Agriculture global practice discussion paper Gender and Agricultural Risk A Gendered Approach to Agricultural Risk Assessments and Management Strategies April 2017 Agriculture Global Practice Discussion Paper Gender and Agricultural Risk A Gendered Approach to Agricultural Risk Assessments and Management Strategies April 2017 Cover photos: Courtesy of Flickr. Contents Acknowledgments v Acronyms and Abbreviations vii Chapter One: Introduction 1 Chapter Two: Agricultural Sector Risk Assessment 3 2.1 The Risk Management Framework 4 2.2 Why Incorporating a Gender Lens Is Important 5 Chapter Three: Gender-Differentiated Impacts and Responses to Agricultural Risk 7 3.1 Asset Endowment 8 3.1.1 Context-Specific Differences 8 3.1.2 Asset Ownership 8 3.1.3 Joint Ownership of Assets 10 3.1.4 Gender-Differentiated Use of Income and Wealth 11 3.2 Division of Labor and Responsibilities 12 3.2.1 Coping Strategies 13 3.2.2 Risk-Sharing within the Community 14 3.2.3 Degree of Labor Market Participation 15 3.3 Knowledge and Information 16 3.4 Gender and Resilience 17 Chapter Four: An Operational Approach to Gender-Based ASRAs 19 4.1 A Gender-Based Approach: Women’s Role in Supply Chains 19 4.1.1 Women in Production 20 4.1.2 Women in Post-Harvest Activities 20 4.1.3 Women in Transportation 20 4.2 The Risk Assessment Process Flow 21 4.3 STEP 1: Desk Assessment 22 4.3.1 The Output of the Desk Assessment 22 4.3.2 Crop Selection and Gender Focus 23 4.3.3 Analysis of Existing Material 23 4.3.4 Risk Quantification 24 4.3.5 Understanding Capacity to Manage Risk 26 4.3.6 Proactively Assessing Specific Vulnerable Groups 27 4.4 STEP 2: Stakeholder Interviews 28 4.5 STEP 3: Risk Prioritization 30 4.5.1 Prioritization Matrix 30 4.5.2 From Commodity Risk to Sector Risk 31 4.6 STEP 4: Gender-Smart Solutions 33 4.6.1 Understanding Gender-Based Constraints 33 Gender and Agricultural Risk iii 4.6.2 Identifying Gender-Smart Solutions 34 4.6.3 Applying Criteria to Prioritize Gender-Based Constraints to ARM 37 4.6.4 Integrating ARM and Social Protection 37 Chapter Five: Report Findings and Conclusions 39 References 41 Annex A: Risk Management Strategies and Instruments 45 Annex B: Analysis of Existing Material 47 Annex C: Methodology for Estimating Indicative Value of Production Losses in Agricultural Crops 49 Annex D: Guidelines for Assessing Risk and Capacity to Manage in Focus Groups with Farmer 54 Boxes Box 2.1:  Distinguishing Between Risks, Constraints, and Trends 3 Box 3.1: Household Surveys and Asset Ownership 9 Box 3.2: Review of Asset Ownership Worldwide 10 Box 3.3: The Legal Framework for Asset Ownership 11 Box 3.4:  Gender-Differentiated Effects of Drought on the Performance of Women’s Tasks in Pastoralist Households in Kenya 13 Box 3.5:  Illustration of Gender-Differentiated Responsibilities in West Africa 14 Box 4.1:  Questions and Checklist for Background Research for a Gender-Differentiated ASRA 24 Box 4.2: Gendered Line of Enquiry to Establish Capacity to Manage Risk 26 Box 4.3: Gender-Based Line of Enquiry for ASRA Fieldwork 28 Box 4.4: Gaps in Land Ownership and Control 36 Box 4.5: ARM and Social Protection 37 Figures Figure 2.1: Components of Arm 4 Figure 2.2: Arm Framework 5 Figure 4.1: The Sequential Asra Flow Process 22 Figure 4.2: Components of Risk Prioritization 30 Figure 4.3: Reprioritization of Risks from Commodity to Sector Level 31 Figure 4.4:  Illustrative Value Chain and Possible Entry Points for Removing Gender-Based Constraints 36 Figure A.1: Risk Management Layers 46 Tables Table 4.1: Illustration of Stakeholders’ Vulnerability Mapping in Livestock 27 Table 4.2: Illustration of Risk Identification in a Cotton Supply Chain 29 Table 4.3: Illustration of Risk Prioritization for Men in Livestock 30 Table 4.4: Illustration of Risk Prioritization for Women in Livestock 31 Table 4.5: Illustration of Sector-Level Risk Prioritization for Men 32 Table 4.6: Illustration of Sector-Level Risk Prioritization for Women 32 Table 4.7:  Summary of Gender-Based Constraints in Agricultural Supply Chains 34 Acknowledgments This report was prepared by a team led by Vikas Choud- and Lucia C. Hanmer (Lead Economist, GCGGR) from hary (Senior Economist) and comprising Sanna-Liisa the World Bank Group. Grateful to Amy Gautam for edit- Taivalmaa (Senior Gender Specialist) and Carlos E. Arce ing the document. (Consultant), all from the Agriculture Global Engage- ments Unit (GFAGE) of the Food and Agriculture Global This work would not have been possible without the Practice of the World Bank. numerous contributions from all the agricultural special- ists who participated in conducting Agricultural Sector The team is thankful to Preeti S. Ahuja (Practice Manager, Risk Assessments during the last eight years—there are GFAGE) and Marc P. Sadler (Adviser, GFAGE) for their too many to list them all. The team would also like to encouragement and support to address the challenges of extend its special appreciation to the stakeholders of agri- unpacking the gender differences in agricultural risk man- cultural supply chains in Africa, Latin America, and East- agement, and incorporating the findings into practical ern Europe who actively engaged in the development of tools while advising developing countries. The team also agriculture sector risk assessments in those regions, enrich- benefitted by early comments from Markus P. Goldstein ing this experience. Their active participation encouraged (Lead Economist, AFRCE), Niklas Buehren (Economist, the World Bank team to undertake further research and GCGDR), Asa H. Giertz (Senior Agriculture Economist, develop this paper. GFA13), and Eija Pehu (Adviser, GFAGE), who provided guidance on the concept and direction of the research. Generous funding for this activity was provided by the Dutch Ministry of Foreign Affairs, whose flexibility, Special thanks to peer reviewers for their thoughtful com- encouragement, and financial support allowed for the ments when revising various versions of this work. They gathering of significant experience that served as the basis are James Tefft (Senior Economist, GFAGE), Niklas for this work. Buehren (Economist, Gender Innovation Lab, GCGDR), 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, denomi- nations, 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. Gender and Agricultural Risk v Acronyms and Abbreviations ARM Agricultural Risk Management ASRA Agricultural Sector Risk Assessment FAO Food and Agriculture Organization of the United Nations GBC Gender-Based Constraint GDP Gross Domestic Product NGO Nongovernmental Organization UN United Nations Gender and Agricultural Risk vii CHAPTER ONE Introduction The objective of this paper is to develop an approach for integrating a gender dimen- sion into Agricultural Sector Risk Assessments (ASRAs). The focus is on the approach because both agricultural risk and gender issues are contextual—thus the outcomes of a gender-based risk assessment are highly dependent on the social, cultural, geo- graphic, and economic contexts of the setting in which the assessment is applied. Each ASRA must be tailored to capture gender differences that arise due to a country’s unique combination of risks, cultural differences, institutional arrangements, and fis- cal constraints. The target users of this approach are agricultural development practi- tioners in charge of making decisions about, planning, and/or conducting an ASRA and designing (and implementing) the corresponding risk management strategies. For over 10 years the World Bank has assisted developing economies in designing agricultural risk management (ARM) products, tools, and strategies. A wealth of expe- rience and lessons learned were gained over the last decade by conducting a number of ASRAs across several countries and agricultural commodities. This work extended and refined the understanding of agricultural risks, their impacts, and their transmis- sion across the sector, and the efficacy of different strategies to manage them. Cap- turing this experience, the World Bank published in 2015 “Agriculture Sector Risk Assessment. Methodological Guidance for Practitioners” on how to assess risks and develop risk management solutions in the agriculture sector (World Bank 2015a). While development practitioners welcomed that guidance, it failed to address issues related to gender differences and their implications for risk management. This paper addresses that omission. The conceptual framework and step-by-step illustrations presented herein are based on lessons learned by the World Bank and other institutions in assisting developing economies in conducting sector and key commodity supply chain risk assessments (World Bank 2015a). The paper also relies on recent work by international institu- tions and academic research that offer insights into the complexities of gender issues in relation to agricultural shocks. These experiences inform this paper’s case for a gender-based approach to ARM and help to illustrate a practical, stepwise approach for conducting gender-based ASRAs. Gender and Agricultural Risk 1 Many ways exist to assess agricultural risk, with various The organization of the paper reflects the challenges of degrees of sophistication and a variety of methods. The incorporating a gender dimension into ASRAs. Those conceptual framework adopted in this paper is highly challenges translated into answering the following ques- illustrative in nature. It (i) offers a checklist of issues to tions: (i) Is there evidence of gender-differentiated impacts consider to identify and assess gender differences while of agricultural risk and of gender-differentiated responses prioritizing risk, and (ii) serves as the basis for formulating to agricultural risk? If yes, then (ii) Is there a justification gender-smart solutions to agricultural risk. for conducting a gender-based approach to risk manage- ment? If yes, then (iii) How should it be done? The outcome of this work will complement the current operational ARM framework and toolkit applied by the As such, Chapter 2 offers a summary of the conceptual World Bank, International Fund for Agricultural Devel- framework for risk management, including addressing opment (IFAD), and other agencies, and will assist in why a gendered approach to ASRAs is essential. Chap- focusing interventions aimed at strengthening agricultural ter  3 shows the available evidence that agricultural risk systems’ resilience. This work is part of a wider initiative has a gender-differentiated impact on farming households led by the Agriculture Global Practice of the World Bank and that farming households adopt a gender-differentiated Group to develop analytical frameworks and tools for response to risk. This provides the justification for incor- assessing agricultural risks and to design risk management porating a gender lens in the operational approach to strategies in developing countries. This paper should be ASRAs. Chapter 4 describes a step-by-step operational considered a companion to the 2015 World Bank publica- approach for doing so, and includes guidance for formu- tion “Agriculture Sector Risk Assessment. Methodological lating gender-smart policies and strategies for strengthen- Guidance for Practitioners.” ing agricultural systems’ resilience. Chapter 5 presents a summary of findings and concluding remarks. 2 Agriculture Global Practice Discussion Paper CHAPTER TWO Agricultural Sector Risk Assessment Conventional analyses of agricultural value chains typically emphasize productivity and value addition, often through identification and removal of critical constraints or bottlenecks, and aim at efficiency gains and/or distributional purposes. This approach tends to ignore the issue of volatility in both production and markets and its effect on supply chain performance; it also does not address the incidence, allocation, or impli- cations of risk or how incentives to add value among different actors (i.e., smallholders, women, and small and medium-size enterprises) might explicitly alter risk patterns. Whereas the definition of risk implies the presence of uncertainty and the probability of losses (e.g., from a pest outbreak), the definition of constraint implies a certain fac- tor known to cause suboptimal performance in agriculture (e.g., poor soil) (Box 2.1). This is important because a risk management strategy aimed at reducing volatility will provide sustainability to agricultural investments targeting increases in productivity and efficiency. Both strategies are typically mutually reinforcing. The World Bank’s conceptual framework for ARM looks at risk from a broad sector perspective and considers it as a decision support tool for designing strategies to man- age risks in the agriculture sector. An ASRA is simply an orderly process to analyze, identify, and prioritize risk, and serves as the basis for the design of risk management strategies. The ASRA’s primary Box 2.1. Distinguishing between risks, constraints, and trends It is important to differentiate “risks” from “constraints” and “trends.” “Risks” are uncertain events that have the probability to cause losses. The element of uncertainty is present. As a symptom, yield volatility might be caused by a drought or a pest or disease outbreak. “Constraints” are conditions that lead to suboptimal performance. For example, low yield (symptom) might be caused by lack of access to inputs or poor technology. The element of certainty is present. “Trends” are longer-term or “chronic” patterns (reversible or irreversible) that provide context. For example, declining yield (symptom) might be caused by structural changes in agriculture or changes in climatic patterns (e.g., desertification). Source: World Bank (2015a). Gender and Agricultural Risk 3 objective is to assist policy makers to simplify and better a more dynamic flow around the needs of various stake- comprehend the complexity of ARM, by following a sys- holders, who are at the center of the process (Figure 2.2. tematic approach to prioritize solutions to mitigate, trans- ARM framework). fer, and/or cope with agricultural risk. Highlighting the types and orders of magnitude of risks and targeting the This conceptual framework has been applied in the con- most vulnerable stakeholders can improve planning and duct of risk assessment in various countries. However, it investments to strengthen resilience in agriculture. is imperative to disaggregate stakeholders, shown in Fig- ure 2.2 as producers and the commercial sector, to capture 2.1 The risk management gender differences in the analysis. The justification is based on the premise that distinctive gender differences arise in framework how agricultural risk impacts women and men, and in how A holistic perspective for assessing agricultural risk (Fig- women and men respond to risk (as shown in Chapter 3). ure 2.1. Components of ARM) needs to: Sex-disaggregated data can be captured, analyzed, and 1.  Understand the different tasks in the risk manage- incorporated in the various steps of the ASRA, including ment process how risk affects assets, livelihoods, and various stakehold- 2. Consider the full range of risks—production, mar- ers’ capacity to manage and recover from external shocks. ket, and enabling environment Developing gender-based solutions in agricultural supply 3. Engage all stakeholders affected by agricul- chains is a necessary strategy to address an increasingly tural risks—producers, commercial sector, and volatile global context and to open new opportunities for government strengthening the resilience of smallholder supply chains. 4. Analyze different strategies to manage risk— Annex A provides a more detailed explanation of inte- mitigation, transfer, and coping grated risk management strategies and instruments. 5. Suggest an action plan—including policy reforms, pubic investments, and knowledge transfer. However, risks and the capacity to manage them vary a great deal across countries, commodities, and regions. In More detailed explanations of these various components practice, ASRAs might involve greater or lesser attention to are presented in Chapter 4 in reference to the operational particular types of risk, depending upon the contexts and approach. objectives of the work. The composition and structure of the various supply chains comprising the agriculture sector As risk management is not a single step but a process, in a country and participation by gender in agricultural and as its components are not linear, they are better illus- supply chains will determine the degree of risk exposure trated as a cycle of tasks (Identification, Assessment, and for women and men participating in agriculture—as pro- Management) that interact with the other components in ducers, traders, input suppliers, and/or service suppliers. Figure 2.1. Components of ARM TASKS RISKS STAKEHOLDERS STRATEGIES INTERVENTIONS • Identify • Production • Producers • Mitigation • Policy reform • Assess • Market • Commercial sector • Transfer • Investments • Manage • Enabling • Public sector • Coping • Technical environment assistance Source: World Bank 2015a. 4 Agriculture Global Practice Discussion Paper Figure 2.2. ARM framework INSTRUMENTS Investments Technical assistance PRODUCERS Policy Men farmers Women farmers STAKEHOLDERS Producers COMMERCIAL SECTOR Commercial sector Men-led businesses RISKS Public sector STRATEGIES Women-led businesses Production Mitigate Market Transfer Enabling environment Cope Source: Adapted from World Bank 2015a. 2.2 Why incorporating tend to cope with risk differently given their asset endow- ments, their use of income and wealth, and the responsi- a gender lens is bilities they adopt within the house and the community. important Understanding the different roles and situations of women A better understanding of gender differences while assess- in their participation along supply chains in terms of their ing agricultural risk can allow for more comprehensive access to and control of productive resources, services, and effective agricultural resilience policies and interven- and employment opportunities is critical for assessing and tions, avoiding disruptions in rural livelihood strategies. prioritizing agricultural risk and for identifying gender- Understanding the root causes of gendered differences smart solutions for managing risk and strengthening resil- when households are facing risk is essential if risk mitiga- ience. Resilience can be strengthened in many different tion investments and risk coping programs are to reduce ways and at different levels through political, economic, rather than reconstruct people’s risk in future risk events. sociological, and technological interventions. Everyone operating in the agriculture sector faces many Women farmers typically have different constraints than types of risk that are often interrelated, including those men when facing risk, and the feasible options open to posed by markets and prices, policies, institutions, and women also differ. In certain contexts, social norms or production. Recent emerging evidence (discussed in barriers increase the complexity of the challenge for Chapter 3) indicates that women farmers are more highly women to mitigate risk or cope in a manner that pre- exposed to agricultural risks than men for many of the serves their agricultural and nonagricultural livelihood same reasons that farm productivity is lower for women strategies. For example, social norms may prevent women than men—namely, women have fewer endowments and from pursuing off-farm activities to diversify their sources entitlements, they have less access to information and ser- of income in the aftermath of an exogenous shock—and vices, and they are less mobile. Likewise, women and men Gender and Agricultural Risk 5 consequently influence women’s level of vulnerability, potential for women to benefit from risk mitigation strat- incomes, and ability to pay for the cost of diversifying egies could be overestimated, and the potential for men their farming practices. In some countries, only men have and households as a whole to benefit underestimated— the right to cultivate certain crops or to access markets, providing a misleading indication of what ARM strate- leaving women marginalized in terms of diversification gies can achieve. To truly capture and address agricultural strategies and market opportunities essential for manag- risk, incorporating gendered-based impacts and responses, ing risk. Another consideration is that many agricultural assessment of those differences must be an integral part practices that strengthen resilience require a high invest- of ASRAs and the findings incorporated in corresponding ment of time or labor (to build stone bunds and terraces, ARM strategies. for example) and thus are often not prioritized if women’s agricultural activities are undervalued. It is no longer justifiable to treat “farmers” as a homog- enous group of stakeholders. It is imperative to analyze Additionally, because of women’s primary roles in both the differential impacts and differentiated responses that natural resources management and in family care, they agricultural risks exert over women, affecting their liveli- also respond differently to shocks than men do. Women hoods and their capacity to recover from external shocks. bear the greater burden of agricultural risk in smallholder Interventions that are based on false assumptions or that farming, primarily through greater expenditures of effort: disregard the gendered dimensions of agricultural risk walking greater distances to fetch water, working harder will fail to strengthen the resilience of rural livelihoods. in the field because of changing weather patterns, and Recognizing these issues opens the possibility of designing developing diverse coping strategies to feed their families. ARM strategies that acknowledge who the most exposed Thus women should be key players in any strategy aimed and vulnerable are and how they may be affected. It at strengthening farming systems’ resilience. ensures development of more targeted, relevant, and early responses for those groups of individuals. If these prevailing differences across gender lines are not taken into account when assessing agricultural risk, the 6 Agriculture Global Practice Discussion Paper CHAPTER THREE Gender-Differentiated Impacts and Responses to Agricultural Risk ASRAs conducted by the World Bank to date have adopted an approach that places priority on high-value commodities and considers stakeholders in supply chains as the unit of analysis. The result is that greater attention has been devoted to the analysis of risk for high-value supply chains, which are mostly represented by men. Even when ASRAs focused expressly on domestic crops and livestock for reasons of food security, the emphasis was on farmers in general, without unpacking the differences that a gen- dered lens can provide. Some efforts have been made in the design of ARM strategies and policies to iden- tify and target specific exposed, vulnerable farmers. Farmers, however, are part of households made up of individuals, and an intervention may affect household mem- bers differently. It is important to understand how proposed interventions are likely to work and the contexts in which certain social, economic, or political factors may affect women and men differently, rather than affecting the household as a whole. Understanding how exogenous shocks impact women and men is a key issue to address if risk mitigation measures and risk coping programs are to reduce rather than reconstruct people’s risk in the agriculture sector. A review of the evidence shows that aggregate economic shocks do not have homogenous effects on farmers’ house- holds. Gender matters in explaining differential effects, in terms of both the direct (or first-round) effects of an economic shock and of households’ response strategies (or second-round effects). Moreover, these effects vary across countries and stages of development. Past empirical studies typically presumed that households behave as though they are single individuals, as the assumption of a “unitary household” is convenient and innoc- uous in many contexts (Udry 1996). However, a growing number of studies since the early 1990s have found strong evidence to the contrary (Strauss and Thomas 1994). These studies conclude instead that the aggregate demands generated by households Gender and Agricultural Risk 7 should be modeled as the outcome of some interaction 3.1.1 Context-specific differences between household members (women and men) with In a revealing piece of work for Uganda and Bangladesh, diverse resources, preferences, and responses. Quisumbing, Kumar, and Behrman (2011) found that while many shocks are similar in both countries, com- Gender relations are socially constructed under different monly experienced shocks do not necessarily have the geographic, cultural, political-economic, and social con- same effects across countries or on men’s, women’s, and ditions. Regarding agricultural risk, gender relations have jointly owned assets within countries. Land and assets complex social consequences for women and men that in general are relatively well-insured against food price present elaborate challenges for analysis in a single model increases in Bangladesh, but jointly held assets and wives’ or framework. However, it is critical to assess the gender- assets in Uganda are negatively affected by food price differentiated impacts of agricultural risk as well as the increases. Weather shocks negatively impact husbands’ clearly gender-differentiated responses for coping with it. assets in Bangladesh and wives’ assets in Uganda. Reflect- The empirical evidence around gender differences asso- ing differences in country and context, dowry and wed- ciated with three fundamental issues bolsters this asser- ding expenses take their toll on wives’ land in Bangladesh, tion: (i) asset endowment; (ii) the division of labor; and and illness shocks also have a large detrimental impact on (iii)  knowledge and information. Analysis of these three wives’ assets there, while death negatively affects wives’ issues illustrates why and how women and men within assets in Uganda. The small impact of weather-related farming households are impacted differently by and shocks on wives’ assets in Bangladesh relative to Uganda’s respond differently to agricultural risk under particular may reflect lower direct exposure to agricultural risk, as contexts. These three topics are the focus of the remain- Bangladeshi women rarely cultivate land independently. der of this chapter. Differences in the relative impact of shocks, and their 3.1 Asset endowment consequences for various assets depending on whether Responses to shocks and the ability to cope with vulner- men or women own them, show that responses to shocks ability are very much dependent on assets, and possession are context-specific, as are gendered responses to shocks. of or access to liquid assets is particularly important to avoid impoverishment. 3.1.2 Asset ownership At the individual or household level, shocks such as the Significant evidence on gendered vulnerability to shocks presence of a drought frequently result in asset sales. comes from studies that examine differences between Therefore, information on ownership of and access to male- and female-headed households. Kumar and Qui- assets is important to assess stakeholders’ capacity to man- sumbing (2014) found, for example, that in Ethiopia age risk. The findings can be integrated into risk man- and Bangladesh, after controlling for numerous factors, agement strategies to assist women’s acquisition of and female-headed rural households were more likely to control over key assets that can be used for coping with report a reduction in living standards or asset holdings as agricultural risk. a result of the 2007–2008 food price increases than male- headed households. An increasing body of literature shows that household welfare is not equivalent to the welfare of the individuals The next four subsections look more closely at the issue of within it (Haddad, Hoddinott, and Alderman 1997; Duflo asset endowment and its relationship to agricultural risk and Udry 2004; Sen 1990; Folbre 2001). Simply collect- vis-à-vis: (i) context-specific differences; (ii) the amount ing information on total household assets and dividing the and types of assets that men and women own; (iii) indi- total by the number of adults in the household presumes vidual or joint asset ownership; and (iv) differentiated uses that each individual has equal access to household wealth of income and wealth. All of these factors help to explain and will benefit equally from the fruits of that wealth. It how men and women are impacted by agricultural risk implicitly assumes that ownership and control over assets and how they respond to shocks. 8 Agriculture Global Practice Discussion Paper within the household will not affect decision making and household expenditures should be gender-differentiated outcomes. As numerous studies have shown, this is clearly based on asset ownership and asset control among the indi- not the case. viduals composing a household. Based on this evidence, important efforts are already being made to systemati- For women, a primary source of economic vulnerabil- cally measure asset ownership from a gender perspective ity is divorce or the death of a husband. Indeed, house- (Box 3.1). hold dissolution—whether due to divorce, separation, or death—is increasingly common (World Bank. 2008). In Women in many countries are far less likely than men to many countries, widowhood and divorce are associated own or control productive assets. In addition, women may with female poverty (Dercon and Krishnan 2000). To the not receive the benefits of assets held by men even when extent that assets provide economic security and a safety they live in the same household (Deere and Doss 2006). net, researchers have started trying to understand how Government policy, social norms, intra-family arrange- ownership and control of property are distributed among ments, and the market determine ownership and accumu- women and men and which individual members are bet- lation of assets. Gender biases in each of these different ter positioned to cope with changes in household assets institutions and practices limit women’s ability to obtain when facing agricultural risk. and keep assets. Women face greater risk of poverty and economic vulnerability than do men, and women’s lack of Evidence reviewed shows that men and women use asset ownership exacerbates this situation. income and wealth in different ways. A few studies have shown that household expenditures differ depending on A review of asset ownership worldwide provides an inter- the assets brought to marriage by each spouse (Fafchamps esting picture of the distribution of asset ownership in var- and Quisumbing 2005; Quisumbing and Maluccio 2003) ious countries.1 Box 3.2 offers the following global picture. and that the current asset distribution by sex affects house- hold expenditure patterns on food, health, education, and 1 Note that this is a rapidly evolving area of research and that these examples are household services (Thomas 1999; Katz and Chamorro for illustrative purposes to make the point of gender differences in asset owner- 2003; Doss 2006a; Quisumbing, Estudillo, and Otsuka ship and control of assets. For more precise statistics, please refer to special- ized agencies systematically evaluating these issues. In any case, ASRAs need to 2004). Therefore, the impact of agricultural risk on identify these differences within the context in which they are applied. Box 3.1. Household surveys and asset ownership Despite the existing evidence on the gender gap in asset ownership and wealth, the required information is currently not avail- able for the overwhelming majority of countries. This situation is in part due to knowledge gaps in preferred questionnaire design and respondent selection protocols for capturing individual-level data on ownership of and rights to assets. With this in mind, the United Nations Evidence and Data for Gender Equality (EDGE) project and the World Bank Living Standards Measurement Study (LSMS) program established a partnership in March 2014 to implement a randomized household survey experiment that documents the relative effects of different approaches to survey respondent selection and questionnaire design on individual-level measurement of ownership and control of assets. The household survey experiment, known as the “Methodological Experiment on Measuring Asset Ownership from a Gender Perspective” (MEXA), was implemented by the Uganda Bureau of Statistics (UBOS) during the period of May–August 2014 with in-country training, survey management, field supervision, data processing, and quality control support from the LSMS. MEXA was conducted successfully using the World Bank’s Survey Solutions Computer-Assisted Personal Interviewing (CAPI) software. The MEXA analysis yields an extensive set of findings that underlie recommendations for survey implementers in specific areas of data collection on ownership and control of physical and financial assets at the individual level. Subsequent phases of this experi- mental survey design detailing the experience and its applications in household surveys promise the construction of large national databases that will be extremely useful for risk analysis. Source: World Bank 2016. Gender and Agricultural Risk 9 Box 3.2. Review of asset ownership worldwide Land Ownership. Most data on asset distribution by sex in developing countries refer mainly to land, not surprising since in developing countries land is the most important component of wealth, especially in rural areas. Across the world, women account for less than 20 percent of the world’s landowners (UN Women 2012). A sizeable gender gap in land ownership exists in Africa, though the data on land ownership in Africa are hard to interpret, since much land is held collectively or is untitled. Recent data in FAO’s Gender and Land Rights Database show that women’s land own- ership is still irregularly dispersed. For example, women in Nigeria and Honduras own 4 percent of plots; in Nicaragua and Uganda, women’s ownership is as high as 20 percent and 18 percent, respectively (FAO 2010–11). The categories of private, communal, and state-owned land include a range of overlapping rights that add layers of complexity to any analysis of land “ownership.” Livestock. Livestock offer women a path toward additional income and food security (FAO 2013). Sales of meat, eggs, and milk can increase women’s decision-making and economic power, particularly important when coping with risk. In turn women fre- quently use the money to purchase food, household items, or medical treatment or to pay school fees that conceivably benefit the household (IFAD 2010). A general pattern is for men to own large livestock, particularly work animals, while women own smaller livestock and yard animals. Women face several challenges specific to their gender that constrain their ability to succeed as livestock keepers. In addition to dif- ficulties already discussed, access to training, extension services, markets, financial services, and occupational health hazards make it harder for women to improve quality of life by raising livestock (FAO 2013). Other inequalities include that women have less access to improved livestock species and enjoy fewer rights (ILRI 2010). Pastoral women are further marginalized even as they account for a necessary part of agriculture, raising stock in parts of the world where conventional farming is impossible. They build homes, tend to livestock, raise and educate their children, and manage domestic tasks such as cooking and fetching water and firewood. Women’s responsibilities prevent their mobility, so they are less likely to receive relief food or to venture out to look for work in towns (Anderson and Brouch 1999). In many developing countries women are responsible for milking and processing livestock products as well as selling them at market, but do not necessarily have control over money from sales. Methods to increase income and access must take into account the roles and responsibilities of women and men, social norms, and customary laws (IFAD 2012). Nonfarm Business Assets. Business assets usually provide a stream of income that provides security to the owners. Micro- finance programs throughout the world have focused on increasing women’s access to business capital for purchase of business assets. Notwithstanding the large amount of international research on microenterprises and informal sector businesses, little of this research has focused on whether a gender gap exists in business assets. In an early study in Ghana, Doss (2006b) found that although women are more likely than men to own business assets, the mean value of business assets owned by men is much higher than that owned by women. 3.1.3 Joint ownership of assets However, in general, it has been found that where men Most research clearly shows that men own the majority of and women have different rights to the same asset, men individual assets. The data collected by International Food tend to have more and stronger rights than women. For Policy Research Institute (IFPRI 2015) to identify lessons example, a wife often has the right to use her husband’s learned from eight worldwide projects show early interest- land (van den Bold et al. 2013; Gilligan et al. 2013). ing insights into the issue of jointly owned assets and reveal Women control milk for home consumption, but men a considerable amount of joint ownership of different control income from milk sales to collection centers (John- assets. A significant share of household land is under joint son et al. 2013). Women can use a pump but not loan it ownership, especially in Africa. It seems that joint owner- out to others without permission (Njuki et al. 2014). Even ship is even more important for livestock and consumer in examples where husbands and wives discuss what to do durables, with the share of jointly owned animals close to and make decisions together, when they cannot agree, it or even exceeding what is owned by men individually and is almost always the man who has the final say. In most always greater than that owned solely by women. cases, men feel they own all household assets by virtue of 10 Agriculture Global Practice Discussion Paper being heads of households. As one respondent in a focus risk coping strategy. However, family members differ in group discussion for a Kickstarter project said, “Men have their contributions to household revenue and in their the right to sell all assets, even those owned by women” control of its use. Where women and men have different (Njuki et al. 2014). But in the project, when a man sold an preferences, household expenditures will vary depend- asset without his wife’s permission, she was able to appeal ing on how control over income is distributed within the to a project authority for its return. When a woman’s household. name was on a land title, it could not be sold without her permission. These cases were possible because women A large body of evidence shows that in many parts of the were aware of their rights and had access to a means of world, men and women spend money differently: women defending them. are more likely to spend the income they control on food, health care, and education of their children (Haddad, Although societal norms govern the gendered distribution Hoddinott and Alderman 1997; Lundberg, Pollak, and of assets, often embodied in a legal framework (Box 3.3), Wales 1997). these are by no means immutable. ARM strategies can reinforce agriculture development strategies, thereby Using a longitudinal data set of 957 households in rural shifting the gendered asset distribution.2 Bangladesh constructed with 10-year survey intervals between 1996–97 and 2006–07, Quisumbing (2011) found that within a household, men’s and women’s non- Gender-differentiated use 3.1.4  land assets are drawn down for different types of shocks: of income and wealth wives are responsible for paying for illness-related shocks, Assets can serve as a buffer when households face eco- while husbands’ assets are drawn down to pay for daugh- nomic shocks. Assets can be used to smooth consumption ters’ dowries and weddings. This study also looked at the and prevent households from potential long-term impacts impact of shocks on land held by men, by women, and of shocks. Therefore, asset accumulation is an important jointly. Consistent with findings on non-land assets, hus- bands’ landholdings are drawn down with wedding and 2 A specialized reference on gender legal constraints can be found on the World dowry expenses, while wives’ landholdings increase with Bank’s “Women, Business and the Law” website. The website collects data on laws and regulations constraining women’s entrepreneurship and employment. deaths in the household (possibly due to inheritance). The dataset illuminates how government policies limit women’s full economic Jointly held household land—as opposed to individually participation through unequal laws and a business environment that does not owned land—appears to be better insulated from shocks. support the businesses in which women participate. Box 3.3. The legal framework for asset ownership National constitutions, civil codes, and legislation shape the policy context of women’s de jure property rights. While most countries no longer deny women ownership rights over assets, many national laws are still inconsistent with international legal frameworks, including the Convention for the Elimination of all Forms of Discrimination Against Women (CEDAW), to which most countries are signatories. The framework that determines women’s property rights—particularly of married women—is the combination of legal marital and inheritance regimes, sometimes referred to as family law. The legal marital regime defines the property rules governing assets acquired prior to or during the marriage. Legal marital regimes can be differentiated as to whether: (i) the assets acquired prior to marriage remain individually owned or are pooled to form community property during the marriage; (ii) the assets acquired dur- ing the marriage (through wages, salary, rent, interest, etc.) are in fact joint assets or owned individually by the person generating the income; and (iii) the assets inherited during the marriage belong to the individual or the married couple. The right to marital assets extends beyond the marriage and determines what happens to the assets upon dissolution of the marriage through divorce or death of a spouse. Source: IFPRI 2015. Gender and Agricultural Risk 11 More directly related to the rural context, evidence from example, in Uganda, 75 percent of agricultural producers Malawi and Uganda showed that women are likely to are women; in Ghana women make up 50–70 percent of spend more of their income on food compared to men the agricultural labor force, but earn less than 10 percent while men are likely to spend more of their income on of its income (Aduamoah-Addo 2016). In areas where assets than women. On average, women spend 23 percent migration and HIV (human immunodeficiency virus) are of their earnings on food and 14 percent on assets while affecting rural demographics, agriculture is becoming men spend only 8 percent of their income on food and feminized, as women increasingly become major actors 25 percent on assets (Njuki et al. 2011). In Côte d’Ivoire, in the sector. In many countries, women also play active better-than-average rainfall associated with high yields roles as traders, processors, laborers, and entrepreneurs, of women’s crops shifts their bonus expenditure toward despite facing many obstacles (compared to their male purchasing food (Duflo and Udry 2004). Asset owner- counterparts) in market access and bearing heavy respon- ship, in particular, is among the factors that may influence sibilities in their reproductive roles. However, the design women’s control over income and bargaining power in of many development policies and interventions contin- household negotiations (Doss 1999; Thomas, Contreras, ues to assume incorrectly that farmers and rural workers and Frankenberg 2002; Quisumbing and Maluccio 2003). are mainly men, who thus become the main recipients of programs aimed at reestablishing agricultural productive Gender-based differences in responsibilities for coping capacity in the aftermath of exogenous shocks, leaving with agricultural shocks have several implications for women as recipients of food aid. long-term asset accumulation. For example, men have a long time to accumulate the assets needed to pay for a The results of a study conducted by Wawire (2011) on dowry or wedding, which are anticipated events. In con- gender roles and risk in the Turkana District of Kenya cor- trast, food shortages come on quickly and unexpectedly; roborated those of other studies on the gender division of dealing with them is the responsibility of women. As a labor: women’s roles revolve around the homestead while response to agricultural risk, women are more likely to those of men feature outside the house. However, Turkana spend the income they control on food, health care, and women’s roles go beyond the homestead to include roles education of their children, using wealth and income for in livestock production and cultural activities. The effects smoothing consumption at the onset of crisis, whereas of drought and resultant famines pose challenges to both men’s assets are not drawn upon until the severity of a men and women in Turkana District, but as a result of catastrophic event merits their use. Thus men and women their prolonged stay in this harsh environment, residents may have different capabilities and roles in managing risk have devised coping strategies. Box 3.4 addresses some and coping with shocks. details of the findings to illustrate how men and women face risk in this particular context. 3.2 Division of labor and responsibilities Evidenced gathered for a World Bank (2012a) report on the 2009 economic crisis showed that women con- In many parts of the world—for example, Sub-Saharan tinued to bear the burden of household responsibili- Africa and South Asia—although women are the main ties despite experiencing longer hours in paid work and farmers or producers, their roles are largely unrecog- income-generating activities. In many communities, nized and often escape the objectives of public policy. interviewees noted that when the time women spent in The growing proportion of women in agriculture is one income-generating activities was added to the time they of the most striking trends of recent times, and a large spent cooking, cleaning, caring for children, and handling body of literature has debated the “feminization” of other household needs, women worked much longer hours labor markets. Women account for around 40 percent of than men. This was particularly apparent in the Central Africa’s agriculture labor force (World Bank and the ONE African Republic, where an interviewee noted, “It is the Campaign 2014), but this figure varies by country. For sleep which drags her away from her daily housework,” 12 Agriculture Global Practice Discussion Paper Box 3.4.  Gender-differentiated effects of drought on the performance of women’s tasks in pastoralist households in Kenya Obtaining water: Interviewees revealed that obtaining water for household use and for drinking purposes is very difficult during the drought period. Lack of rain means the rivers dry up. This creates competition for the available water sources, which are usually so far that women and girls have to walk long distances to get to them. They walk up to 2–10 km in the Turkwell area and 30 km in the dry Kapua area. To beat the long queue at the water points, women start the trip before dawn. This poses security problems because they may be attacked by bandits or rapists on their way. To counteract this, women walk in groups. Transporting the water for long distances is also a problem because women carry the water on their heads in heavy traditional wooden troughs originally meant to be carried by donkeys, before the drought killed them. The few people using boreholes/shallow wells face similar short- ages because the wells dry up due to too many users. If they break down, they are not repaired because the majority cannot afford the maintenance costs. Provision of fuel: Women have to walk long distances to get firewood because of slow tree regrowth during the dry season. Construction of houses: The construction materials (twigs and leaves) are from a wild plant called egol (dumb palm). Construction becomes tedious during the dry season. Women have to keep dismantling, loading, unloading, and reconstructing new houses as the migrations become more frequent. Provision of food for the family: Food becomes scarce during the drought period. Women have to rely on alternative sources of food since the usual foodstuffs (milk, meat, and blood) are no longer easily available. Wild fruits are found in the fields while maize meal is bought at the markets or supplied by famine relief agencies. Even those living near the river are not better off because it dries up. Farming activities are reduced to small plots along the river where water forms pools (amokolol). Alternatively, people use boreholes. Consequently, only vegetables are grown on a small scale. Most women complain that cooking the “new” kinds of goods (maize and beans) is more time- and fuel-consuming compared to cooking milk, blood, and meat. Provision/decision making: Men feel that their role as heads of households is adversely affected during the drought period because of reductions in family resources. There is no money to provide food, clothing, and even school fees for family members. There are therefore fewer chores to distribute and delegate. Most of the men interviewed feel that women’s status in the family is uplifted because they receive famine relief food. Many men interviewed feel threatened by this because their role as providers is undermined by the relief food controlled by women. This creates conflict in some families. Provision of water for livestock: Women are responsible for provision of water to livestock left at the homestead; they obtain this water from boreholes. During the dry period, the water table of the wells lowers and thus women have to scoop these wells almost on a daily basis. The wells are as deep as 10 meters. This task is laborious and dangerous because the wells may collapse. Apart from providing water for the stock left behind, women must take on additional roles because most of the time men migrate in search of livelihoods. Women have to perform the role of household heads. Procurement of pasture and water: This role is performed by men. Water and pasture become scarce during the dry season, which necessitates migrating to new locations and walking for long distances. Men’s role as security providers intensifies during this period because insecurity mounts as the movements become frequent and long, especially from the neighboring Pokots. Even in homes, security needs intensify during drought. Source: Wawire 2011. and in Kenya, where rural women reported 18-hour observed worldwide in the various case studies conducted working days. for the abovementioned World Bank report (2012a).3 In Bangladesh, women commonly reported that they ate after everyone else was fed, even if it was not enough food 3.2.1 Coping strategies Reducing the quality of food and the number of meals is the most typical behavior-based coping response— 3 During sudden food crises, women, children, and infants can become quickly malnourished, according to the World Food Programme (2016). Women and and often the first one used—for households responding WFP. Helping Women Help Themselves. WFP website: https://www.wfp.org/ to an economic crisis. This specific coping response was our-work/preventing-hunger/focus-women/helping-food-assets. Gender and Agricultural Risk 13 Box 3.5. Illustration of gender-differentiated responsibilities in West Africa In West African cocoa farm units, men and women earn separate incomes within the same household. Men control most high- income cocoa crops, while women cultivate smaller plots of consumption crops, which provide less income. Money collected from cocoa crops is not always shared equally among household members, but women and men have specific obligations to pay for ordi- nary expenses in support of the household. This varies within each home, but men (high earners) typically pay for large expenses and women (low earners) pay for smaller expenses. However, because women tend to have less income to apply toward necessary expenditures and therefore less excess money to retain, they have less power in decision making and bargaining, and household food security is negatively influenced. Source: Kiewisch 2015. for nourishment. Gender-based distribution of food is long-term consequences for children’s nutrition and cog- a coping strategy commonly used to handle food crises nitive development as well as for economic outcomes, thus (Lambrou and Nelson 2010). weakening households’ resilience (Alderman, Hoddinott, and Kinsey 2006). Reducing nonfood consumption (e.g., of soap and coal), working more hours, and diversifying sources of income 3.2.2 Risk-sharing within (e.g., by entering a new informal occupation) are common the community reactions nearly everywhere. Migration is also prevalent, The 2012 World Bank study found that assistance for sometimes including reverse movement to the home area women often comes from relatives, friends, and neigh- and migration of men toward urban centers as they seek bors. Relief frequently comes from informal groups orga- work. Women respond in general by expanding their nized around mutual solidarity, often along occupational duties and reducing consumption. lines. The mere idea of belonging to informal community groups seems to make people feel stronger and helps them Women’s subordination and economic insecurity (e.g., get through hardships. Moral and financial help from reli- contingent labor, home and care work, lack of credit and gious organizations is common, but help from nongovern- savings, etc.) are factors that explain not only their differ- mental organizations (NGOs) is less frequent. NGOs had ent vulnerability to risk but they also underlie how women a presence in additional study sites, but did not always aid respond differently to regain the initial pre-shock position. in a form that the respondents perceived as important or In many countries, women are often confined to the pro- helpful for coping. Government assistance was important duction of subsistence foods, left with little choice but to for coping only in sites in the former socialist countries of respond to risk by lowering consumption. Box 3.5 illus- Eastern Europe, Central Asia, and Mongolia—countries trates another example of gender-differentiated responsi- that had large national social protection programs in place bilities within households in West Africa. prior to the crisis (World Bank 2012a). More specifically, in comparative studies across male- and Kumar and Quisumbing (2014) found that women look female-headed households, Kumar (2014) showed that outside the household for insurance mechanisms. It is female-headed households are more susceptible to a shock argued that women share their response to risk with other like a rise in food prices and tend to eat less preferred foods women in the village while men have a wider and less and cut back on quantities served, because they are gener- defined approach to risk-sharing. Indeed, cash transfers ally the member responsible for providing food. from the spouse and the extended family seem not to be Such coping strategies in severe situations, especially responsive to shocks, but those from nonfamily friends are. for pregnant and lactating women, can have adverse 14 Agriculture Global Practice Discussion Paper Women’s behavioral responses to responding to shocks Though much of the existing literature has focused exclu- (tightening their household responsibilities and reduc- sively on urban areas, Lim’s (2000) study of the Philip- ing consumption) place them in the most vulnerable of pines showed that women’s entry into the labor force is situations. To the extent that women often resort to these much more marked in urban areas; however, in rural mechanisms, it is important to recognize that these coping areas, women’s overall labor force participation has not strategies have long-term effects on the nutritional status increased but their work hours have increased significantly. of children, and thereby reduce resilience. The absence In fact, women’s work hours seem to have replaced men’s of appropriate social safety net policies and ARM strate- work hours in agriculture. Similarly, exploring state-level gies compounds this problem. data in India, Bhalotra (2010) found that recessions are associated with an increase in rural women’s labor sup- ply. She argued that this result suggests the dominance of 3.2.3 Degree of labor market participation the added worker over the “discouraged worker” effect in Shocks can induce rural household members to increase rural India; that is, economic downturns increase house- their participation in the labor market if household hold poverty. responsibilities and labor market conditions allow for it. In Similarly, Kumar and Quisumbing (2014) reported that times of economic crisis, women typically enter the labor shocks can induce households to increase their labor force (the “added worker” effect) in response to declin- supply to compensate for the increased expenditure or ing household income (a second-round effect). This effect reduced income caused by the shock (Berloffa and Mod- appears to be particularly strong in lower-income house- ena 2009). In some contexts, cultural and gender role bar- holds and in lower-income economies, where an informal riers restrict women from entering the labor market. Even labor market or a rural sector can absorb additional work- when women can enter the labor market, multiple fac- ers (World Bank 2012b). tors do not work in their favor. A gender wage differential Gender and Agricultural Risk 15 exists in labor markets in the developing world (whereby Because women are more likely than men to leave their women get a lower wage for the same job) and women are natal villages to marry in many countries, marriage, as often subject to sexual and physical abuse (Garcia, Her- well as marriage-related migration, has a prominent role nadez, and López-Nicolàs 2001; Hinks 2002). Increased in resilience (Kumar 2014). labor supply by women in response to a shock, whether in the local labor market or outside, can have significant 3.3 Knowledge implications for children in these households, particularly for adolescent girls, who then have to take on domestic and Information responsibilities (Holmes, Jones, and Marsden 2009). The issues of knowledge and access to agricultural tech- nology have been brought forward as key determinants In crises, women’s small-scale trading networks can be of gender gaps in agricultural productivity, recently ana- damaged, thus reducing a principle source of income. lyzed and summarized in a World Bank (2014) research Conflict situations, however, can limit men’s and open publication for six Sub-Saharan African countries. The women’s access to markets. Men may flee, join armed same reasons that explain the gender gap in agricultural groups, face imprisonment, or be killed during conflicts, productivity in that report can also explain women’s lower which puts women under greater pressure, even exposing capacity to manage risk. Although the intention is not to them to sexual violence and abuse. generalize the findings of one country to others, this sub- section presents some highlights of the findings that are In extreme cases, conflicts and shocks increase women’s important for ASRAs. use of transactional sex as a risk-coping mechanism. Recent research in rural Tanzania showed that shocks The World Bank (2014) report maintains that knowledge lead to a tripled increase in paid sex and that as income and training on farming methods and techniques are criti- goes down, sexually transmitted infections (STIs) surge cal for all farmers, but that it is particularly important to (de Walque, Dow, and Gong 2014). target women farmers. Women farmers tend to receive second-hand information from husbands and friends if Gender-based violence is another form of bias limiting they are not the head of their household, may not attend women’s opportunities and how they respond to crises. field training activities due to household responsibilities or Violence against a woman takes many forms. It includes mobility constraints, and may not be able to interact with but is not limited to physical, sexual, or psychological male extension agents due to cultural norms. harm. Women can be subjected to violence within the family, from the larger community, or as state-sanctioned For example, women farmers in Tanzania receive fewer abuse (WHO 2009). Gender-based violence often results extension services than men, and this difference contrib- from unequal power relations between men and women. utes to the country’s gender gap: women produce 14 per- cent less per acre than men. Similarly, extension services Migration is a coping strategy often used if the local labor do not lead to the same returns for women farmers in market opportunities are not sufficient or perceived ben- Ethiopia, who produce 23 percent less per acre than men, efits from migration outweigh the costs of entering the or in Uganda, where women produce 13 percent less per market locally. Migration may increase resilience for both acre than their male counterparts. This evidence sug- origin households and migrants. People may move to gests that these services are less effective for women or are pursue better opportunities but also to escape economic, poorly attuned to their needs. political, or social distress. Migration benefits the origin household because of potential remittances, but more Informal social networks play a critical role in the immediately because there is one less mouth to feed during exchange of agricultural information and the adoption hard times. The family’s choice of a migrant is gendered, of agricultural technologies among farmers. Cultural with families investing in different children’s migration. norms, such as restrictions on women’s interactions with 16 Agriculture Global Practice Discussion Paper men outside the household, as well as time and mobil- Second, women in the roles of caretaker and food pro- ity constraints may limit avenues for female farmers to vider for their families adopted alternative strategies for access public extension and formal agricultural informa- food and goods resourcing and for childcare. Women tion services. Women’s networks tend to differ from men’s spent longer hours obtaining items needed to sustain (for one thing, their networks are smaller), and research families, and bore the stress of helping children cope with suggests that women may rely more heavily on them for the new reality. While not directly related to agriculture, accessing agricultural information, particularly from this global analysis highlighted and recognized the central other women. significance of women in enabling the “resilience” with which societies managed this crisis. Across the countries profiled in the World Bank (2014) report, women’s lower levels of education hamper both Existing surveys and monitoring mechanisms are not likely their access and returns to agricultural resources. The to routinely measure these pressures exerted on women. gender gap in human capital observed today is partly Thus policies and strategies need to recognize them and due to women’s lack of access to education in previous gender-differentiated interventions designed to strengthen decades. Although girls’ school enrolment rates have the capacity of various members of a household to absorb increased markedly, offering the promise that future gen- shocks. In agriculture, strengthening women’s resilience erations of women farmers will not face the same obsta- to shocks and economically empowering women need to cles to productivity, today’s adult female farmers continue be taken into account in key risk management policies to have lower education levels. and strategies, and gender-smart solutions to agricultural shocks must be implemented. 3.4 Gender and resilience Overall, the main policy and operational implications are The World Bank (2012a) analysis of the 2009 economic that gender-differentiated impacts to shocks, the causes of crisis observed that women played a major role in shock differentiated responses, and the effects on well-being and absorption, possibly more so than during previous crises. resilience need to be identified. A gendered assessment Researchers offered two reasons for why women shoul- of agricultural risk will fill an information gap and help dered the consequences of the global crisis and helped to inform policy and better target the operational tools foster community resilience. First, the financial crisis and interventions designed to improve ARM practices in heavily affected the formal job sector, which employed a developing countries. These are precisely the objectives of large portion of women. Women lost their jobs, and as the a gender-based ASRA. recovery started, those positions did not re-emerge. Gender and Agricultural Risk 17 CHAPTER FOUR An Operational Approach to Gender-Based ASRAs Chapter 4 provides practical guidelines on how to integrate gender issues into the pro- cess of conducting ASRAs as advised by the World Bank and other development part- ners in developing countries. It first identifies the existing gaps in taking into account gender differences in the agriculture sector. It then addresses the various entry points in the process of risk assessment where a gender lens could be introduced in a practical way. It next presents the issues of vulnerability and understanding capacity to manage risk. The chapter ends with examples of best practices related to gender-based solu- tions to risk management that illustrate the design of gender-based risk management strategies. 4.1 A gender-based approach: Women’s role in supply chains The end product (output) of the ASRA process is a set of actions that require special attention to reduce vulnerability to shocks associated with the identified key priority risks. As resources are scarce, decision makers need to see an explicit risk prioritization and strategy identification process. The risk prioritization helps justify the proposed interventions in terms of reduction of agricultural income volatility, consumption smoothing, food security, protection of vulnerable stakeholders, and agricultural resil- ience. The measures can then be incorporated into government plans and budgets as part of government agricultural policy and strategies. Risks faced by agricultural stakeholders can be classified primarily into three catego- ries: production, market, and enabling environment risks. Depending on the market integration of any particular supply chain and its context, each type of risk can be present, dominant, or absent. Each can also affect unique segments of the supply chain or the entire chain. Many women farmers are more vulnerable to agricultural risk compared to men for many of the same reasons that farm productivity is lower for women than men, as discussed in Chapter 3. Gender and Agricultural Risk 19 The types of risk that impact women and men most and upgrade value chain performance and build input mar- their individual (or joint) capacity to manage them are best kets, benefiting women and input supply companies at the assessed in this framework by analyzing each agricultural same time. supply chain individually. Women participate in various modalities along the different phases of value addition in 4.1.2 Women in post-harvest agricultural supply chains, from farm to fork. Many face activities constraints and risk in ways that are very distinct from While women are generally, though not universally, those of men. Recent research by the International Finan- responsible for key processing activities, specific roles in cial Corporation (IFC 2016) illustrates the dimension and post-harvest and storage seem to be highly variable across diversity of activities performed by women in agricultural regions and supply chains. However, a few cross-cutting supply chains in developing economies, as follows. lessons emerge. Research has found that women are more likely to participate in processing activities as employees Women comprise over 40 percent of the agricultural labor of larger firms, rather than as individual entrepreneurs. force worldwide as farmers, entrepreneurs, and laborers, Also, where post-harvesting activities are not mechanized, and are significant contributors to agribusiness supply they are more likely to be carried out by women. chains. In addition to women’s roles and employment on large commercial farms, the perspective of women as Post-harvest processing is typified by high levels of varia- small-scale farmers needs to be brought to the frontline tion between men and women in different commodities, of analysis. Women’s activities in agricultural production even within the same region. For instance, in Bangladesh, vary greatly across commodities and regions. For example, women provide 5 percent of the labor in harvesting and Indonesian women provide the majority of the labor in threshing for rice, while in Assam, India, women provide rice farming, but less than one-third of the labor for rub- 60 percent of the labor. For other post-harvest activities, ber (IFC 2016). However, women are oftentimes paid less Bangladeshi women provide 51 percent of the labor, while than men for the same work, and are overrepresented in women in Assam provide 90 percent. A gendered supply informal, unpaid, part-time, and seasonal work. Because chain mapping should pay particular attention to the divi- women are an important labor source in agricultural pro- sion of labor within each stage. duction, leveraging their potential and providing them with risk management tools like access to assets, as well as As a rule, women’s activities can be divided into (i) energy- training, land, and inputs, can help the agriculture sector heavy activities, with prominent ones including milling increase productivity and better manage risk. and de-hulling of grains and walking with loads, and (ii) time-heavy activities, with prominent ones including The following subsections present an overview of the walking, waiting, and manual milling. In particular, the many ways women participate along supply chains, as major staple crops (maize, paddy, sorghum, millet, and producers, in post-harvest activities, and to a lesser extent, cassava) constitute a group of core crops for which pro- in transportation. duction and manual processing are significant, with a high involvement of women in post-harvest activities. Also, as 4.1.1 Women in production a rule women’s access to storage tends to be lower than Women play a variety of roles in input provision and use. that of men because of the required access to transport They are active as small-scale farmers for own consump- and financing. tion, selling in the market, and providers of agricultural inputs to agribusinesses. They act as agro-input retailers 4.1.3 Women in transportation and agro-dealers, and they are hired as extension workers Women in agribusiness are likely to be excluded from and rural agro-agents. Women’s activities are well-suited transportation of goods to market or from marketing or to assist input supply companies and enable an effective sales of goods, even when women are the main produc- and wide reach of companies’ products to large con- ers of those goods. Where women are involved, their sumer markets. Women represent significant potential to 20 Agriculture Global Practice Discussion Paper sales opportunities are more likely to be confined to local agricultural supply chains. Given the complexities of markets rather than regional or international ones. This agricultural production, processing, and commercializa- results in poor access to networks and is reinforced by tion, there is no shortage of risks and potential solutions infrastructure and trade systems that tend to inadvertently to manage them. Moreover, effective risk management disadvantage women. generally requires close cooperation between the various actors involved in different political economy scenarios, 4.2 The risk assessment who usually defend their particular vested interests. The process flow facilitation process requires risk assessment teams to assume a neutral position and play the role of “honest bro- An ASRA is devised as a consultative and time-bound ker” among parties, with the explicit mandate to answer process to be carried out over an estimated 12-month the following questions in a gender-based approach: period. This rapid assessment draws upon available quan- 1.  What are the key agricultural risks faced by women titative data and qualitative information collected through and men, and who is mostly affected? stakeholder interviews and dialogue. 2. What are the optimal and practical solutions to manage women’s key risks? The ASRA is considered as the process of identifying and 3. What are the gender gaps in current ARM prioritizing the major risks that typically explain (depend- strategies? ing on government objectives) agricultural gross domestic 4. What is the action plan to strengthen women’s and product (GDP) volatility, food insecurity issues, and vul- men’s resilience to shocks in agriculture? nerability of the rural poor. In the past, this same process was established and applied to assess risk for agricultural An assessment team typically comprising between three to enterprises in various countries. The approach proposed four agribusiness specialists is needed to conduct the over- herein adapts the process to include gender-smart solu- all ASRA. Each specialist is required to assess and prior- tions as part of the formulation of ARM policies and itize risk and corresponding solutions in particular supply interventions. Thus for each of the supply chains under chains, including answering the questions stated above, and assessment, it is important to: to follow the gender-based approach outlined in this paper. 1.  Understand very well the institutional and socio- The team in a subsequent step aggregates the findings of economic context of women participating in individual supply chains into a sector-wide assessment. A agriculture gender specialist is needed as part of the core assessment 2. Identify and assess women’s and men’s current team to lead and guide the various members of the team risk management practices and their roles and on the process and outcomes regarding gender. constraints 3. Assess women’s and men’s capacity to manage risk The ASRA process is basically an approach whereby the 4. Identify a package of solutions in consensus with risk assessment team establishes the guidelines and sys- stakeholders that will strengthen stakeholders’ tematically facilitates the discussion among stakehold- current risk management practices. ers to prioritize risks and corresponding solutions. This facilitation also includes the collection and processing of The result of an ASRA should be a set of practical risk qualitative and quantitative information that informs the management measures that stakeholders agree upon and assessment process and forms the basis for discussions that will contribute to a more resilient agriculture sector. among stakeholders. Early consultations with stakeholders will enhance own- ership of the process and facilitate incorporation of the Figure 4.1 shows a simplified step-by-step version of an recommendations into government plans. ASRA process. It starts with data assessment for identify- ing and assessing risks for major individual commodities’ Risk analysis can be complex since it involves assessment supply chains followed by interviews with key stakeholders. of several stakeholder groups participating in various It uses the findings to build a risk prioritization matrix at an Gender and Agricultural Risk 21 Figure 4.1. The sequential ASRA flow 4.3 STEP 1: Desk process assessment The first step of the ASRA process is to gather back- Risk Prioritization ground information on the agriculture sector, collect rel- evant data (quantitative and qualitative) for risk analysis, and conduct a desk-level assessment before stakeholder Step 1: interviews in fieldwork. The main objective of this step is • Risk profile Desk data to understand the structure and dynamics of the agricul- assessment • Risk quantification ture sector, to identify major risks for agricultural supply chains, and to obtain the information to start unpacking Step 2: and quantifying the causes of risk. A gender-based back- Stakeholders ground assessment would include, among others: interviews 1.  An understanding of the relative weight, role, and structure of the sector within the broader economy Step 3: • Frequency of events 2. The composition of and stakeholders’ participa- Prioritization • Intensity of events tion in the main supply chains, differentiated by matrix • Capacity to manage gender and region • Special vulnerable 3. Familiarization with agricultural policies, strate- farming groups Step 4: assessment gies, and programs—including those addressing Gender vulnerability issues smart solutions 4. Identification of historical agricultural GDP vola- Source: Adapted from World Bank 2015a. tility and preliminary identification of key agricul- tural risks (by commodity, region, and stakeholder aggregated sector level at the end of the process. Individ- groups—including women and men) ual supply chain risk assessments that are later aggregated 5. Preliminary identification of levels of vulnerabil- at sector levels serve as the backbone of the approach. This ity among various stakeholder groups, differenti- process allows obtaining the information that is needed to ated by region and gender when possible. identify and design solutions tailored to managing agricul- This exercise allows the team to get familiar with the tural risk. country context, the broader economy, and with the agri- This sequential flow process has proven very practical in culture sector in particular. conducting ASRAs to date. It should be adapted to the particular circumstances of the country where it is applied 4.3.1 The output of the desk and should incorporate a gender-differentiated approach assessment to risk management in every step along the process flow. A gender-based outcome of the desk assessment will high- The step-by-step activities shown in Figure 4.1 are sequen- light the main gender issues of the ASRA and deliver sex- tial: the findings of each step inform and serve as the basis disaggregated data when possible. The findings of this for the next. Though the full description of activities in desk work serve as the basis for formulating preliminary this process can be found in World Bank (2015a), the fol- hypotheses regarding risk prioritization and the potential lowing subsections offer suggestions on how to incorpo- solutions. The subsequent field interviews serve to con- rate a gender lens along the four key steps of the ASRA.4 firm hypotheses and fill information gaps, particularly on the capacity of women and men participating in the agri- cultural supply chains to manage risk. 4 See World Bank (2015a) for a full description of the ASRA guidelines. 22 Agriculture Global Practice Discussion Paper By the time the desk study is complete, the team should be tobacco, cocoa, cotton) and domestically consumed crops able to produce a gender-differentiated output of: (i.e., maize, sorghum, millet, rice). The group of agricul- 1.  A timeline of major shock events that have caused tural crops to be assessed should be representative of the volatility in the country’s agricultural GDP or food sector as a whole. Women usually actively participate in imbalances one form or another along those supply chains. However, 2. A list of risk events and their frequency of additional crops representative of women’s agricultural occurrence activities and employment should be added to this group 3. Quantification of production losses for each com- if necessary. modity due to those risks in terms of yield losses (volume) and monetary value losses 4.3.3 Analysis of existing material 4. An assessment of the importance of price vola- Sources of information are numerous and vary from tility of agricultural commodities in terms of the one country to the next. They include government agen- magnitude of shock to the sector cies like Central Banks, Ministries (Finance, Agricul- 5. An assessment of major current public and pri- ture, Environment, Livestock, Rural Development), and vate sector interventions addressing risk. other government agencies like those involved in disaster prevention and early warning systems. Farmers’ asso- The importance of the background assessment is that it ciations and commodity boards publish annual reports sets the stage for the rest of the ASRA. It is at this early assessing performance, constraints, and risks. Similarly, stage that most of the working hypotheses of risk, impact, regional and international development institutions and and capacity to manage them start to be developed. The donor-supported research can prove valuable for this rest of the steps along this process use the assumptions initial assessment. Issues that are thoroughly researched and findings revealed in this initial background research. include: (i) demand conditions; (ii) supply chain structures; It is therefore imperative to properly integrate a gender (iii)  stakeholder roles; (iv) performance; (v) governance dimension at the onset. If planning ahead during the and coordination mechanisms; and (vi) public sector poli- background research and information-gathering phase cies and interventions.5 is overlooked, the team will not capture the informa- tion needed for the risk prioritization and corresponding More and more studies—either at the micro or macro gender-smart solutions developed later in the process. level—examine the gender dimension of agricultural production, processing, and distribution. One reason why 4.3.2 Crop selection and gender gender has not been a prominent part of the literature focus is lack of empirical information on individual ownership A developing economy’s agriculture sector is often based of land, housing, livestock, and agricultural productive on production of a wide range of agricultural products. activities. Moreover, most data are typically collected at Given the time limitations to assessing all crops, filtering the farm level, which gives a partial picture of individual- criteria to choose representative crops can be useful. The level productive agricultural activities, and fails to capture team can rank commodities in terms of: the information needed for analyzing issues concerning 1.  Their relative importance to export earnings gender differences. As a result, policy makers have only a 2. Their contribution to food security and limited understanding of issues regarding women’s par- 3. The participation of women and men. ticipation in agriculture. However, an important effort has been made over the last decade to disaggregate the col- A simple rule of thumb used in prior assessments to lection and processing of sex-disaggregated data that can arrive at a representative group of commodities is that the be used to support the risk prioritization process and the final mix of commodities represents around 80 percent design of risk management strategies. Box 4.1 presents of agricultural GDP. Typically, the group of commodi- ties will be composed of export commodities (i.e., coffee, 5 An extended explanation of these analyses can be found in Annex B. Gender and Agricultural Risk 23 Box 4.1. Questions and checklist for background research for a gender-differentiated ASRA When conducting the background research for an ASRA, using a gender-focused checklist can help ensure that the assessment team collects the information it needs to incorporate a gender dimension from the start. These questions are for illustrative purposes and may vary from country to country depending on the circumstances and required depth of the risk assessment in question, but should generally be guided by two overarching questions: »» What constraints limit women’s full involvement along all parts of the value chains in question? »» What are the differences between men and women in their capacity to manage agricultural risk? Information from a gender perspective to gather during a background research should include, but is not limited to, the following: »» National and cultural policies around asset ownership (i.e., women’s ability to legally own assets without men’s permission, joint ownership, ability to make asset-related decisions) »» National and cultural policies and practices around women’s access to land, mobile assets, and finance/loans »» National and cultural policies around inheritance »» Women’s flexibility and possibilities to seek employment, attend trainings and meetings, and organize childcare »» Women’s mobility to travel for jobs, trainings, market sales, milk delivery, etc. »» Women’s ability to travel alone »» Gender differences in access to assets (physical and financial) »» Gender differences in access to technology and information »» Gender differences in roles played in the supply chains »» Gender differences in education and literacy and numeracy skills of participants in supply chains Source: Authors. a checklist for information gathering in the background 4.3.4 Risk quantification research. The task of quantifying losses caused by different risk events is a key step in the ASRA process, but it is not Descriptive statistics help explain the status of men, straightforward. Financial losses attributed to agricultural women, girls, and boys in a society. Increasingly, interna- risks are caused by a variety of shocks related to pro- tional statistical compendiums offer data disaggregated by duction, market, or enabling environment factors. The sex and often by age.6 However, when analyzing specific impacts of some risks can be complicated to quantify in supply chains with clear geographical expressions, it is monetary terms, whereas others cannot be quantified at important to gather subnational data. Country statisti- all. Assigning proxy values can provide an estimate of the cal offices provide some information disaggregated by sex magnitude of financial losses to allow comparison of risks, (e.g., farm ownership) in annual or quarterly reports, but providing the information needed for their prioritization. it is necessary to dig deeper with ministry officials or to reach out to cooperatives and associations to obtain more Quantifying losses associated with a risk event enables nuanced information. comparison or risk ranking as an intermediate step in the risk prioritization process. Loss quantification provides an 6 They include: the World Bank’s Gender Data, World Development Indica- order of magnitude of the indicative losses in terms of tors, and Enterprise Surveys, and Women, Business, and Law; FAO’s Agri- agricultural GDP, which helps to justify investment in risk gender Statistics Toolkit; World Economic Forum’s Annual Global Gender Gap Report; UN statistics on the situations of women and men; United management solutions that reduce the impact of exter- Nations Development Programme International Human Development World nal shocks and strengthen resilience. In fact, losses are a Values Survey; surveys by Gallup and others as well as regional sources, such key piece of information for any cost-benefit analysis for as the Southern African Development Community (SADC) Gender Protocol investing in agricultural resilience. Having loss estimations Barometer. 24 Agriculture Global Practice Discussion Paper per supply chain before the field interviews makes the of risk impacting women because they typically par- fieldwork much more efficient and helps facilitate discus- ticipate in low-value crops, and the estimation based on sions with stakeholders over the causes of risk and their yield losses does not necessarily represent the importance capacity to manage. of their exposure and vulnerability to risk. Information about the number of women participating in each supply Several methods can be used to quantify risk losses, rang- chain and their importance in terms of income and food ing from calculating the simple variation from a mean security will also need to be used for risk prioritization, not value to using sophisticated statistical software applica- just the estimated losses. tions as is commonly done in the insurance industry. How- ever, the focus herein is on conducting a simple time series Stakeholder interviews will also provide important infor- analysis of crop yield volatility to arrive at an indicative mation for assessing the capacity of women and men value of direct losses caused by production risks. Market in each supply chain to manage risk, a key variable for risks (price volatility) and enabling environment risk need assessing and ranking the importance of risk. Having this to be assessed in different ways. A full description of risk information for every commodity under study is essential quantification can be found in World Bank (2015a), but for the introduction of a gender lens in this approach to the subsections below present some suggestions vis-à-vis risk management. gender differentiation. 4.3.4.2 Market and Enabling Environment Risk 4.3.4.1 Production Risk Quantifying market (price) risk could become a complex As strengthening resilience is a key policy objective in most exercise with no straightforward simple ways to assess the developing countries, the estimation of production losses causes of volatility. For the risk prioritization process, the is an instrumental step to rank, prioritize, and identify the team needs to distinguish between internationally traded solutions to address risk in agriculture and achieve more export commodities (i.e., cocoa, cotton, coffee, soybean) resilience to shocks (see Annex C for a detailed meth- and those that are mostly traded domestically (i.e., food odological approach to estimate indicative loss values in crops), as explained in World Bank (2015a). ASRAs also supply chains). In this regard, loss estimations disaggre- rely heavily on qualitative measures to evaluate price vola- gated by gender provide useful information to prioritize tility, exposure, and stakeholders’ capacity to manage risk. the attention of those risks that impact women in dif- ferent degrees of severity. When historic estimated loss Depending on the magnitude, price shocks affect income, calculations are made individually for each supply chain consumption, and ultimately the livelihood of farming under study and the results plotted in a matrix in accor- households. Women’s capacity to cope with exogenous dance with severity and frequency of impacts, the results price shocks needs to be assessed at various levels of the can be presented separately for men and for women. If supply chains, and differentiated from men’s capacity sex-disaggregated production data are not available, the to manage. Individual interviews and focus group dis- results of the calculations can be estimated for women by cussions can prove useful for gathering the information applying the proportion of women farmers in that supply needed to inform the overall ASRA and the risk prioriti- chain. zation in particular, as will become more apparent in the next section. However, to introduce a gender lens to this analysis, dis- aggregating the loss quantifications in terms of gender is While many production, market, and enabling environ- informative, though not determinant. The reason is that ment shocks have longer-term consequences and losses, the quantification of women’s losses alone is not neces- for simplicity’s sake, it is helpful to restrict the assessment sarily the best variable to capture the relative importance only to the immediate direct impact. Identifying the net Gender and Agricultural Risk 25 multiplier effects in the economy requires general equilib- more fully capture and analyze the impact of risk and rium models that go beyond the scope of the rapid assess- assess vulnerability. ment made in an ASRA. Capacity to manage risk is used to inform the risk prioriti- zation and the design and targeting of risk management 4.3.5 Understanding capacity interventions. Stakeholders with low capacity to manage to manage risk agricultural risk are vulnerable. This capacity is assessed Whereas understanding stakeholders’ capacity to man- both during the background work prior to the field visit age risk is of overall importance to the ASRA, it is a cen- and mostly during the field visit through stakeholder tral point for a gender-based risk assessment. Depending interviews. Box 4.2 offers suggested questions to answer on the size (by area cultivated or by value) of the supply to identify risk profiles of women and their capacity to chains, the magnitude of the impacts, and the frequency manage agricultural risk. of occurrence, the assessment of aggregate losses can sometimes mask the importance and scope of impacts on A risk profile of stakeholders in assessed supply chains can women if data are not sex-disaggregated. map stakeholders and the way they manage different agri- cultural risks. As an illustration, the risk profile summary Even if sex-disaggregated data were to be used, the over- shown in Table 4.1 lists the impacts of risk and risk man- all magnitude of losses in women’s participation in agri- agement capacities for herders with different herd sizes for culture might not appear prominent in terms of overall a hypothetical livestock supply chain. The introduction of a agricultural GDP volatility given women’s participation in gender lens requires identifying women herders and assess- low-valued crops. Likewise, their risk will not rank high in ing their capacity to manage risk, or identifying the role a risk prioritization exercise based solely on these grounds. women play within a herding household and assessing how women are impacted by risk and how they cope. In this Consequently, a separate assessment of women’s capacity example, though women’s herding units comprise only a to manage risk can reveal gender-differentiated vulner- small proportion of total livestock units, their low capacity abilities to risk and the implications for income, con- to manage places those groups in a situation of high vulner- sumption, and ultimately livelihoods. The assessment of ability. This information informs the prioritization exercise women’s capacity to manage risk (as is done for other and final design of risk management policies and strategies. groups of stakeholders) is therefore an important step to Box 4.2. Gendered line of enquiry to establish capacity to manage risk Understanding risk profiles entails (i) analyzing the roles of different stakeholders for each supply chain under assessment in a gender-disaggregated enquiry, and (ii) understanding their risk management capacities. To guide the assessment of stakeholders’ risk profiles, the team should aim to answer the following broad questions: »» Who is involved in the value chain analyzed (different stakeholders, segments of population, gender roles, etc.)? »» What risks have the greatest impact on women? »» What is the differentiated exposure and impact of risk for women and men? Are there regional differences? »» What are women’s current risk management practices in terms of risk mitigation, risk transfer, and/or risk coping strategies? »» How do women and men manage risks, and are their instruments effective? Why or why not? »» What are the limitations of women’s current risk management practices? Why are some risks not being managed? »» What is the capacity of supporting institutions to manage key risk predominantly faced by women? The assessment team needs to address those questions during the field interviews with stakeholders along each supply chain. For smallholder women farmers, who are often the most vulnerable and least vocal, focus group discussions are an important technique to discuss their risks and vulnerabilities, mitigation strategies, and coping mechanisms. Source: Authors. 26 Agriculture Global Practice Discussion Paper Table 4.1. Illustration of stakeholders’ vulnerability mapping in livestock Risk Management Category Share of Women of Herders/ Herding Men (15% of Animals Characteristics Household (85% of Pastoralists Units) Pastoralists Units) Larger-scale These units provide full-time 6% Can move livestock long No women participate herders (>20) employment for family members distances if necessary; winter in this group and adequate incomes. Good preparation with hay and fodder access to pastures and inputs. production Medium-scale These units have adequate 25% Some capacity to prepare hay Limited capacity to herders (15–20) incomes to support herding and fodder but vulnerable to prepare fodder but households but units are of dzuds7 in extreme event years vulnerable to dzuds middle or lower wealth levels and even in mild years have limited access to capital. Small full-time These herding units are poor, No data Highly vulnerable to dzuds of Highly vulnerable to herders (<15) without alternative sources of moderate to extreme magnitude dzuds of any magnitude income, and are often dependent on state support. Would like to exit the sector if possible. Periodic herders This group enters and exits No data Highly vulnerable to dzuds of Highly vulnerable to (often <15) livestock production depending any magnitude dzuds of any magnitude on economic circumstances. For this group, herding is a safety net. Source: Authors. 4.3.6 Proactively assessing specific undertaken as part of the ASRA, and the findings incor- vulnerable groups porated in the final sector-level risk prioritization.7 In some cases the risk assessment team will need to under- take particular (ad hoc) assessments in reference to policy Similarly, identifying those supply chains with high partic- concerns about vulnerable population groups. Certain ipation of women merits special (ad hoc) risk assessments risks may have relatively low impacts on the sector or a to influence the design of the relevant risk management geographical region, but relatively high impacts on a par- policies and strategies. In countries where the initial back- ticularly vulnerable group of stakeholders. This means ground research shows that women farmers in certain that applying prioritization filters of high impacts and low communities or regions are in vulnerable positions, efforts capacity to manage (as described above) might not capture need to be deepened with auxiliary secondary research or those policy concerns. For example, losses in Region A may information that can identify them more precisely. This outweigh those of Region B in terms of monetary value, will enable targeted interventions designed to assist those but if a great majority of the country’s rural food insecure groups of households even when the impacts of risk appear population lives in Region B and is dependent on low- marginal at the aggregate sector level. Even small shocks value crops prone to risk-related production losses, policy to the livelihood of vulnerable farming communities can makers may want to prioritize risk management interven- tions in Region B to address food security concerns, rather 7 A dzud is a Mongolian term for a severe winter in which a large number of than the magnitude of value lost. Special assessments of livestock die, primarily due to starvation due to being unable to graze, in other vulnerability driven by policy concerns will need to be cases directly from the cold. Gender and Agricultural Risk 27 have a catastrophic impact for large groups of households, The main activities typically covered during the fieldwork but these impacts are usually masked at the cumulative include: sector level. 1.  Data mining. This is a valuable opportunity to fill gaps regarding sex-disaggregated data. Some In this way, the results of the risk prioritization process pieces of information exist only in hard copy and will incorporate not only strategies to manage the major only by visiting local institutions can the team risk of magnitude of losses, but also the policy angle of access them. For example, commodity boards’ gender considerations. annual reports have valuable information about supply chain dynamics reported on an annual 4.4 STEP 2: Stakeholder basis, including causes for drops in volume, and interviews logistics and market issues. Agricultural colleges and NGOs undertake research on agricultural During the fieldwork for stakeholder interviews, the team supply chains that can be differentiated by gender. establishes direct, one-to-one contact with various women Local offices of international development insti- and women’s groups participating in the assessed supply tutions (some of which have gender as a priority chains. This is done to obtain the narrative and fill the policy theme) sponsor and/or undertake research gaps left during the desk assessment. Moreover, it is useful on agricultural productivity, supply chains, climate for identifying the storyline. The team seeks to: change, and food security. Similarly, disaster risk 1.  Corroborate the timeline of events for each supply management agencies collect data on vulnerable chain as it affects women and men populations, design coping strategies, and deliver 2. Confirm or identify the causes of risk events relief programs. These latter entities usually disag- 3. Test whether the estimated losses can be validated gregate data by sex. against actual losses 2. Individual interviews. The team typically interviews 4. Assess stakeholders’ capacity to manage risks (both representatives of each level of value chain par- women and men). ticipant for each supply chain under assessment. This means a high volume of interviews, since Box 4.3 shows a checklist of issues for team members to the ASRA will cover farmers, processors, input address during the fieldwork. Box 4.3. Gender-based line of enquiry for ASRA fieldwork The following activities are part of the line of enquiry for team members participating in the ASRA. These can be used as a check- list for interviews and focus group discussions: »» Identify the causes of losses and women’s attribution of them (single or multiple causes) »» Assess how losses affected women participating in the supply chains »» Corroborate the frequency of those events »» Determine if losses were evenly distributed by area, by farmer groups, and/or by gender »» Establish how different stakeholders (women and men) managed risks »» Find out how shocks were absorbed by women and men »» Determine if any women’s enterprises went out of business »» Get a sense of long-term threats to livelihoods »» Assess capacity to manage risks by existing institutions (any gender bias) »» Test if magnitudes of losses estimated during the desk assessment are correct »» Analyze government’s current strategies to respond to shocks »» Elicit women’s perceptions of risk priorities »» Identify women’s suggested solutions Source: Authors. 28 Agriculture Global Practice Discussion Paper suppliers, government agencies, financial inter- planning is imperative. Annex D presents guide- mediaries, service providers, traders, and export- lines for conducting focus group discussions with ers. Some value chains are clustered around farmers. particular regions and it is possible to efficiently 4. Validation workshops. During the fieldwork, the team meet most stakeholders by travelling to the region. holds workshops with key stakeholder representa- Most government agency representatives can be tives of each supply chain under study to validate met in the national or provincial capital cities. the findings of the ASRA, as well as to advance The story on risk for each supply chain can be a list of potential solutions. This step’s value is in identified by interviewing both women and men smoothing out any subjective views held by the entrepreneurs and assessing their exposure and assessment team by sharing the methodology and individual capacities to mitigate, transfer, or cope findings with stakeholders. The conclusions of with agricultural risk. these workshops for each supply chain will later 3. Focus groups. Group discussions are useful for serve as the basis for the sector risk prioritization addressing particularly large, homogenous groups (see next step on risk prioritization). (i.e., women farming rice, women in drought-prone areas, women in export crops, etc.). Focus groups By the end of the field mission each team member has facilitate identification of groups’ and communi- a partial view of agricultural risk because the team has ties’ exposure to risk and capacities to mitigate and focused on analyzing specific commodities in selected ter- cope with risk collectively and individually. Local ritories. The preliminary findings of these exercises for gender specialists could be very helpful in advis- each supply chain can be summarized in a matrix by plot- ing the team for each supply chain where it will be ting risks in terms of: (i) the level of intensity (capacity to more critical to interview women, men, and differ- produce losses), (ii) frequency of occurrence, and (iii) the ent kind/sizes of agricultural enterprises, as well distribution of impacts across stakeholders. Table  4.2. as how to choose the more representative groups, illustrates a summary risk identification matrix for a hypo- and the way to convene for focus group discussions thetical supply chain. (as women may have restricted mobility). Early Table 4.2. Illustration of risk identification in a cotton supply chain Frequency Intensity Identified Risks of Events of Losses Who Suffers Most Production risk •  Pest and diseases Medium High Women farmers •  Weather risk High High Farmers (women and men), ginners •  Chemical poisoning High High* Women pickers Market risk •  International price volatility High High Ginneries and exporters •  Exchange rate Low Low Ginneries •  Ginners credit default Low Low Bank •  Farmers credit default Low Medium Bank •  Domestic price volatility High High Women farmers and pickers Enabling environment risk •  Port delays Low Low Exporters •  Sudden changes in orders of chemicals Low High Women farmers** •  Carrying large quantities of cash Low Low Ginners  High in individual terms as idiosyncratic risk *  **As with less access to alternatives to manage Source: Authors. Gender and Agricultural Risk 29 4.5 STEP 3: Risk 4.5.1 Prioritization matrix prioritization Team members can plot the results of the prioritization exercise for each supply chain in a gender-differentiated The prioritization process is a team and stakeholders’ fashion, as illustrated in Table 4.3 for men and Table 4.4. exercise. As a result of the desk assessment and fieldwork, for women. The plotting of risk in these tables should the team has already identified the risks for each com- consider stakeholders’ differential capacity to manage the modity in terms of: (i) frequency of occurrence; (ii) sever- identified risks. For example, if the team finds that stake- ity of impact (intensity); and (iii) stakeholders’ capacity to holders already have high capacity to manage a particu- manage the identified risks; and has reviewed (iv) special lar identified risk, this particular risk will not be ranked ad hoc risk assessments of vulnerable farming groups among the most important to cause high impact. (women and men) that might have been conducted. Fig- ure 4.2 illustrates the components of risk prioritization. While women face the same risks in this example, their risks are concentrated more in the upper right corner of Figure 4.2. Components of Risk the matrix. These women’s enterprises are more vulner- Prioritization able to risk than are men’s, mostly because women have less capacity to manage risk in this example. This can be because their farming is less technologically advanced, Potential to produce Frequency of events losses (intensity) they are not recipient of extension services, and lack the knowledge and/or financing for improved animal health care. The use of gender-differentiated prioritization matrices helps enormously in prioritizing risks for each Risk supply chain under assessment. Prioritization Matrix If team members use these tables in a standardized man- Special vulnerable ner, results can be shared and discussed among the team Stakeholders’ capacity farming groups members and among stakeholders in an easy-to-under- to manage risk risk assessment (women and men) stand fashion, and adjustments can be quickly made (women and men) accordingly. Source: Authors. Table 4.3. Illustration of risk prioritization for men in livestock Impact/Probability of Event Low Moderate High Highly probable Milk contamination Foot and mouth (1 in 3) Drug and inputs disease contamination and adulteration risk Probable Drought Power cuts (1 in 5) Occasional Glut (price risk) (1 in 10) Remote Maize feed shortages Aflatoxins (1 in 20) Source: Authors. 30 Agriculture Global Practice Discussion Paper Table 4.4. Illustration of risk prioritization for women in livestock Impact/Probability of Event Low Moderate High Highly probable Foot and mouth Milk contamination (1 in 3) disease Drug and inputs contamination and adulteration risk Probable Power cuts Maize feed shortages Drought (1 in 5) Aflatoxins Occasional Glut (price risk) (1 in 10) Remote (1 in 20) Source: Authors. 4.5.2 From commodity risk a sector perspective. This involves choosing only the risks to sector risk located in the right upper corner of each commodity’s Translating individual commodity risk prioritization into risk prioritization matrix and relocating them in a single an aggregate sector risk prioritization is also a collective sector-aggregated risk prioritization matrix. This process exercise. Prior to this stage, team members have assessed involves not just familiarization with the risk assessment risks for individual commodities but have not yet estab- done for each commodity, but a change in perspective lished a broader sector perspective. After completing an from the commodity supply chain to a broader agriculture individual prioritization matrix for each commodity, the sector perspective. Figure 4.3 illustrates the process of risk team proceeds to reclassify those risks in terms of fre- aggregation and reprioritization from the commodity quency, severity of impact, and capacity to manage from level to the sector level. Figure 4.3. Reprioritization of risks from commodity to sector level COMMODITIES SECTOR Prioritization PRIORITIZED VULNERABLE GROUPS RISKS ONLY! Special gender/vulnerable groups risk assessments Source: Authors. Gender and Agricultural Risk 31 As mentioned earlier, the findings of any special risk risk prioritization matrix. As with individual commodity assessment on women and/or vulnerable farming groups risk, separate sector prioritization matrices for men and motivated by policy considerations need to be accounted women can facilitate the design of gender-smart solutions in the final sector prioritization matrix to inform policies to agricultural risk. and strategies to strengthen resilience of vulnerable farm- ing systems. An illustration of such prioritization matrices is shown in Table 4.5 and Table 4.6 for men and women, respectively. The final outcome of the aggregated risk prioritization Though these are simplified illustrative matrices, they process is a list of key priority risks that can explain the need to be adapted to the context of every country, and causes of agricultural GDP volatility, food insecurity, can be designed as complex as needed. In this hypotheti- and livelihoods disruptions; it is these risks that must be cal example, women are concentrated in primary activities addressed to reduce the overall impact of risk in the agri- in food crops and as laborers in export crops with virtu- culture sector. The highest priority risks are those located ally no participation at other levels of the supply chains. in the boxes in the upper right corner of the aggregated Also, a high concentration of unmanageable production Table 4.5. Illustration of sector-level risk prioritization for men Impact/Probability of Event Low Moderate High Highly probable Flea beetle infestation (S) Ginners counterparty risk/ Cotton price volatility (Cot) Maize streak diseases (M) side selling (Cot) Coffee price volatility (Co) Fungal diseases (e.g., Bull work outbreak (Cot) Yellow moto virus (R) powdery mildew) (C) Probable Aflatoxin (M) Erratic rainfall (R) Drought (R) Regulatory risk (Co) Occasional Counter party risk Pests (Rodents, army (Ginners)/International worms, quealea birds) (R) buyers (Cot) Note: Cot = Cotton, Tob = Tobacco, Co = Coffee, C = Cashew nuts, R = Rice, M = Maize, S = Sesame. Source: Authors. Table 4.6. Illustration of sector-level risk prioritization for women Impact/Probability of Event Low Moderate High Highly probable Brown stick disease (Ca) Flea beetle infestation (S) Drought (M) Aflatoxin (M) Mosaic disease (Ca) Probable Wild animals (Ca) Insects and pests (e.g., Excess rainfall (Ca) beetle, army worm) (B) Occasional Diseases (B) Note: Cot = Cotton, Ca = Cassava, Tob = Tobacco, Co = Coffee, M = Maize, S = Sesame, B = Beans. Source: Authors. 32 Agriculture Global Practice Discussion Paper risk appears in the upper right corner, revealing very low 4.6.1 Understanding gender-based capacity to manage, exposing high vulnerability to food constraints insecurity. This information can be extremely useful for Earlier chapters argued and showed evidence indicating designing gender-smart solutions to agricultural risk, that the fundamental reasons for women’s lower capac- informing government policy and interventions. ity to manage agricultural risk are the same as those for women’s lower levels of agricultural productivity (even 4.6 STEP 4: Gender-smart operating under similar conditions as men in some cases). solutions Therefore, in addition to the solutions to agricultural risk informed by the ASRA (World Bank 2015a), the risk Whereas the ASRA is a simple-to-understand, logical pro- management strategy needs to remove gender-based con- cess to prioritize risk, it is more complex to standardize straints (GBCs) to risk management. In this way, the risk the process to identify and prioritize the solutions into a solutions can be integrated with overall policies and strat- single approach. This is mostly because every country has egies to reduce the gender gap and increase resilience for a unique way of decision making that depends on public smallholders in agricultural systems. policies, institutional strengths, fiscal constraints, and the political economy. However, risk management does not Chapter 3 demonstrated that the ASRA methodology start from zero. Stakeholders in all countries already have basically utilizes supply chains as the focus of risk assess- interventions and instruments with various temporal and ment and prioritization. It is therefore practical and spatial features to incorporate risk management activi- logical to also use supply chains as the entry point for ties. The design of a strategy and an action plan there- identifying GBCs to risk management and incorporating fore requires focusing on identifying the gaps of current the findings into the overall sector risk management poli- interventions and designing a package of solutions that cies and strategies. addresses the main underlying causes of risk. The next subsections present some guidelines to complement the Table 4.7 summarizes the most critical GBCs to risk man- ASRA approach outlined in World Bank (2015a). agement in agricultural supply chains as identified in ear- lier chapters. Gender and Agricultural Risk 33 Table 4.7. Summary of gender-based constraints in agricultural supply chains Productivity and Vertical Empowerment and Horizontal Linkages Linkages Business Enabling Environment Access to assets, including: Active participation, including: Laws, policies, and institutions: Land Membership Formal discrimination in law and policy Labor Decision making Cross-sectoral or cross-jurisdictional Capital Leadership in decision making inconsistencies Inputs Unequal enforcement Technology Information Education Source: Authors, partly adapted from USAID 2013. The assessment of GBCs aims to link gender issues to the ture) may require efforts to address laws regarding prop- ASRA by identifying GBCs and incorporate appropriate erty rights; in areas where the definition of a household measures in the government policies and investment plans. excludes women from participation in farmers’ groups, women’s inclusion in risk management initiatives will be restricted. 4.6.2 Identifying gender-smart solutions Lessons from development partners suggest that partici- Gender and agricultural risk are contextual, so gender- patory, inclusive approaches aimed at building adaptive based solutions to agricultural risk also need to fit the capacity, such as farmer-to-farmer extension or farmer- context of the country and situations to which they are led innovation, are scalable. But individual innovations— applied. This subsection provides some key references including some that are particularly attractive to for illustrative purposes to guide the identification of key women—are difficult to scale out because they are suited gender-smart solutions when designing risk management to highly specific environments and contexts. Moreover, it strategies based on the findings of the previous steps. is valuable to recognize that women make an active and important contribution to climate adaptation based on Sex-disaggregated data in countries as diverse as Kenya, their local knowledge and capacity, and that it is limiting Senegal, Uganda, and Bangladesh show that both men and simplistic to view them as passive victims of climate and women are taking up new agricultural practices change (Otzelberger 2011). Successful adaptation pro- that are likely to enhance their resilience to the effects of jects increase women’s opportunities to add value to their drought (World Bank 2015b). These practices must be agricultural activities—for example, through agricultural reinforced if agriculture is to withstand the effects of cli- processing and marketing—and diversify their income- mate change while bringing about improved productivity earning opportunities (Njuki et al. 2011). In other words, and food and nutrition security, and increased economic they promote transformational change in agriculture, growth.8 Policy changes are particularly critical for adopt- ing an effective and sustainable gender-based approach to acknowledge women’s role in that process, and strengthen risk management. For example, securing women’s right to resilience of farming systems. own land (and to thus protect their investments in agricul- Opportunities for women are not equal where legal gen- der differences are prevalent. Since 2009, Women, Busi- 8 Such as modifications in planting dates or changes in crop varieties, but also ness and the Law at the World Bank has collected data practices that lead to more transformative change, such as diversified livelihoods and an increase in assets. about legal restrictions on women’s entrepreneurship and 34 Agriculture Global Practice Discussion Paper employment to inform policy discussions and promote affect women’s incentives or capacity to work or to set up research on the linkages between the law and women’s and run a business. economic opportunities.9 The data focus on seven indica- tors: (i) accessing institutions, (ii) using property, (iii) get- A growing global consensus recognizes the intrinsic and ting a job, (iv) providing incentives to work, (v) going to instrumental importance of securing rights to land and court, (vi) building credit, and (vii) protecting women other productive resources in the eradication of poverty from violence. New areas covered within these indicators and reduction of gender inequality (World Bank 2014). include legislation on issues such as nondiscrimination in Women’s ability to exercise agency over land and housing access to credit, care leave for sick relatives, the legal age is determined by the interplay of laws—including statu- of marriage, and protection orders for victims of domes- tory, customary, and religious laws—and social norms. tic violence. Control over land and housing has instrumental value. Women who have more control over land—whether Equality of opportunity allows women to make the through inheritance, land titling, improved documenta- choices that are best for them, their families, and their tion, or stronger communal rights—tend to have greater communities. However, opportunities for women are economic opportunities, mobility outside of the home, not equal where legal gender differences are prevalent. and decision-making power. (See Box 4.4 for illustrations Such restrictions constrain women’s ability to make eco- of the importance of land ownership as a contributor to nomic decisions in a variety of ways, and can have far- gender disparities.) The World Bank’s (2010) “Toolkit for reaching consequences. Moreover, they are associated Integrating Gender-Related Issues in Land Policy and with real economic outcomes. Many laws remain restrict- Administration Projects” provides guidelines for a gender ing women from taking certain actions. Women, Business analysis of the socioeconomic and cultural conditions in and the Law examines 11 areas where women may face a project area, including with regard to statutory and cus- constraints on their legal capacity to act or ability to con- tomary property rights, land policies and legislation, land duct transactions. Each action or transaction is examined administration institutions, and land market transactions. separately for married and unmarried women. Approaches based on information and communication Many laws continue to prevent women from improving technology (ICT), including radio, TV, cellphones, and their own well-being and that of their families by working social media, promise to enhance women’s access to risk or running a business. Datasets such as the World Bank mitigation practices and weather and climate informa- Group’s Enterprise Surveys and Doing Business have led tion, reduce the perceived risks, and strengthen women’s the way in providing information on the challenges con- participation in commodity value chains. Like many other fronting firms and entrepreneurs in starting and expand- studies, a 2011 World Bank study found a high demand ing their businesses and creating jobs. But women often for extension information among women farmers; that face additional constraints in starting businesses and level of demand presents an opportunity to train agricul- navigating the workforce. How can governments improve tural extension officers to use ICTs to reach an increased women’s access to entrepreneurial and employment activ- number of women farmers more cost-effectively. The ities? Answering that question requires understanding study provided a useful step-by-step guide to introducing many factors—from access to education and health care, ICT-based solutions with a gender focus in agricultural to social and cultural norms, and many things beyond. projects (World Bank 2011). One important factor is how laws, regulations, and institu- tions differentiate between women and men in ways that Promoting associativity and identifying key interventions along the supply chain could be another approach to identify strategies aiming at assisting women in removing gender- based constraints to ARM tools. Figure 4.4 illustrates one 9 Women, Business and the Law 2016: Getting to Equal is the fourth in the series. such approach for identifying entry points for removing http://wbl.worldbank.org/~/media/WBG/WBL/Documents/Reports/ GBCs in agricultural supply chains at various levels. 2016/Women-Business-and-the-Law-2016.pdf Gender and Agricultural Risk 35 Box 4.4. Gaps in land ownership and control Twenty years ago, Argawal’s study of rural South Asia identified gaps in land ownership and control as the most important con- tributor to disparities in economic well-being, social status, and empowerment (Agarwal 1994). More recent studies reinforce and extend this finding with the following associations: »» In Vietnam, women with a joint title are more aware of legal issues, are more likely to proactively seek a Land Tenure Cer- tificate, have more say in the use and disposition of land, and are more likely to earn independent incomes than women who are not on the title (World Bank 2008). »» In Peru, squatter households given property titles experienced a 22 percent reduction in fertility rates, and women who received a joint title were two times less likely to have a child than women in families in which the title was in the male part- ner’s name only. Receipt of titles also allowed women to seek paid work instead of spending time safeguarding their land against property invasion (Field 2007). »» In Nepal, women who own land are significantly more likely to have a final say in household decisions, and children of mothers who own land are less likely to be underweight, with associated benefits for almost all maternal and child nutritional outcomes (Allendorf 2007). »» In Ecuador, joint land ownership increased women’s participation in household decisions about crop cultivation (Deere and Twymen 2012a, 2012b). »» In rural Karnataka, India, ownership of land and housing improved women’s mobility outside the home and their ability to make decisions about their work, health, and household spending (Swaminathan, Lahoti, and Suchitra 2012). »» Expanding women’s ownership of land and housing is not a panacea, nor is land legislation alone. Access to credit, markets, education, extension services, technology, personal mobility, and public voice all influence women’s ability to claim and make use of property rights (Spichiger et al. 2013). Source: World Bank 2014. Figure 4.4. Illustrative value chain and possible entry points for removing gender-based constraints Provide opportunities for Assist women’s groups Assist women to overcome women to gain access to to purchase equipment mobility constraints and inputs and market information to expand processing social barriers LOCAL AND PRODUCERS AND TRADERS AND PROCESSORS INTERNATIONAL OPERATORS EXPORTERS MARKETS Different types of associations Improve participation of Build horizontal linkages that Encourage more women-owned women in association favor women and women’s enterprises to join trade associations leadership roles enterprise participation Source: USAID 2013 36 Agriculture Global Practice Discussion Paper 4.6.3 Applying criteria to prioritize 4.6.4 Integrating ARM and social gender-based constraints protection to ARM While ARM focuses on strengthening stakeholders’ resil- The term GBC includes both the factors (i.e., discrimina- ience along agricultural supply chains, when the intensity tory land laws) and the measurable disparities that result of exogenous shocks reaches catastrophic levels, agricul- (i.e., women hold only a small fraction of land titles). The tural measures can only achieve so much. Recognizing team can use gender analysis guidelines to identify GBCs, those limitations, the team will need to integrate ARM drawing on information collected in the background strategies with broader disaster risk management and research. These include: (i) identifying conditions of gen- social protection policies and activities. Box 4.5 illustrates der disparity; (ii) identifying the factors that cause the gen- the justification for doing so. der disparities; and (iii) formulating a “cause and effect” hypothesis. The findings of the ASRA process will inform the final package of ARM strategies that are suggested in any To prioritize activities aimed at removing GBCs to risk particular country. The factors that will influence the management, a set of criteria can be applied to the list choices of risk management strategies are usually based of identified GBCs to risk management. Criteria include: on (among others): (i) practicality of implementation; 1.  Those that restrict a more efficient response to (ii)  affordability; (iii) potential impact (short, medium, shocks or place disproportionate costs or weights or long term); (iv) complementarity with public policies; on women in the face of risk and (v) the political economy. A gender-differentiated risk 2. Those that exclude women’s access to (and dis- management strategy will need to incorporate specific posal of) assets programs and projects that address the vulnerabilities 3. Those that discriminate against women’s partici- of stakeholders arising from gender differences. The risk pation in technology, information, and higher- management assessment process discussed herein should value markets capture the relevant information and prioritize the solu- 4. Those that interfere with the achievement of more tions that will remove GBCs and allow all stakeholders livelihood resilience to shocks. access to risk strategies and tools. Box 4.5. ARM and social protection In many countries, a policy priority is to keep households from losing their asset base below a certain threshold, and to likewise ensure consumption is not destabilized after a disaster (Carter et al. 2006; Heltberg et al. 2009). Doing so is only possible if social protection interventions can be scaled up or introduced rapidly after a shock. Clarke and Hill (2013) investigated the case of Ethio- pia and Malawi and found that the cost of a drought to households increases from zero to about US$50 per household if support is delayed by four months after harvest and to about US$1,300 if support is delayed by six months or more. This rapid increase is due to distress sales and loss of assets (especially livestock). Acting rapidly implies: (i) scaling up social protection immediately after a disaster; (ii) targeting the affected population; (iii) enhanc- ing livelihoods to make them more resilient to shocks, and (iv) having stronger institutions for managing risks and crises (Kuriakose et al. 2013; World Bank 2013). In a world in which climate change makes such disasters more frequent or intense, the effect on poverty could increase significantly, making policies to support affected vulnerable households even more important. Source: World Bank 2015b. Gender and Agricultural Risk 37 CHAPTER FIVE Report Findings and Conclusions This paper examines the conceptual basis and available empirical evidence sustaining the thesis that agricultural risk has a differentiated gendered impact on rural house- holds involved in agricultural activities. The emerging research strongly suggests that this area deserves more attention to enable practitioners to effectively incorporate those effects into agricultural policy and risk management strategies. Review of the evidence shows that aggregate economic shocks do not have homogenous effects on farming women and men. Gender matters in explaining differential effects, in terms of both the direct, or first-round, effects of the economic shock, and of households’ cop- ing strategies, or second-round effects. Moreover, these contrasts vary across countries and stages of development. Women farmers typically face different constraints than men when facing risk, and the feasible options open to women also differ. In certain contexts, social norms or barriers increase the complexity of the challenge for women to mitigate risk or cope in a man- ner that preserves their agricultural and nonagricultural livelihood strategies. If these prevailing differences across gender lines are not taken into account when assessing agricultural risk, the potential for women to benefit from risk mitigation strat- egies could be overestimated, and the potential for men and households as a whole to benefit underestimated—providing a misleading indication of what Agricultural Risk Management (ARM) strategies can achieve. To truly capture and address agricultural risk, incorporating gender-based impacts and responses, assessment of those differ- ences must be an integral part of Agricultural Sector Risk Assessments (ASRAs) and the findings incorporated in corresponding ARM strategies. Understanding the different roles and situations of women in their participation along agricultural supply chains in terms of their access to and control of productive resources, services, and employment opportunities is critical for assessing and prioritiz- ing agricultural risk and for identifying gender-smart solutions for managing risk and strengthening resilience. And recognizing the root causes of such gender differences is essential if risk mitigation investments and risk coping programs are to reduce rather than reconstruct people’s risk in future risk events. Gender and Agricultural Risk 39 This paper shows why and how gender-differentiated Risk management does not start from zero, fortunately. considerations can be included in agriculture sector risk Stakeholders in all countries already have interventions assessments and strategies. A single blueprint for a risk and instruments with various temporal and spatial features management roadmap is not feasible because of the to incorporate risk management activities. The design of diversity of risks, cultural and social contexts, supply a strategy and an action plan therefore requires identi- chain structures, and overall country contexts. Nonethe- fying the gaps of current interventions and designing a less, despite the diversity of approaches, some systematic package of solutions that addresses the main underlying processes and checklists can be adopted by all risk man- causes of risk. A gender-differentiated risk management agement approaches. strategy must thus incorporate specific programs and projects that address the vulnerabilities of stakeholders The step-by-step approach described in this paper intends arising from gender differences. The risk management to fill the operational gap of integrating a gender dimen- assessment process discussed herein should capture the sion into ASRAs. The focus is on the approach because relevant information and prioritize the solutions that will both agricultural risk and gender issues are contextual— remove gender-based constraints and allow all stakehold- thus the outcomes of a gender-based risk assessment are ers access to risk strategies and tools. highly dependent on the social, cultural, geographic, and economic contexts of the setting in which the assessment Overall, the main policy and operational implications are is applied. Each ASRA must be tailored to capture gender that gender-differentiated impacts to shocks, the causes of differences that arise due to a country’s unique combina- differentiated responses, and the effects on well-being and tion of risks, cultural differences, institutional arrange- resilience need to be identified. A gendered assessment ments, and fiscal constraints. of agricultural risk will fill an information gap and help to inform policy and better target the operational tools The gender-based approach to agricultural risk is sim- and interventions designed to improve ARM practices in ply an orderly process to analyze, identify, and prioritize developing countries. These are precisely the objectives of risk, and serves as the basis for the design of policies and a gender-based ASRA. interventions to manage agricultural volatility and food security and to strengthen farming systems’ resilience to Introducing a gender-differentiated approach to ARM is shocks. The primary objective of a gendered approach a key issue for development effectiveness. In the face of to agricultural risk is to assist development practitioners multiple risks, the resilience of farming households is criti- to simplify and better comprehend the complexity of cal for collective action, coordination, and public inter- gender-based ARM by following a systematic approach vention. One cannot understand the current and future to prioritize solutions to mitigate, transfer, and/or cope potential of the agriculture sector in developing countries with agricultural risk. Highlighting the types and orders without understanding the differential ability of women of magnitude of risks and targeting the most vulnerable and men to anticipate and respond to agricultural shocks. stakeholders can improve planning and investments to strengthen resilience in agriculture. 40 Agriculture Global Practice Discussion Paper References Aduamoah-Addo, Lois. 2016. “Culture, Rural Women and Land Rights in Ghana.” Arti- cle 474618 at http://www.ghanaweb.com/GhanaHomePage/regional/Culture- rural-women-and-land-rights-in-Ghana-474618. Alderman, Harold, John Hoddinott, and Bill Kinsey. 2006. “Long Term Conse- quences of Early Childhood Malnutrition.” Oxford Economic Papers 58 (3): 450–474. Anderson, David, and Vigdis Brouch. 1999. The Poor Are Not Us: Poverty and Pastoralism. Eastern African Studies. Athens: Ohio University Press. Berloffa, Gabriella, and Francesca Modena. 2009. “Income Shocks, Coping Strat- egies, and Consumption Smoothing. An Application to Indonesian Data.” Uni- versity of Trento. http://www.siecon.org/online/wp-content/uploads/2012/08/ Berloffa-Modena.pdf. Bhalotra, Sonia. 2010. “Fatal Fluctuations? Cyclicality and Infant Mortality in India.” Journal of Development Economics 93 (1): 7–19. de Walque, Damien, William H. Dow, and Erick Gong. 2014. “Coping with Risk: The Effects of Shocks on Reproductive Health and Transactional Sex in Rural Tanza- nia.” Policy Research Working Paper 6751, World Bank, Washington, DC. Deere, Carmen Diana, and Cheryl R. Doss. 2006. “The Gender Asset Gap: What Do We Know and Why Does It Matter?” Feminist Economics 12 (1&2): 1–50. Dercon, Stefan, and Pramila Krishnan. 2000. “In Sickness and in Health: Risk-sharing within Households in Rural Ethiopia.” Journal of Political Economy 108 (4): 688–727. Doss, Cheryl R. 1999. “Intrahousehold Resource Allocation in Ghana: The Impact of the Distribution of Asset Ownership within the Household.” In Food Security, Diversi- fication and Resource Management: Refocusing the Role of Agriculture?, ed. G. H. Peters and von Braun. Aldershot, UK: Dartmouth Publishing. Doss, Cheryl R. 2006a. “Women’s Access to Land in Ghana.” Paper presented at the IAFFE meetings, Sydney, Australia, July 5–9. Doss, Cheryl R. 2006b. “The Effects of Intrahousehold Property Ownership on Expenditure Patterns in Ghana.” Journal of African Economies 15 (1): 149–180. Duflo, Esther, and Christopher R. Udry. 2004. “Intrahousehold Resource Allocation in Côte d’Ivoire: Social Norms, Separate Accounts and Consumption Choices.” NBER Working Paper No. W10498, National Bureau of Economic Research, Cambridge, MA. Gender and Agricultural Risk 41 Fafchamps, Marcel, and Agnes R. Quisumbing. 2005. Recent Evidence.” Discussion Paper No. 20, Interna- “Assets at Marriage in Rural Ethiopia.” Journal of Devel- tional Livestock Research Institute, Nairobi, Kenya. opment Economics 77 (1): 1–25. International Financial Corporation (IFC). 2016. Invest- FAO. 2013. Understanding and Integrating Gender Issues into ing in Women along Agribusiness Value Chains. International Livestock Projects and Programmes: A Checklist for Practitioners. Financial Corporation, Washington, DC. Rome, Italy: The Food and Agriculture Organization International Food Policy Research Institute (IFPRI). of the United Nations. 2015. “Gender, Assets, and Agricultural Development. Folbre, Nancy. 2001. The Invisible Heart: Economics and Fam- Lessons from Eight Projects.” Discussion Paper 01436, ily Values. New York: The New Press. International Food Policy Research Institute, Washing- Food and Agriculture Organization (FAO). 2010–2011. ton, DC. The State of Food and Agriculture. Women in Agriculture: Clos- International Fund for Agricultural Development (IFAD). ing the Gender Gap in Development. Rome, Italy: The Food 2010. Livestock Thematic Papers. Gender and livestock: Tools and Agriculture Organization of the United Nations. for Design. Rome, Italy: International Fund for Agricul- Garcia, Jaume, Pedro J. Hernandez, and Angel Lopez- tural Development. Nicolas. 2001. “How Wide Is the Gap? An Investi- Johnson, Nancy, Jemimah Njuki, Elizabeth Waithanji, gation of Gender Gap Differences Using Quantile Marihno Nhambeto, Martha Rogers, and Eliza- Regression.” Empirical Economics 26 (1): 149–167.  beth  H. Kruger. 2013. “The Gendered Impacts of Gilligan, Daniel O., Neha Kumar, Scott McNiven, J. V. Agricultural Asset Transfer Projects: Lessons from the Meenakshi, and Agnes R. Quisumbing. 2013. “Bar- Manica Smallholder Dairy Development Program.” gaining Power and Biofortification: The Role of Gen- CAPRi Working Paper 115, International Food Policy der in Adoption of Orange Sweet Potato in Uganda.” Research Institute, Washington, DC. IFPRI Discussion Paper 1353, International Food Pol- Katz, Elizabeth, and Juan Chamorro. 2003. “Gender, icy Research Institute, Washington, DC. Land Rights, and the Household Economy in Rural Haddad, L. and R. Kanbur. 1990. “How Serious Is the Nicaragua and Honduras.” Paper presented at the Neglect of Intra-Household Inequality?” Economic Journal. annual conference of the Latin American and Carib- Haddad, Lawrence, John Hoddinott, and Harold Alder- bean Economics Association, Puebla, Mexico, October. man (eds.). 1997. Intrahousehold Resource Allocation in Kiewisch, Elizabeth. 2015. “Looking within the House- Developing Countries: Methods, Models, and Policy. Baltimore: hold: A Study on Gender, Food Security, and Resilience Johns Hopkins University Press for the International in Cocoa-Growing Communities.” Gender & Develop- Food Policy Research Institute. ment 23 (3): 497–513. Hinks, Timothy. 2002. “Gender, Wage differentials and Kumar, Neha. 2014. “Agricultural Insurance: A Gender Discrimination in the new South Africa.” Applied Eco- Perspective from Rural Bangladesh.” Unpublished, nomics 34 (16): 2043–2052. International Food Policy Research Institute, Washing- Holmes, Rebecca, Nicola Jones, and Hannah Marsden. ton, DC. 2009. “Gender Vulnerabilities, Food Price Shocks Kumar, Neha, and Agnes Quisumbing. 2013. “Gendered and Social Protection Responses.” Background Note, Impacts of the 2007–2008 Food Price Crisis: Evidence Improving Social Protection Effectiveness: Understand- Using Panel Data from Rural Ethiopia.” Food Policy 38: ing Social and Economic Risks and Gender Inequali- 11–22. ties Series. Overseas Development Institute, London, Kumar, Neha, and Agnes Quisumbing. 2014. “Gender, England. Shocks, and Resilience.” 2020 Conference Brief, Interna- IFAD. 2012. Livestock Thematic Papers. Women and Pastoral- tional Food Policy Research Institute, Washington, DC. ism. Rome, Italy: International Fund for Agricultural Lambrou, Yianna, and Sibyl Nelson. 2010. “Farmers in a Development. Changing Climate—Does Gender Matter? Food Secu- International Livestock Research Institute (ILRI). 2010. rity in Andhra Pradesh, India.” Rome, Italy: Food and “Livestock and Women’s Livelihoods: A Review of the Agriculture Organization of the United Nations. 42 Agriculture Global Practice Discussion Paper Lim, Joseph. 2000. “The Effects of the East Asian Crisis Family Decisions.” Draft Chapter for the Handbook of on the Employment of Women and Men: The Philip- Development Economics. pine Case.” World Development 28 (7): 1285–1306. Thomas, Duncan. 1999. ”Intrahousehold Resource Allo- Lundberg, Shelly, Robert Pollak, and Terence Wales. cation: An Inferential Approach.” Journal of Human 1997. “Do Husbands and Wives Pool Their Resources? Resources 25 (4): 635–64. Evidence from the United Kingdom Child Benefit.” Thomas, Duncan, Dante Contreras, and Elizabeth Fran- Journal of Human Resources 32 (3): 463–480. kenberg. 2002. “Distribution of Power Within the Njuki, Jemimah, Susan Kaaria, Angeline Chamunorwa, Household and Child Health.” UCLA, manuscript. and Wanjiku Chiuri. 2011. “Linking Smallholder Udry, Christopher. 1996. “Gender, Agricultural Produc- Farmers to Markets, Gender and Intra-Household tion, and the Theory of the Household.” Journal of Dynamics: Does the Choice of Commodity Matter?”” Political Economy 104 (6): 1010–1046. European Journal of Development Research 23 (3): 426–443. UN Women. 2012. “Facts & Figures.” United Nations Njuki, Jemimah, Elizabeth Waithanji, Beatrice Sakwa, Entity for Gender Equality and the Empowerment Juliet Kariuki, Elizabeth Mukewa, and John Ngige. of Women at http://www.unwomen.org/en/news/ 2014. “A Qualitative Assessment of Gender and Irriga- in-focus/commission-on-the-status-of-women-2012/ tion Technology in Kenya and Tanzania.” Gender and facts-and-figures. Technology Development 18 (3): 303–340. United States Agency for International Development Otzelberger, Agnes. 2011. “Gender Responsive Strategies (USAID). 2013. “A Guide to Integrating Gender into on Climate Change: Recent Progress and Ways For- Agricultural Value Chains.” Based on Promoting Gender ward for Donors.” Brighton, United Kingdom, Insti- Equitable Opportunities in Agricultural Value Chains: A Hand- tute of Development Studies. book. Washington. DC.: USAID. Quisumbing, Agnes. 2011. “Do Men and Women Accu- van den Bold, Mara, Andrew Dillon, Deanna Olney Mar- mulate Assets in Different Ways? Evidence from Rural cellin Ouedraogo, Abdoulaye Pedehombga, and Agnes Bangladesh.” Discussion Paper 01096, International Quisumbing. 2013. “Can Integrated Agriculture- Food Policy Research Institute, Washington, DC. Nutrition Programs Change Gender Norms on Land Quisumbing, Agnes, and John A. Maluccio. 2003. and Asset Ownership? Evidence from Burkina Faso.” “Resources at Marriage and Intrahousehold Alloca- IFPRI Discussion Paper 01315, International Food tion: Evidence from Bangladesh, Ethiopia, Indonesia, Policy Research Institute, Washington, DC. and South Africa.” Oxford Bulletin of Economics and Statis- Wawire, Violet, K. 2011. “Gender and the Social and tics 65 (3): 283–328. Economic Impact of Drought on the Residents of Quisumbing, Agnes R., Jonna P. Estudillo, and Keijiro Turkana District in Kenya.” Gender Issues Research Report Otsuka. 2004. Land and Schooling: Transferring Wealth Series, No. 21. Mimeo. Organisation for Social Science across Generations. Baltimore, MD: Johns Hopkins Uni- Research  in Eastern and Southern Africa. OSSREA versity Press. Publications. Addis Ababa.  Quisumbing, Agnes, Neha Kumar, and Julia Behrman. WHO. 2009. Promoting gender equality to prevent vio- 2011. “Do Shocks Affect Men’s and Women’s Assets lence against women. Series of Briefings on Violence Differently? A Review of New Literature and Evi- Prevention: The Evidence. Geneva, Switzerland: dence from Bangladesh and Uganda.” Discussion World Health Organization. Paper 01113, International Food Policy Research Insti- World Bank. 2008. World Development Report: Agriculture for tute, Washington, DC. Development. Washington, DC.: The World Bank. Sen, Amartya. 1990. “Gender and Cooperative Con- World Bank. 2009. Gender in Agriculture Sourcebook. Wash- flicts.” In Persistent Inequalities, ed. Irene Tinker. New ington, DC.: The World Bank. York: Oxford University Press. World Bank. 2010. “Toolkit for Integrating Gender- Strauss, John, and Duncan Thomas. 1994. “Human Related Issues in Land Policy and Administration Proj- Resources: Empirical Modeling of Household and ects.” World Bank, Washington, DC. Gender and Agricultural Risk 43 World Bank. 2011. “Gender and Asset Ownership: A Agriculture Organization, International Food Policy Guide to Collecting Individual-Level Data.” Policy Research Institute, Africa Development Forum. Wash- Research Working Paper 4704, World Bank, Washing- ington, DC.: The World Bank. ton, DC. World Bank. 2015c. ICT in Agriculture. Connecting World Bank. 2012a. Living Through Crisis. How the Food, Fuel, smallholders to knowledge, networks, and institutions. and Financial Shocks Affect the Poor. New Frontiers of Social E-Sourcebook. Washington, DC.: The World Bank. Policy. Rasmus Heltberg, Naomi Hossain, and Anna World Bank. 2016. “Methodological Experiment on Reva, Editors. Washington, DC.: The World Bank. Measuring Asset Ownership from a Gender Perspec- World Bank. 2012b. World Development Report: Gender Equal- tive (MEXA). Technical Report. Living Standards ity and Development. Washington, DC.: The World Bank. Measurement Study (LSMS).” Washington, DC.: The World Bank. 2014. Voice and Agency: Empowering Women and World Bank. Girls for Shared Prosperity. Washington, DC.: The World World Bank and the ONE Campaign. 2014. Leveling the Bank. Field: Improving Opportunities for Women in Africa. Michael World Bank. 2015a. Agriculture Sector Risk Assessment. Meth- O’Sullivan, Arathi Rao, Raka Banerjee, Kajal Gulati, odological Guidance for Practitioners. Agriculture Global and Margaux Vinez (Eds). Washington, DC.: The Practice. Washington, DC.: The World Bank. World Bank. World Bank. 2015b. Confronting Drought in Africa’s Drylands. Raffaello Cervigni and Michael Morris (Eds). Food and 44 Agriculture Global Practice Discussion Paper Annex A Risk Management Strategies and Instruments The vulnerability of individual stakeholders and the agriculture sector as a whole depends on the nature of the risks (i.e., their correlation, frequency, timing, and sever- ity) and the effectiveness of the risk management instruments in use. It is unrealistic to suppose that all risks can be managed, as one solution or product cannot serve as a “silver bullet” for all risks in all circumstances. Indeed, a complex variety of strategies are discussed at length within the existing literature. Rather than reviewing all possible risk management strategies, the strategic framework used in these guidelines presents a simplified approach for risk management strategy development for illustrative and practical purposes. The conceptual framework can be adapted to be as complex or as simple as needed given a country’s circumstances. Following the assessment of risks and analysis of stakeholder vulnerability, risk man- agement strategies can be proposed. A practical way to identify solutions is by classify- ing possible risk management strategies into three categories: mitigation, risk transfer, and coping. The appropriate set of strategies depends in part on participants’ capacity to effectively use them. 1.  Risk mitigation (ex-ante): Risk mitigation strategies are actions taken prior to a risk event to reduce the likelihood of risk or the severity of losses. They are particularly useful for risks that occur with relatively high frequency but with lower impact intensity. Risk mitigation options are numerous and var- ied. Examples include: adoption of improved agronomic practices such as soil drainage and mulching, conservation farming, and the use of short duration and disease- and stress-resistant cultivars; irrigation and flood control infra- structure; soil and water conservation measures; changes in cropping patterns; crop and livestock diversification; income diversification; improved early warn- ing systems; and modern information and decision support systems. 2. Risk transfer (ex-ante): As not all effects of realized risks can be mitigated, risk transfer tools and mechanisms transfer the potential financial consequences of particular risks from one party to a willing third party, usually for a fee or premium. These mechanisms usually trigger compensation in the case of a risk-generated Gender and Agricultural Risk 45 loss (e.g., purchasing insurance, re-insurance, finan- and gaps in their current risk management strategies to cial hedging tools). While insurance and hedging enable them to adapt to a changing risk landscape. are well-known forms of risk transfer, in developing countries the use of informal risk transfer within fam- Output and priority measures ilies and communities is also extremely important. The end product (output) of the ASRA process is a set 3. Risk coping (ex-post): Some risks cannot be of actions that require special attention to reduce vul- mitigated or transferred, so risk coping strategies are nerability to shocks associated with the key priority risks. needed to help stakeholders better absorb and recover Moreover, resources are scarce and stakeholders need to from their impacts. These instruments improve the see an explicit risk prioritization and strategy identifica- affected population’s resilience to withstand and cope tion process in which the proposed interventions provide with events through ex-ante preparation to sustain pro- the clear returns in terms of agricultural growth, poverty duction and livelihoods following an event. Examples reduction, food security, or other objectives pursued by a include some form of compensation (cash or in-kind), country’s agricultural policies. The measures can then be social safety net programs, buffer funds, savings, stra- incorporated into government plans and budgets. tegic reserves, and livelihood recovery programs (e.g., government assistance to farmers, debt restructur- Risk management strategies are operationalized by three ing, contingent financing). Such interventions are main categories of instruments that can be planned, often financially beneficial, and the ability to quickly budgeted, and implemented for: respond to events often reduces losses. 1.  Policy reform: Improved risk management often entails policy reforms (e.g., legal or regulatory Figure A.1 illustrates these risk management strategies in reforms to improve access to agricultural inputs; the context of increasing layers of risk depending on the changes in information policy to make agricultural probability of occurrence (frequency) and the intensity or information easily accessible to all; changes in gov- potential to cause losses (severity). ernment policy related to price formation, govern- ment procurement, or strategic grain reserves). The combination of different activities selected to man- 2. Agricultural investment: While policy reforms age risk ultimately depends largely on the findings of the mainly require political will, other risk mitiga- ASRA, the characteristics of the identified risks, various tion measures can be costly. Examples are financial actors’ existing capacity to manage risk, and the fiscal investments in irrigation infrastructure, research into constraints to implementing an integrated strategy. This drought- and disease-resistant and pest-tolerant culti- framework can be applied to prioritize risks and interven- vars, soil and water conservation, weather infrastruc- tions in a country with many risk management practices ture, or updated agricultural services (e.g., agricultural already in place, whereby stakeholders identify priorities extension systems or disease surveillance systems). Some of these measures may already be part of a government program, with the ASRA simply calling Figure A.1. Risk management layers for additional investments to strengthen capacity in LAYER 3 those areas more vulnerable to external shocks. PROBABILITY LAYER 2 Very low 3. Technical assistance: Technical assistance is frequency, Low very high losses geared towards building local stakeholders’ capac- frequency, Risk mitigation ity (e.g., training in price risk management; fea- medium losses + Risk transfer LAYER 1 Risk mitigation + Risk coping sibility studies for various instruments; flood risk High frequency, + Risk transfer modeling work; development of early warning sys- low losses Risk mitigation tems). Recent developments in information systems addressing agricultural risks can be easily transferred to public and private institutions that can adapt the SEVERITY instruments to a country’s specific conditions. 46 Agriculture Global Practice Discussion Paper Annex B Analysis of Existing Material Preliminary analysis of agricultural supply chains often entails a comprehensive review of existing material (reports of agriculture sector reviews, studies, research the- ses, major news items, trade reports, etc.). Issues to be thoroughly researched include: »» Demand conditions. Identification of major markets of the commodities under study (export, regional, and domestic), end use of the commodity, character- istics of the commodity, quality attributes, demand and supply dynamics, and market trends. »» Supply chain structures. The current structure of supply chains, which includes the flow of goods, information, and finance, and the degree of concentration of different operators (in terms of volume or value). Agricultural supply chains for a single commodity can be composed of various sub value chains with distinct and different participants, depending on market demand. Different levels of technology, integration, women’s participation, and risk can be present. »» Stakeholders and their functions. Identification of the various actors in the agricul- tural supply chains (those directly affecting the flow of goods, as well as those providing support services) and their functions. A disaggregation of the various stakeholders in terms of relative size and gender. »» Performance. Analysis of time-series data of yields, area, and production for each of the agricultural commodities under study. This analysis includes disaggrega- tion in terms of gender and region (provinces). Understanding the frequency, intensity, and causes of yield volatility for the various crops revealed in the anal- ysis of time-series data is one of the key activities to support the findings of the risk assessment.10 »» Governance and coordination mechanisms. The dominant governance structures and coordination mechanisms and how they promote or inhibit the flow of infor- mation, risk sharing, and risk management across the sector. Illustrations of risk timelines for the agriculture sector as a whole and for commodities can be found in Annex C in 10 World Bank (2015a). Gender and Agricultural Risk 47 »» Public sector policies and interventions. The national and objective of risk management in the first place, but regional enabling environments (e.g., national/ addressing risk is oftentimes implicit in their activi- regional agricultural policies, discriminatory gender ties. Also oftentimes implicit are issues of gender regulations, level of trade cooperation/integration, discrimination to access government programs. etc.). Public investments and donor-funded pro- Understanding the existing risk management land- grams in the agriculture sector and their impli- scape lays the groundwork for the team to begin cations in terms of gender participation and risk identifying gaps. Once the literature review is con- management. Typically, government programs in cluded, a gap analysis can be completed during the agriculture are not necessarily designed with the field visits. 48 Agriculture Global Practice Discussion Paper Annex C Methodology for Estimating Indicative Value of Production Losses in Agricultural Crops Gender and Agricultural Risk 49 50 Agriculture Global Practice Discussion Paper Gender and Agricultural Risk 51 52 Agriculture Global Practice Discussion Paper Gender and Agricultural Risk 53 Annex D Guidelines for Assessing Risk and Capacity to Manage in Focus Groups with Farmers11 Basics. A focus group comprises people who have attributes in common, and who are able to provide information and opinions about the topic or subject that is the focus of discussion. Focus groups may be homogeneous or heterogeneous depending on the purpose of the focus group meeting (e.g., to identify a pattern, a homogeneous group is more appropriate, whereas to ascertain perceptions of risks affecting a whole supply chain, a heterogeneous group is more appropriate). Both ways are valid for assessing gendered differences in risk management.11 Although the Ministry of Agriculture staff assisted in the selection of focus groups for the risk assessments, the team should: »» Be careful that there are no significant power differentials among group mem- bers, as this often results in influential people (often men, or better-off individu- als) dominating the discussion. »» Aim to cover a mixed sample of representative farming entities covering (i) small and (ii) medium farmers. Within this continuum there may still be considerable differentiation within groups, which should also be accounted for. For example some small farmers (possibly in different regions) may have low input (e.g., lim- ited inputs, family labor) versus high input use (e.g., moderate input use, hire labor). The level of commercialization may also vary among medium/large enterprises and regions in the country. »» Keep the size of the focus group manageable. A focus group is most effective with 10–15 participants. The team could conduct a minimum of three (3) farmers focus group discussions: on the major cash crop, on all food crops, and one more cash crop at the consideration 11 Adapted from World Bank 2015a. 54 Agriculture Global Practice Discussion Paper of the field team member. In supply chains where women »» Have sufficient stationery, including copies of all participation is high, the team needs to organize special- report formats (to be developed), notebooks, focus ized focus groups with women farmers. on clipboard (for drawing, writing, or posting), tape, pens, pencils, markers, post-its. Purpose. A focus group discussion is one of the tools (alongside literature review, secondary data collection, Developing a facilitator’s/moderator’s guide, or key-informant interviews, etc.) to help assess risk. Its pur- script. The script is a guide for the facilitator to explain pose therefore is to help the team: to individuals the purpose of the group, review the focus »» Identify risks (three principle types of agricultural group rules, and other information that may be important risk: production, market, and enabling environ- to provide to the participants. Suggested script/guide template ment risk) for all field teams to follow: »» Analyze and quantify risks (primary stakeholder; Opening (10 Minutes): losses; frequency; underlying causes; risk transmis- sion); prioritize risks (frequency of occurrence and “Hello. I will introduce myself and then I would like severity of impact); and each of you to tell us your name, what you do, and »» Identify current capacity to manage and vulner- whether you have participated in a similar discus- ability (i.e., sale of assets, borrowing, sale of labor, sion before. My name is (. . .). [Have the focus group reduce consumption, government handouts, members introduce themselves]. Today we would like migration, etc.). to have a conversation with you about risks in your agricultural activities, how they impact you, and The purpose of the focus group discussions with farmers how you manage. We have been asked by the gov- is to assess their capacity to manage by: identifying farm- ernment to assess the risks affecting the agriculture ers’ perceptions on the risks that they face in the supply sector and suggest measures to better manage them. chain, examining how these risks have negative impacts, We have selected major food and cash crop supply and identifying gaps in how these risks could be managed chains across the country, and just like we are here more effectively. today with you, there are other teams in other parts of the country doing the same with farmers in [list Stage One: Preparing some commodities and some of the other regions]. What we are trying to accomplish before we leave for a Focus Group here today is to get a better understanding of the risks Discussion that you face in the supply chain, to asses what is your Prior to engaging in a focus group discussion, the field capacity to manage these risks, and how they could team needs to: be managed more effectively. We plan that this will »» Identify and focus on the key purpose of this discus- take us approximately two hours. Are there any ques- sion, and key questions (please see guide/script below) tions?” [Respond to participant questions] »» Appoint a facilitator/moderator who will guide the “Before we continue, let’s agree on the rules of the content and the process of the discussion discussion [decide together on whether you want focus group »» Appoint a note-taker to record the focus group members to participate by raising their hands to speak; write their feedback (the facilitator/moderator can take notes, ideas on a piece of paper, etc.] I would like to tell you that too, but should be careful that it does not nega- everything we discuss here will be kept confidential. tively affect the flow of the discussion if the facilita- We are not here to check on licenses or for tax collec- tor stops to write things down) tion purposes. We will summarize the things you tell us »» Develop a facilitator’s/moderator’s guide, or script and combine it with other focus groups we are meet- »» Confirm whether a translator is needed (a field ing across the country. My job here today is to make team member should be able to assume this respon- sure we discuss all of the issues we planned to discuss, sibility too, if he/she is capable of doing so) and my colleague here will help us by taking notes. Gender and Agricultural Risk 55 “Let’s begin.” (40 minutes) engage in wealth breakdown: the group can be asked to con- sider 2 sized groups, ranging from the small poorest to the [All discussions should cover four main areas: (i) livelihood larger-farm units. Visually this could be represented with profile; (ii) risk exposure; (iii) capacity to manage risk; and colored stickers or similar. This differentiation can prob- (iv) gaps and opportunities to address risk, and be structured: ably help us at the end of the process identify easier how first a question, then a probe. Please see Stage Two]. important are the expected losses for different participants Closure (10 minutes) in the chain, relative to their assets, livelihood/enterprise “Are there final questions? [Respond to questions] Let strategies, and performance outcomes. me re-iterate what I think we heard the risks that you face in the supply chain are, and how these risks and Proceed with the opening according to the script, and start the negative impacts could be managed more effectively discussion with the agreed method of participation. [re-iterate]. Thank you for participating in our focus All field teams should be prepared to discuss the five main group today. We will be collecting feedback from all areas identified above as (i) livelihood profile; (ii) risk exposure; the focus group discussions undertaken across the (iii)  risk management; and (iv) gaps and opportunities to address country, and together with other analysis included it risk. Below is a table with indicative questions under each in our discussion with the Government.” area. This is not a questionnaire to be strictly followed and completed. It is meant to help the facilitator/moderator Stage Two: Conducting to keep the discussion organized. Questions asked should Focus Group Discussion in general be open-ended to stimulate both individual Since we are aiming to understand what risks and par- contributions and group interaction. Questions should ticipants capacity to manage them, it might be worth at not elicit “yes/no” answers. the beginning of the discussion to have the focus group participants Agricultural/Livelihood Profile Question: What are the main crops you produce for own food consumption; Probe: % area of value, total area farmed & owned (ha), rank them in for cash; for export. importance. Question: Is livestock important for you? Probe: Do you consume livestock products that you produce? Do you derive income from livestock products? What percentage of income? Question: What percentage of your household income is from crops, Probe: Try to find out what role agriculture plays in the household livestock, other? Is your average level of production and revenue in a given livelihood. year? What has been the trend in recent years? Question: What inputs do you use? Where do you source inputs from? Probe: Fertilizer, seeds, pesticides. Check for reliability, quality, and utilization. Question: Do you have regular input procurement arrangements? With Probe: Timely provision of inputs, cost factors, logistics issues. Formality of whom? How effective are existing input arrangements? arrangements. Question: Do you have fixed selling (contract) arrangements with processors Probe: Formality of contracting, length of trading relationships. Logistics. or other intermediaries? How often are these negotiated? Question: Is there any type of irrigation used? What type of irrigation Probe: Furrow, drip, overhead, etc. is in use? Question: What are the factors that motivate planting of crop? What are the Probe: Return to assets v. risk management, substitute, complementary alternatives? goods. Question: What access do you have to local markets and traders? What is Probe: Formal v. informal markets. the distance to the nearest trading center? Question: What access do you have to financial resources? Probe: Formal v. informal finance (credit, trade finance, personal). Question: Are you a member of a cooperative/organization? What are the primary benefits of this relationship? 56 Agriculture Global Practice Discussion Paper Risk Exposure Question: In broad terms, what are the main sources of risk you face in: Probe: Nature of risk (production, market, enabling environment). Probe   Sourcing inputs? against specific risk factors impacting on farm kevel e.g.: weather, price,  Production? environment, labor standards, logistics, operational, trade policies.   Sales/Marketing of goods? Question: What is the magnitude of the direct negative impacts that Probe: e.g., High, medium, or low impacts on: income, food consumption, potentially arise from these risks? indebtedness, wellbeing. Temporal impacts? Question: What crops are more vulnerable to risks? Probe: Probe against specific production risks (weather, SPS), market risks (prices),and enabling environment. Question: What are the three main sources of risk that most concern you? Probe: Ranking of potential problem “areas.” Question: Of the risks identified, what is their frequency? Probe: Seasonal, annually, 1 in 2 years, 1 in 5, 1 in 7. Question: How would you describe the potential severity of impact and Probe: Expected loss—minimal, low, medium, high, very high. expected losses arising from major risks? Question: Overall, are conditions in the supply chain, and your position in Probe: Get a sense of long-term threats/opportunities to participants. particular, deteriorating/improving in recent years? Note: You may wish to probe specific risk aspects related to seasonality dimensions, contracting arrangements, direct impact of weather and environmental factors. You are trying to corroborate frequency of risk events; identify the causes of losses; could be single cause, or multiple causes (percentage). Risk Management Question: What is done to address problems in advance of a risky event? Probe: Ex ante risk management strategies—investments in infrastructure, How long have these actions existed? technology, management practices, financial instruments, organizational arrangements. Question: What is done to address problems after a risky event? Probe: Ex post risk management strategies—reallocation or sale of assets (livestock), seek employment/migration, transfers, resort to savings, borrow from friends or family, etc. Question: How effective have these actions been? What actions have been Probe: Ex ante and ex post risk management strategies. most effective? Least effective? Why? Question: Who typically provides these actions? Probe: Self-made decisions, decisions by farmer organizations, formal v. informal mechanisms. Question: What interventions have been supported by public sector “agents” Probe: Role of public sector/Government agencies v. market-based actions v. (including donors/NGOs) to manage problems? community-level. Ex ante v. ex post risk. Question: How effective have public interventions been? Which are more/ Probe: Timing, targeting, delivery aspects. less effective? Question: What has recent experience illustrated about farmer capacity to Probe: Ability to manage risk on own v. need for external “partners.” withstand major deviations, disruptions, and disasters in the supply chain? Question: What information sources, if any, are used to predict/assess the Probe: Early warning information, price tracking, local knowledge. potential frequency/magnitude/severity of problems? Question: How would you describe overall access to credit and insurance? Probe: Availability, affordability or credit and timely/“fair” payment of What are the benefits/costs from credit and/or insurance? insurance. Note: You are trying to understand impact of events (short term, medium term); how losses are transmitted upstream and downstream along the supply chain; how different stakeholders manage these risks; how small/medium/larger farmers manage in terms of mitigation, transfer or coping; how shock was absorbed; were there any stakeholders out of business. Gender and Agricultural Risk 57 Gaps and Opportunities to Address Risk Question: What are the main lessons learnt from past experiences in risk management? Question: What options could be explored to manage risks affecting farmers Probe: Production/Market/Enabling environment. Opportunities and better? constraints. Question: What are the perceived potential options for managing problems jointly with other supply chain entities? Question: What roles could others play? Probe: Community; Public sector/Government; Private sector, NGOs, Donors. Remember: opportunity to double-check major findings, and also identify top-of-mind patterns and themes. You are taking time out of focus group members’ busy schedules. Keep it within the time limit you promised it Stage Three: Reporting will take! Focus Group Discussion As soon as possible after ending focus group discussion Ending Focus Group Discussion (ideally that same evening with the team) proceed with Proceed with the closure according to the script, and debrief reporting the results of the focus group discussion, using with focus group discussion participants immediately after this format (copies of the report format should be pro- each focus group when information is fresh. It is a good vided to all field teams). Focus Group Discussion No. ____________________ Region Name: ___________________________ City/Village Name: ___________________________ Livelihood zone: ___________________________ Date: ___________________________ Team Member Names: ___________________________ 58 Agriculture Global Practice Discussion Paper Livelihood Profile Main Food Crop Production (Rank Importance of Small Medium Contribution to Diet) Farmers Farmers Comments 1 2 3 4 5 *Rank from 1: Most important to 5: Least important Main Cash Crop Production (Rank Importance of Small Medium Contribution to Income) Farmers Farmers Comments 1 2 3 4 5 Small Medium Livestock (Types/Size Held) Farmers Farmers Comments Narrative description of supply chain segment interviewed: Explain main characteristics of supply chain segment (participants, interactions) and general trend. Gender and Agricultural Risk 59 Risk Exposure Food Crop Small Medium Comments (Impact: Moderate, (Rank Exposure to Risk) Farmers Farmers Considerable, Critical, Catastrophic) 1 2 3 4 5 *Rank from 1: Most exposed/Most vulnerable to risky event to 5: Least exposed/Least vulnerable to risky event Food Crop Small Medium Comments (Impact: Moderate, (Rank Frequency of Risk) Farmers Farmers Considerable, Critical, Catastrophic) 1 2 3 *Rank from 1: High Probability to 3: Low Probability Cash Crop Small Medium Comments (Impact: Moderate, (Rank Exposure to Risk) Farmers Farmers Considerable, Critical, Catastrophic) 1 2 3 4 5 *Rank from 1: Most exposed/Most vulnerable to risky event to 5: Least exposed/Least vulnerable to risky event Cash Crop Small Medium Comments (Impact: Moderate, (Rank Frequency of Risk) Farmers Farmers Considerable, Critical, Catastrophic) 1 2 3 *Rank from 1: High Probability to 3: Low Probability Risks for Food Risks for Cash Crops Frequency Impact Crops Frequency Impact 60 Agriculture Global Practice Discussion Paper Current Risk Management Practices Capacity to Manage Risk Cash Crop Mitigation with These Strategies Risk Event (e.g., Drought (It Would Be Good to (e.g., Drought, Tolerant Seed Coping (e.g., Have This by Small Output Price, etc.) Varieties, etc.) Transfer Borrowing, etc.) and Medium Farmers) Capacity to Manage Risk Food Crop with These Strategies Risk Event (It Would Be Good to (e.g., Drought, Have This by Small Output Price, etc.) Mitigation Transfer Coping and Medium Farmers) Gender and Agricultural Risk 61 Suggested Gaps and Opportunities to Address Risk Opportunities Cash Crop for Public Sector/ Opportunities Opportunities Risk Event Gaps Government for Private Sector for NGOs/Donors Opportunities Food Crop for Public Sector/ Opportunities Opportunities Risk Event Gaps Government for Private Sector for NGOs/Donors Narrative description: A sense of prioritization? Let’s discuss the dynamics of the format of interaction. group and farmers raise their hands for suggested answers; Options: (i) written questions with closed answers to fill in (iii) Open questions to the group and open answers and by farmers like multiple choice; (ii) Open questions in a team just take note; (iv) etc. 62 Agriculture Global Practice Discussion Paper A g r i c u lt u r e g l o b a l p r a c t i c e d i s c u s s i o n p a p e r 1818 H Street, NW Washington, D.C. 20433 USA Telephone: 202-473-1000 Internet: www.worldbank.org/agriculture Twitter: @WBG_agriculture