. In fact, in Survey 4 where there is a peak in use of SGs for borrowing, the majority of those reporting borrowing from an SG are RCPB members, several of whom belong to SEMUS (who like ODE facilitates SGs). These results also demonstrate that as resources become available to them (for example, grain harvest), they reprioritize the way they manage any particular shock. For example, they appear to substitute the reduction of food consumption with the sale of grain when grain is available to help cover the costs of a shock. Future analysis regarding these decisions will likely confirm whether this is actually the case. When shocks occurred, most households reported feeling very saddened and stressed as the events have further impoverished them. As one client put it, My life curve remains constant. Another said, It hurts psychologically. Some feel ashamed in front of their creditors. The disappointment is palpable. However, by Survey 4 when harvest has begun, while households continued to face shocks, some felt less affected. One person shared, This has not affected us negatively. Indeed, for now, the situation has improved a lot because of the grain (harvest). Understanding Resilience in Burkina Faso 32