Measuring livelihood resilience: The Household Livelihood Resilience Approach (HLRA)

World Development 107 (2018) 253–263

Contents lists available at ScienceDirect

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

Measuring livelihood resilience: The Household Livelihood Resilience Approach (HLRA) Amy Quandt Jornada Experimental Range, New Mexico State University, P.O. Box 30003, MSC 3JER, Las Cruces, NM 88003-8003, United States

a r t i c l e

i n f o

Article history: Accepted 28 February 2018

Keywords: Sustainable livelihoods approach Livelihoods Resilience Development Agroforestry Kenya

a b s t r a c t The concept of resilience, and livelihood resilience more specifically, is growing in prominence with international development and humanitarian organizations that aim to measure and build resilience to specific disturbances such as floods or droughts. However, measuring livelihood resilience is a difficult task, and practical methods to measure livelihood resilience, as well as analyze and visualize the data are needed. In this paper, I introduce the Household Livelihood Resilience Approach (HLRA), which draws from the sustainable livelihoods approach and it’s five capital assets to measure resilience. However, unlike other approaches that use the five capital assets such as utilized by Zurich Insurance Group, Ltd and the IFRC, the HLRA goes farther to help visualize the results and identify specific actions to build resilience. This paper illustrates the effectiveness of the HLRA through an empirical case study where this approach was used to measure livelihood resilience in Isiolo County, Kenya, and the effectiveness of agroforestry in building livelihood resilience for agricultural households. Drawing from this case study, I suggest five ways that the HLRA improves upon previous frameworks, including 1) providing practical methods and tools, not just a theoretical framework, 2) integrating ‘subjective’ measures of livelihood resilience, 3) focusing on the household-scale instead of community-scale or larger, 4) providing methods to analyze, visualize, and interpret results of livelihood resilience measures, and 5) highlighting the importance of human agency, power, and access to assets. The HLRA has the potential to help organizations identify specific interventions that could help build livelihood resilience for the most vulnerable groups of people within a community. Published by Elsevier Ltd.

1. Introduction Over the last few years the concept of resilience has gained prominence with international development and humanitarian organizations (Jones and Tanner, 2015; Walsch-Dilley et al., 2016). The notion of resilience is being employed with the aim to sustainably manage resources for both ecosystem function and human development and wellbeing (Berkes et al., 2003). However, Tanner et al. (2015) propose that the lens of resilience ‘‘requires greater attention to human livelihoods if it is to address the limits of adaptation strategies and the development needs of the planet’s poorest and most vulnerable people (pg. 23).” Thus, the concept of livelihood resilience specifically has been growing as livelihoods are increasingly caught in major global transitions in climatic, economic, and social systems. For example, livelihood resilience is acknowledged both explicitly and implicitly in a range of the United Nation’s Sustainable Development Goals for 2030 (Bahadur et al., 2015). Resilience-building efforts are also key aspects of E-mail address: [email protected] https://doi.org/10.1016/j.worlddev.2018.02.024 0305-750X/Published by Elsevier Ltd.

humanitarian organizations including the International Federation of Red Cross and Red Crescent (IFRC) and Zurich Insurance Group, Ltd (Zurich). However, resilience is often criticized for being difficult to empirically measure (Carpenter, Walker, Anderies, & Abel, 2001), thus creating a challenge when aiming to understand or promote livelihood resilience. While methods for measuring livelihood resilience have been proposed by others, many of these methods have not been empirically tested (Speranza, Wiesmann, & Rist, 2014), focus more on social resilience compared to overall livelihood resilience which includes environmental and natural indicators important to livelihoods (Obrist, Pfeiffer, & Henley, 2010), while others focus on only a subset of indicators such as Lindstädter et al. (2016) who focus their analysis solely on agronomic and institutional indicators. Further, much of the discussion on livelihood resilience takes place at the community-level or larger scales, and often overlooks the importance of household-level analyses and human agency in building livelihood resilience (Berkes and Ross, 2013; Brown and Westaway, 2011; Cabell and Oelofse, 2012; Cutter et al., 2008; Elasha et al., 2005; Erenstein, Hellin, & Chandna, 2010; Lindstädter et al., 2016; Szoenyi et al.,

254

A. Quandt / World Development 107 (2018) 253–263

2016). For example, in practice, resilience strategies in Africa are widely applied at the community-level (Otsuki et al., 2017; United Nations Development Program 2010). This paper aims to fill these gaps and introduces one innovative methodological and analytical approach for measuring livelihood resilience at the household-scale called the Household Livelihood Resilience Approach (HLRA). This method uses on-the-ground quantitative and qualitative data to measure livelihood resilience between different households and individuals within a community. It also provides analytical methods to identify the most vulnerable groups and identifies specific pathways for building livelihood resilience. This approach draws from the sustainable livelihoods approach and five capital assets to measure resilience as is done by Zurich and the IFRC, but it improves upon previous approaches in five major areas. First, the HLRA provides not only a theoretical framework for measuring resilience, but also a practical method and applicable tools for measuring livelihood resilience. Second, it includes ‘subjective’ measures of resilience, moving beyond top-down objective measures of livelihood resilience and allowing for context-specific indicators of livelihood resilience. Third, it focuses on the household and individual scale and the importance of understanding that communities are heterogeneous and diverse individuals or households within a community have different levels of livelihood resilience. Fourth, it outlines methods for analyzing, visualizing, and interpreting the results, and identifying specific actions to build resilience. Lastly, the HLRA highlights the importance of human agency and power relationships in livelihood resilience and access to assets with which households build their livelihood resilience. This paper illustrates the effectiveness of the HLRA through an empirical case study where this approach was used to measure livelihood resilience in Isiolo County, Kenya and the effectiveness of agroforestry in building livelihood resilience. 1.1. Resilience Resilience is defined by Walker and Salt (2006) as the ‘‘capacity of a system to absorb disturbance and still retain its basic function and structure.” There are two main aims of building resilience: to prevent the system from moving to an undesired, alternative regime in the face of change, and to nurture and preserve the components of the system that build resilience and allow the system to renew and reorganize after a disturbance (Walker et al., 2002). Broadly, the concept of resilience is a promising tool for exploring adaptive changes towards sustainability because it provides a way for analyzing how to maintain stability in the face of change (Berkes et al., 2003). In order to effectively build resilience, it is important to ask the questions ‘resilience of what’, ‘resilience to what’, and ‘resilience for whom’ (Lebel et al., 2006; Walker and Salt, 2006; Nelson and Stathers, 2009)? Much has been written about the development and applications of resilience thinking, and I will not repeat this work here (for a more comprehensive review see Walker and Salt, 2006; Gunderson and Holling, 2002, to name a few). 1.2. Livelihood resilience While resilience thinking has been praised by some, it has also attracted some criticism. For example, it is often seen as highly context specific which challenges its implementation through policy mechanisms (Cooper and Wheeler, 2015). Resilience thinking has also largely focused on natural systems and is often criticized for ignoring the social or political side of social-ecological systems (Brown, 2014). One response to these criticisms has been the development of a livelihood perspective in resilience thinking. Tanner et al. (2015; 23) define livelihood resilience as ‘‘the capacity

of all people across generations to sustain and improve their livelihood opportunities and well-being despite environmental, economic, social, and political disturbances.” Focusing on livelihood resilience places people in the center of analysis and highlights the role of human agency, rights, and capacity to prepare for, and cope with shocks (Tanner et al., 2015). A livelihood resilience approach expands our understanding of resilience beyond ‘‘. . .technical approaches to minimizing harm and loss by bringing issues of people’s lives, rights, justice, politics, and power to the fore” (Tanner et al., 2015, 23). Society’s ability to manage resilience resides in actors, social networks, and institutions (Lebel et al., 2006). A livelihood approach also strengthens resilience theory by acknowledging that people’s circumstances, cultures, values, and perceptions impact their ability to adapt (Enns and Bersaglio, 2015). Further, concentrating on livelihood resilience helps address the question of resilience of what and resilience for whom by focusing on the resilience of people’s livelihood strategies. Central to livelihood resilience are the coping strategies used by households or individuals during times of stress. These coping strategies can be spontaneous, but often involve planning and preparation for certain shocks. Coping strategies are specific responses or activities used to adjust to changing conditions, both short and long-term, and do not only happen during periods of stress, but are often intensified in such events (Adger, 2003; Mosberg and Eriksen, 2015). Building livelihood resilience means that a given household’s livelihood strategies and activities are better prepared to cope and manage the impacts of shocks, navigate uncertainty, and adapt to changing conditions (Marschke and Berkes, 2006). According to Allison and Ellis (2001), the most robust livelihood system is one displaying high resilience and low sensitivity, while the most vulnerable displays the opposite. Shocks to livelihoods can come from the environmental realm, such as climate change, or from the political-economic system, including crop price fluctuations or political instability. Lastly, it is important to differentiate livelihood resilience from social resilience. Cote and Nightingale (2012) define social resilience as the ability of communities to cope with stresses as a result of social, political, or environmental change. Social resilience is said to be increased through improvements in communications, risk awareness, the development and implementation of disaster plans, the purchase of insurance, and sharing of information (Cutter et al., 2008). While not mutually exclusive, livelihood resilience applies a broader set of indicators to resilience measurement, such as natural and human assets, while indicators of social resilience can be more narrowly focused, and in many cases is similar to social capital within the sustainable livelihoods framework. For example, when measuring the resilience of the social system, Lindstädter et al., 2016 only include agronomic and institutional indicators. The multi-disciplinary, broad set of contextual indicators is one advantage of the HLRA. 1.3. Sustainable livelihoods The concept of sustainable livelihoods was first introduced by the Brundtland Commission on Environment and Development, advocating sustainable livelihoods as a broad goal for poverty eradication (Krantz, 2001). The sustainable livelihoods approach developed as a form of livelihood analysis that has been used by a number of development organizations including the Department for International Development of the United Kingdom (DFID), the United Nations Development Program, CARE, and Oxfam (Adato and Meizen-Dick, 2002). The sustainable livelihoods approach is methodologically based in participatory research, applied anthropology, and rapid rural appraisal (Chambers, 1994; Krantz, 2001; Thulstrup, 2015). This approach states that livelihoods should be considered in terms of people’s access to capital assets

Cooperative societies, selfhelp groups

Education, skills, knowledge, health, nutrition, and labor power

Networks that increase trust, ability to work together, access to opportunities, reciprocity; informal safety nets; and membership in organizations

Knowledge, sills, health, labor availability

Adherence to rules, relationships of trust, mutuality of interest, leadership, kin and ethnic networks, social organizations

Labor including skills, knowledge, ability to work

Access to markets, representation and access to the ‘state’

Included in financial capital

Capital base including cash, credit, savings and basic infrastructure and production equipment and technologies

Skills, knowledge, ability of labor, and good health

Social resources including networks, social claims, affiliations, associations

Included in financial capital

Financial/Economic Capital

Human Capital

Social Capital

Physical Capital

Households assets, agricultural implements, infrastructure

Transportation, roads, buildings, water supply, sanitation, energy, technology and communication

Irrigated area, farm mechanization, distance to nearest town, access to paved roads

Female literacy, immunizations, work participation, population density

Savings, credit, as well as inflows such as state transfers and remittances Credit, savings, remittances Infrastructure and tools/equipment

Environmental services, natural resource stocks such as soil, water, air Natural Capital

Farm size, herd size, bank facilities, credit societies

Erenstein et al. (2010)

Annual rainfall, soil capability index, farm size, herd size

Adato and Meizen-Dick (2002)

Land, water, forests, marine resources, air quality, erosion protection, and biodiversity

Campbell et al. (2001)

Soil fertility, water resources, forest resources, grazing resources, land quantity and quality

Tacoli (1999)

Freshwater availability, land management, agricultural space, land

Scoones (1998) Type of Capital

Table 1 The five livelihood capitals as described by various authors.

A. Quandt / World Development 107 (2018) 253–263

255

(financial, physical, natural, human, and social), the ways in which people combine these capital assets to create livelihoods, and how they are able to enlarge their asset base through interactions with other actors and institutions (Chambers and Conway, 1992; Carney, 1998; Scoones, 1998; Johansson, 2015). At the core of the sustainable livelihoods approach are the five capital assets needed for a sustainable livelihood: financial, physical, natural, human, and social (see Table 1). These capital assets constitute a stock of capital that can be stored, accumulated, exchanged, or allocated to activities to generate an income, means of livelihoods, and other benefits (Rakodi, 1999; Babulo et al., 2008). Livelihood capitals may be accumulated so that reserves and buffers are created for times of stress or shocks (Scoones, 1998). While these five capitals can, and do, overlap, they encompass different types of assets needed for sustainable livelihoods as described by a variety of authors (Tacoli, 1999; Campbell et al., 2001; Adato and Meizen-Dick, 2002; Erenstein et al., 2010). Particularly there are overlaps between financial and natural capital as natural capital can create financial capital and vice versa. A household is assumed to need a balance of these five capitals in order to maintain adaptive capacity and well-being (Jacobs et al., 2015). For example, minimum levels of human and social capital are necessary to effectively make use of natural, physical, and financial capital (Jacobs et al., 2015). However, an accumulation of assets does not automatically shield households from shocks, and research on smallholders in Kenya found that they easily slipped back into poverty, particularly in areas with high physical insecurity (Ulrich et al., 2012). Additionally, there are also tradeoffs between the five livelihood capitals. For example, financial capital may be diminished in order to build up human capital by paying school fees. Weighing the trade-offs between these five types of livelihood capital is an ongoing process for households and individuals. Further, the five capital assets can be constrained by the transformation of structures and processes within the wider society (Obrist et al., 2010). Lastly, according to Bebbington (1999), people’s assets are not merely a means for which they make a living, but they also give meaning to the person’s world, they give people the capability to be and act. The five capital assets are not only inputs into the livelihood system, but are also outputs (Bebbington, 1999). Some of the strengths of the sustainable livelihoods approach is that it draws attention to the multiplicity of assets that people make use of when constructing their livelihoods (Krantz, 2001), and seeks to understand changing combinations of livelihood activities in a dynamic and historical context (Serrat, 2010). It moves beyond a focus on monetary measures to more adequate multi-dimensional understandings of livelihoods. (Rakodi, 1999). However, it has been critiqued for not sufficiently accounting for power relationships and politics (Scoones, 2009), underplaying macroeconomic trends and conflict (Serrat, 2010), a lack of rigorous attempts to deal with long term change (Scoones, 2009), and being expert-driven (Jones and Tanner, 2015). 1.4. Integrating sustainable livelihoods into resilience measurement: The Household Livelihood Resilience Approach Measuring resilience is a difficult task and several authors have put forward ideas about how to measure resilience (Carpenter et al., 2001; Carpenter, Westley, & Turner, 2005; Walker and Salt, 2006; Nadasdy, 2007; Nelson and Stathers, 2009; Leslie and McCabe, 2013). Because resilience is not directly measurable, most of the approaches above make use of quantifiable surrogates or indicators of resilience (Jones and Tanner 2015). Each of these authors use different methods for determining indicators of resilience. For example, Carpenter et al. (2005) talk about four general approaches to determine indicators of resilience including stakeholder assessments, model explorations, historical profiling, and

256

A. Quandt / World Development 107 (2018) 253–263

case study comparisons. However, there is no standard protocol for determining indicators of resilience, nor should there be necessarily. Instead, a comprehensive strategy for ensuring that major social and ecological aspects of resilience are included for measurement protocols is important. In this paper, I propose that using the sustainable livelihoods approach provides one innovative method for determining indicators of resilience. The five livelihood capitals can be used to organize indicators of resilience into these five different categories. As Campbell et al. (2001) state, ‘‘the capital assets approach to livelihoods may be an appropriate organizing principal for the selection of indicators of system performance.” It serves as a way to ensure that a variety of indicators are considered, including material, social, and natural factors that may help to measure and ultimately build resilience. The sustainable livelihoods approach acknowledges that there are important nonmonetary factors to livelihood resilience. Ultimately, resilience is a key component of sustainable livelihoods and vice versa (Thulstrup, 2015). Fig. 1 outlines the conceptual framework used here summarizing how to build livelihood resilience drawing from the sustainable livelihoods approach. Fig. 1 highlights the importance of access to livelihood capitals for building livelihood resilience, as well as how the historical and current context, conditions, trends, and institutions can shape if a household has sufficient access to livelihood capitals to build resilience livelihood strategies, or not. Diversity is also critical for increasing a livelihood’s ability to cope with change (Ellis, 2000; Hodbod and Eakin, 2015). Livelihood diversification can be different in different contexts. For example, it is sometimes used to accumulate resources, while others employ diversification to help spread risk or cope with temporary crises, and furthermore others use it as a response to longer-term declines in incomes or resources or due to large scale economic or environmental changes beyond local control (Hussein and Nelson, 1998). Using the sustainable livelihoods approach to measuring resilience ensures that a diversity of indicators are used to measure livelihood resilience, instead of indicators focusing just on financial capital. Therefore, the accumulation of livelihood capital assets may help increase the ability of households to respond to shocks with a greater diversity of potential decisions and actions. While using a sustainable livelihoods approach to measure resilience has been used in a handful of studies, it has not been widely adopted (Scoones, 1998; Campbell et al., 2001; Elasha et al., 2005; Erenstein et al., 2010; IFRC, 2015; Thulstrup, 2015; Szoenyi et al., 2016). While these studies cited here provide examples of how the sustainable livelihoods approach has been used to measure livelihood resilience, most examples are lacking in one area or another. Here, I will outline some of the major drawbacks or oversights of past attempts to integrate the sustainable livelihoods approach into resilience measurement.

Effective Institutions, Organizations, Governments Context, Conditions, Trends Livelihood able to cope with change or disturbance in context, conditions, trends

Sufficient Access to Livelihood Capital Assets or Resources

Resilient Livelihood Strategies

Fig. 1. Understanding how to build livelihood resilience with the Household Livelihood Resilience Approach.

First, some attempts provide only a theoretical framework instead of practical tools and methods for measuring livelihood resilience. For example, DFID (2011), state that ‘‘resilience enhancing activities can be usefully classified using the ‘assets pentagon’ from the sustainable livelihoods framework – social, human, physical, financial, and natural.” However, they provide no further guidance on how to accomplish this in the field. This makes the framework difficult to operationalize and use. Further, Speranza et al. (2014) developed an indicator framework for assessing livelihood resilience using the five capital assets that does include human agency; however, they admit that the framework has yet to be systematically tested. This is a concern because even a framework that looks effective in theory might not be successful practically. The HLRA has been empirically tested in the case study presented later in this paper. Second, most efforts to measure resilience have largely focused on the use of ‘objective’ frameworks focused on a range of observable socioeconomic variables and large datasets (IFRC, 2015; Jones and Tanner, 2015). Jones and Tanner (2015) advocate for the use of an alternative measurement of ‘subjective’ resilience which stems from the idea that people generally have a good understanding of the factors that contribute to their ability to plan for and cope with disturbance and change. This is a self-assessment of resilience that focuses on a more bottom-up process than the top-down measurements that have traditionally been the method of resilience measurement. Jones and Tanner (2015) state that subjective resilience can add value to objective methods because: 1. people have a good understanding of their capacity to deal with disturbance, 2. subjective measures can help to reduce uncertainty in the selection of indicators, 3. it allows insight to be gained on resilience in contexts where accurate, large datasets are inadequate, and 4. resilience is heavily shaped by sociocultural and psychological factors including risk perception and personal and cultural values. Brown and Westaway (2011) also promote the use of subjective measures which should be viewed as reflexive, dynamic, and differentiated temporally and socially. The idea of subjective resilience can, and arguably should, be integrated into the development of indictors in the HLRA and was used to develop indicators/surrogates of livelihood resilience for the case study that I will present below based on my previous work in the case study area of Isiolo County, Kenya (Quandt and Kimathi, 2016; Quandt and Kimathi, 2017). I will also utilize the idea of subjective resilience in order to validate using the sustainable livelihoods approach to measuring resilience by comparing the results of the HLRA with self-reported levels of well-being. Third, many efforts to use the five livelihood capital assets to measure livelihood resilience focus on the community-scale (Berkes and Ross, 2013). In these frameworks, those measuring livelihood resilience often assume that households within communities have similar access to livelihood assets. However, Agrawal and Gibson (1999) argue that this approach is flawed because communities are rarely made up of a unified, homogenous group of people. For example, Szoenyi et al. (2016) explain how the Zurich Insurance Group is using the five livelihood capitals to measure community resilience to flooding. However, the framework presented by Szoenyi et al (2016) does not acknowledge diversity in livelihood resilience within a community and among groups of people. While measuring community livelihood resilience may meet the goals of certain organizations, the HLRA provides a prospective that focuses on the household, and not an aggregated group of households called a ‘community’. Fourth, there is a current need for effective methods to analyze, visualize, and interpret the results of livelihood resilience measurements. Szoenyi et al. (2016), voice a need for a method for visualizing and analyzing the measurement results, as well as guidance on how to interpret their results. One good example of

A. Quandt / World Development 107 (2018) 253–263

visuals is presented in Erenstein et al. (2010), who created poverty maps aimed at building resilience using the five capital assets. Their maps illustrate how the livelihood capital assets vary in Nepal at the district-level. Using their mapping technique, they found that a solid foundation of all five assets is generally needed for livelihood security and to enable people to rise above the poverty line (Erenstein et al., 2010). However, these maps present data at the district-level, not the household-level. Below, I will present examples of effective visualizations based off the HLRA. Lastly, other efforts to measure livelihood resilience using the sustainable livelihood approach ignore the human agency in building livelihood resilience and focus on presence of assets in a given area, instead of an individual’s ability to access and control those assets as outlined in Fig. 1. This highlights how issues of power are often left out of resilience measurements, when indeed having power and access to livelihood assets is critical. For example, the IFRC (2015) developed a framework to measure resilience using the five capitals to compare the resilience of countries globally by utilizing large-scale, publically available, datasets. This is problematic and, for example, their indicators of financial capital focus on national-level indicators including ‘‘proportion of the population living under $1.25 a day’, ‘population which has an account at a formal financial institution’, and ‘average aid received 2008– 2014’ (IFRC, 2015). While these indicators may have meaning at the national-level, they can be relatively ineffective when the aim is to measure livelihood resilience at the household scale. While a country may have received a significant amount of foreign aid, this does not mean that the benefits have necessarily reached every household within the country or been distributed equally. Further, understanding what types of households or groups of people within a country actually have access to financial institutions and higher paying jobs would be helpful for organizations whose aim is to build livelihood resilience of the most marginalized people within a country. Understanding which groups of people have lower levels of livelihood resilience, and thus arguably have a greater need for resilience-building interventions is a major aim of the HLRA. So, how does the HLRA utilize the five livelihood capitals to measure livelihood resilience? The next section will outline the empirical methods and analysis of research conducted in two communities in Isiolo County, Kenya using the HLRA. Further, the discussion section will use these empirical results to outline how the HLRA improves and builds upon previous attempts of using the livelihood capital assets by focusing on the household-scale, using subjective measures of resilience, and highlighting how to interpret livelihood resilience measurement results.

2. Case study: using the HLRA to measure how agroforestry is building livelihood resilience in Isiolo county, Kenya 2.1. The research This case study focused on if and how agroforestry is building livelihood resilience for smallholder farmers in two semi-arid communities in Isiolo County, Kenya: Kinna and Burat. Isiolo County is located north of Mt. Kenya and is dominated by Turkana, Meru, Somali, Samburu, and Borana ethnic groups. Burat is located 5 km outside of Isiolo Town and agriculture takes place between the Isiolo and Aye Nakore Rivers, which are used for irrigation. Kinna is more rural and borders Meru National Park. Agriculture in Kinna is largely irrigated by canals built by the government as early as the 1960s (personal communication). The history of agroforestry in both Kinna and Burat is complex, but generally the spread of agroforestry has been at the grassroots level. Meru are the only traditionally agricultural ethnic group in these

257

communities and their agroforestry practices have spread to others, from neighbor to neighbor, instead of a larger-scale agricultural intervention. There is not one specific type of agroforestry practice in Kinna and Burat, but the most common include trees on borders and along roads, live fencing, home gardens, and intercropping. This study utilized a mixed-methods research approach that included 20 qualitative case study households (unstructured, qualitative interviews), 339 quantitative household surveys, and key informant interviews. All research participants were located in either Burat or Kinna, and practiced agriculture (which excluded some purely pastoralist households. The qualitative household case studies were selected through combined convenience and respondent-driven sampling (Bernard, 2011). Each household was interviewed three times; this included an initial interview, and an interview during both the rainy and dry seasons. When possible, both male and female household heads were interviewed. The qualitative interviews all took place before the survey and helped inform the livelihood resilience indicators and survey questions. The household surveys represent a statistically representative sample size of households practicing agriculture in Kinna and Burat. Surveys were conducted by enumerators who surveyed every other household along the transect for that day, interviewing either the male or female household head based on who was available and willing to participate. The average age of survey respondents was approximately 42.8 years, the average household had 7.2 members, and the average respondent planted 6.2 different types of crops, including tree species, on their farms. While agroforestry was not the main livelihood of most research participants in Kinna and Burat, it was an important supplemental livelihood activity for some. In Burat, 63.4% of agricultural households practiced agroforestry, and the main trees planted include papaya, banana, and mango. In Kinna, 61.2% of agricultural households practiced agroforestry and the main trees planted include guava, mango, and papaya. 2.2. Methods and results This section will outline the exact methodological approach used in this case study. The indicators of resilience used for this case are shown in Table 2 and the overall HLRA process is outlined in Fig. 2. 2.2.1. Step 1: Individual indicators Indicators of livelihood resilience were developed using both ‘objective’ indicators determined through literature review of livelihood capitals (Tacoli, 1999; Campbell et al., 2001; Adato and Meizen-Dick, 2002; Erenstein et al., 2010), and ‘subjective’ indicators based on the qualitative interviews and previous research which used focused groups and semi-structured surveys to explore local perspectives of adaptation to climate variability (Quandt and Kimathi, 2016). This approach of determining ‘subjective’ indicators of resilience is similar to the stakeholder assessment method proposed by Campbell et al. (2001). The qualitative interviews were key in understanding the local context and understanding the impacts of disturbances such as floods and droughts on individuals, as well as the assets that help individuals build livelihood resilience to these disturbances. Three key indicators of resilience that emerged from previous research and the qualitative interviews included road conditions, access to irrigation, and household labor availability. These indicators, presented in Table 2, are organized around the five livelihood capital assets as has been done in other research (Kusters, Achdiawan, Belcher, & Ruiz Perez, 2006; Babulo et al., 2008; Erenstein et al., 2010). While some of these indicators are ‘contextually’ based, for example ‘Normal and rainy season road conditions’ under Physical Capital, this does not mean that the ‘context’ and responses will be the same for all households

258

A. Quandt / World Development 107 (2018) 253–263

Table 2 Household survey livelihood resilience indicators. Asset Financial Capital

Quantitative Indicator (Independent Variables)       

Salaried job (yes or no) Access to a bank account (yes or no) Remittances (yes or no) Household belongings (# of belongings) Livestock (# of livestock) Size of farmland (# of acres) Ownership of farm equipment (own, rent, borrow pieces of equipment)

Human Capital

 Labor availability (number of household members between 18 – 55)  Education (level of education of respondent)  General health of family (scale of poor to good)  Health problems impact on ability to practice livelihoods (Scale of no to very much)

Social Capital

   

Physical Capital

 Normal and rainy season road conditions (scale of good to bad)  Presence of facilities (schools, hospitals, etc.) near home (yes or no)  Access to irrigation schemes (yes or no)  Ownership of farming equipment (own, rent, borrow pieces of equipment)

Natural Capital

    

Family living nearby (yes, how close) Political influence or power (scale of none to a lot) Participation in groups (# of groups) Participation in agriculture or tree planting group (yes or no)  Strength of relationship with neighbors (# of activities done with neighbors)

Size of farmland (# of acres) Own farmland (yes or no) Diversity of farm crops (# of different crops planted) Livestock (# of livestock) Soil erosion (rank of severity of soil erosion on farm)

within a community. As explained by Agrawal and Gibson (1999), researchers and development practitioners need to move away from thinking of community as ‘homogenous’, and instead recognize internal variability and differences. Thus survey respondents’ perception of ‘Normal and rainy season road conditions’ may vary significantly within a community, particularly based on the physical location of a household compared to roads. Therefore, even ‘contextual’ indicators can and do vary within a geographic community. Importantly, the indicators do not focus only on the availability of resources, but instead focus on access that individuals and households have to assets and resources. This is important because the existence of an asset does not imply that an individual is able

Overall composite asset index

to access that asset, thus addressing issues of power and human agency. These indicators presented in Table 2 were then turned into quantitative questions that were included in the household survey, which was conducted with 339 respondents. 2.2.2. Step 2: Composite asset index Next, a simple additive, or composite asset index was created for each household as outlined in Campbell et al. (2001) and Erenstein et al. (2007). This approach was also utilized for the Human Development Index (UNDP, 1994). In order to create the index, the survey results for each of the indicator questions were converted so that the answer choices for questions were on a scale of 0 to 1. The results were assigned a 1 to represent the most desirable response, and 0 to represent the least desirable response. For example, for the question of if any household member has a salaried job, any ‘yes’ answer was assigned a 1 and any ‘no’ answer was assigned a 0. Questions with multiple answer choices (such as Likert scale-type questions) were assigned values within the range of 0 to 1 (for example, 0, 0.33, 0.66, 1). Thus, it was assumed that higher scores should indicate higher levels of livelihood assets and greater livelihood resilience. The importance of converting the results of each indicator question into a scale of 0 to 1 is that it allows for them to be averaged together, and generally makes it easier to analyze the results. Each indicator was given equal weight to aid interpretation and reduce ambiguity, as done by Erenstein et al. (2007). After the survey results for each question and respondent were converted to fit a scale of 0 to 1, composite asset indexes could be created. To create the composite asset index for each of the five livelihood capital assets the individual indicator scores were averaged for each household. (Campbell et al., 2001; Erenstein et al., 2007). This means that for each household, all the results for each livelihood capital asset were averaged. For example, all the results from questions about natural capital assets were averaged to give the overall natural capital score for that household. This process is outlined in Fig. 2, which shows the individual indicators at the bottom, the scores of which are then averaged to create the composite index score for each of the five livelihood capital assets. This overall process described here was done at an individual/household-level, but scores can also be aggregated to represent larger groups of households. For example, the financial capital index scores for households both with (0.318) and without agroforestry (0.233) helps to illustrate how agroforestry increases a household’s financial capital. This type of analysis highlights that the livelihood activity of agroforestry, is building livelihood resilience by increasing financial capital. Similar analysis can be done with all five types of livelihood capital

Overall Livelihood Capital

Composite asset indexes

Natural Capital

Physical Capital

Human Capital

Social Capital

Individual Indicators

farmland farm crops livestock soil erosion

road conditions irrigation facilities services

labor availability education health

family neighbors groups political influence

Financial Capital

salary remittances belongings livestock bank account

Fig. 2. Schematic representation of data types and linkages for the sustainable livelihoods framework. Step 2 in the text is the creation of the composite asset indexes, and Step 3 is the creation of the overall composite asset index.

A. Quandt / World Development 107 (2018) 253–263

assets to understand how a specific intervention interacts with and builds certain types of capital assets. 2.2.3. Step 3: Overall composite asset index Next, for each survey respondent the scores for the five capitals were also averaged to give the overall livelihood composite asset index. This score represents the overall measurement of livelihood resilience for that survey respondent and can be used as proxies for overall livelihood resilience during analysis. For example, the average livelihood composite asset index for households with agroforestry was 0.440 and without agroforestry was 0.400. This indicates a potential 10% increase in livelihood resilience for survey households with agroforestry, which is statistically significant (pvalue of 0.000). The evidence supporting the link between livelihood capital assets and livelihood resilience was outlined in the introduction (Campbell et al., 2001; Thulstrup, 2015). Thus, results of the HLRA in our case study found that households with agroforestry have greater livelihood resilience. This suggests that promoting agroforestry may be one type of livelihood resilience building effort that could be promoted in Isiolo, Kenya (Quandt, Neufeldt, & McCabe, 2017). 2.2.4. Step 4: Analysis and visualization of results Lastly, visualizations of both the qualitative and quantitative data were created. To analyze the quantitative household survey data, the composite asset indexes and overall composite asset index were used to compare different groups of survey respondents. This process was important in determining which groups of people are more resilience and how (i.e., which capital assets are higher for certain groups of people within a community). This was often done with the use of radar diagrams, or spider charts, to compare and contrast the five capital assets (composite asset index scores) between different groups of survey respondents. For example, in Fig. 3a the five capital assets were compared between households both with and without agroforestry practices. This means that the composite asset indexes for each livelihood capital were averaged for all survey respondents that practiced agroforestry, and all survey respondents that did not practice agroforestry. This analysis and visualization shows that households practicing agroforestry had significantly higher natural (p = 0.0079) and financial capital (p = 0.0000). Further, in Fig. 3b households were aggregated into groups based on how many different tree species the household has on their farm. Fig. 3b shows even more nuance than Fig. 3a and illustrates that greater tree diversity also significantly increases natural (p = 0.0097) and financial capital (0.0000) for a household. Therefore, this analysis not only shows that agroforestry builds livelihood resilience, but which livelihood capital assets are particularly impacted by agroforestry. Thus, for households with relatively lower natural and financial capital, promoting agroforestry as a livelihood-resilience building activity may help them accumulate natural and financial capital assets. Therefore, adopting a certain livelihood practice can in fact increase a household’s access to specific assets. The qualitative interviews were transcribed and coded with help from NVivo software. These codes were then organized using the sustainable livelihoods approach as an organizational framework. These codes or ‘themes’ were used to create the flowchart in Fig. 4. Fig. 4 visualizes how the qualitative data about how a specific intervention, agroforestry, contributes to livelihood resilience by using the five livelihood capitals as an organizing framework. It helps build upon the quantitative results and shows the mechanisms by which agroforestry can build resilience to one type of disturbance: flood (Quandt et al., 2017). This type of analysis is particularly helpful to organizations aiming to understand how to effectively build livelihood resilience in an area to a specific type of disturbance. Shown in the blue bubbles are all the ‘themes’ that

259

were mentioned during the qualitative interviews that relate to the impacts of flood on livelihood capitals. For example, some interviewees discussed how floods have negative impacts on the roads, making transportation difficult or impossible. They discussed how this impacts physical capital (transportation and infrastructure), as well as financial capital because it impeded their access to markets in order to sell their crops, thus having negative impacts on household finances. Shown in the green bubbles are all the ‘themes’ that were mentioned during the qualitative interviews that relate to how agroforestry mitigates the impacts of floods. For example, agroforestry trees provided construction materials; which indirectly mitigated the impacts of floods by increasing financial capital through selling construction materials, and also directly mitigated the impacts of floods by being used directly to rebuild destroyed buildings. This example illustrates one specific mechanism, construction materials for use and sales, through which agroforestry can build livelihood resilience to floods. 2.3. Was the HLRA successful? Now that we have outlined the methods and analysis involved in the HLRA process, it is important to consider how successful it was in measuring livelihood resilience? In order to get some sense of this I compared the overall composite asset index scores from the HLRA with self-reported levels of well-being or ‘subjective’ resilience. During the survey, respondents were asked to rate their current overall living conditions and well-being compared to their neighbors. ‘Neighbors’ was used as a general term, and left up to the interpretation of the respondent. While this is not exactly the same as livelihood resilience, it is a relatively similar measure. This question was asked on a scale of 1 (much worse) to 5 (much better). Next, the livelihood capital asset overall composite scores were compared to the self-reported scores. These HLRA scores and self-reported scores were found to be statistically significantly correlated (chi2 = 57.57, p = 0.000). While neither of these scores or measures of livelihood resilience are perfect, the fact that the HLRA methods used here had similar results to self-reported measures suggests that using the sustainable livelihoods approach to measuring livelihood resilience is an effective methodological approach. 3. Discussion of the HLRA 3.1. Advantages of the HLRA The previous sections have outlined the shortcomings of previous efforts to measure livelihood resilience and explained the step by step process of measuring livelihood resilience using the HLRA by drawing from a case study of agroforestry in Isiolo County, Kenya. This section aims to bring these two discussions together by providing evidence from the case study that highlights the specific advantages of the HLRA over previous efforts to measure livelihood resilience. First, some attempts to measure livelihood resilience only provide a theoretical framework instead of practical methods, such as DFID (2011) and Speranza et al. (2014). The HLRA goes further and explains how to determine both objective and subjective indicators of resilience, how to turn those indicators into quantitative survey questions (Table 2), how to analyze (Fig. 2) and visualize the results (Figs. 3 and 4), and then how to make recommendations from the results. The case study walks the user through how these methods were used for one specific example, illustrates how agroforestry is building resilience for some households, and thus highlights one specific type of intervention (agroforestry) that could be

260

A. Quandt / World Development 107 (2018) 253–263

Fig. 3. Spider diagram comparing the five livelihood capitals between, a. households with and without agroforestry, and b. households with varying levels of tree diversity (number of different tree species). A one-way ANOVA test of variance found that natural capital (p = 0.0079) and financial capital (p = 0.0000) were significantly different between households with and without agroforestry. A second one-way ANOVA test of variance found that natural capital (p = 0.0097) and financial capital (p = 0.0000) were significantly different between the groups of households with varying levels of tree species diversity.

Fig. 4. Flowchart of qualitative data. The blue bubbles are the impacts of flood on the five livelihood capital assets and the green bubbles represent how agroforestry both directly and indirectly mitigates the impacts of floods. For example, trees directly mitigate flood by providing food to fight human hunger and disease, while trees indirectly build livelihood resilience by providing construction materials which can build a household’s physical or financial capital bases.

targeted at households not currently practicing agroforestry to help them build their overall livelihood resilience. Second, most efforts to measure livelihood resilience only rely on objective measures of resilience (IFRC, 2015; Jones and Tanner, 2015). The case study outlines how conducting focus groups and interviews prior to a quantitative survey can help highlight what indicators of resilience are most important to people themselves, an important step as outlined by Brown and Westaway (2011) and Tanner et al. (2015). In the two communities included in the case study, previous research and qualitative interviews revealed that road conditions, access to irrigation, and labor availability were perceived by residents as key indicators of resilience and thus they were included in the indicators of resilience presented in Table 2. Additionally, in the case study, the overall livelihood resilience scores were compared to respondent’s selfassessment of household well-being compared to their neighbors. This was found to be statistically significantly correlated, verifying that the indicators chosen through the HLRA process were relatively accurate measures of household livelihood resilience in the context of agricultural households in Isiolo County, Kenya.

Third, many interventions to build livelihood resilience focus on the community-scale (Berkes and Ross, 2013; Szoenyi et al., 2016), while the HLRA focuses at the household-scale. The results of Isiolo County, Kenya illustrate how the livelihood resilience levels within these communities varied by if a household practiced agroforestry, and the diversity of tree species they planted (Fig. 3). Further, both financial and natural capital were significantly different between households with and without agroforestry. This illustrates that agroforestry is building resilience particularly in the areas of financial and natural capital. These results highlight the importance of not looking at geographic communities as homogenous (Agrawal and Gibson, 1999), and that understanding the variations in livelihood resilience within a community can help to identify which households are most vulnerable and then target specific interventions at specific groups of people within a community. Fourth, Figs. 3 and 4 provide examples of how to visualize measurements of livelihood resilience. Szoenyi et al. (2016) call for a need to better analyze, visualize, and interpret results of resilience measurements, and the case study presented here illustrates how

A. Quandt / World Development 107 (2018) 253–263

the HLRA provides tools to address this issue. One of the major goals of measuring resilience in the first place is to understand what components of the livelihood system help build resilience (Walker et al., 2002), and Figs. 3 and 4 provide two different methods to visualize this information. Lastly, many efforts to measure livelihood resilience ignore human agency, the importance of power relationships, and access to assets. Tanner et al. (2015) promote the livelihood resilience approach because it highlights human agency and capacity to prepare and cope with different shocks. This is important because it shows that people can and do take an active role in building resilience. Measuring livelihood resilience through the five livelihood capital assets can highlight how people actively build and accumulate capital in order to better prepare for shocks. It also ensures the inclusion of non-monetary measures of livelihood resilience including social capital (social relationships and networks), and physical capital (access to education and transportation, for example), which is important in understanding the overall picture of livelihood resilience (Rakodi, 1999; Lebel et al., 2006). The HLRA highlights access to assets as a key to building livelihood resilience (Fig. 1), thus acknowledging to some extent how power relationships can influence access to assets. 3.2. Drawbacks and critiques While using the sustainable livelihoods approach to measure livelihood resilience has many advantages and benefits, there are also drawbacks and critiques of this method. First, it only provides a ‘snapshot’ of livelihood resilience, and not necessarily a dynamic measure of how livelihood resilience is changing. If the goal was to look at how livelihood resilience has changed over time, households would have to be asked the same questions focused on indicators of resilience at two different points in time. Additionally, the method used here does not prioritize indicators of resilience. Some indicators may be more important than others in building livelihood resilience and this method did not take that into account. However, using the same methods, different indicators could be weighted during analysis. So weighing different surrogates is possible using the methods presented here, however it was not done in the case study. A drawback of this approach is that it may be difficult to include large-scale factors into resilience measurement that can have a significant impact on livelihoods including national politics, macroeconomics, or international trade. This is a critique of the sustainable livelihoods more generally (Serrat, 2010), but also applies to the methods described here. Lastly, one critique of measuring resilience is that it is often highly contextualized, and thus difficult to include in policy (Cooper and Wheeler, 2015). This critique holds true for the HLRA because the specific indicators of resilience will vary between contexts. However, the general approach could be used regardless of context. 4. Conclusions Overall, the HLRA provides an innovative, holistic method for measuring livelihood resilience. While there are drawbacks, it has the potential to measure livelihood resilience for research, development, and humanitarian projects. Drawing from the case study, the HLRA was shown to be an effective method for developing indicators of resilience and understanding the role of agroforestry in building livelihood resilience in the two case study communities in Isiolo, Kenya. This paper proposed five strengths of the HLRA. First, the HLRA provides practical methods and tools for measuring livelihood resilience, not simply an untested theoretical framework. Second, it includes ‘subjective’ measures

261

of resilience, moving beyond top-down objective measures of livelihood resilience that ignore people’s own perceptions of their resilience. Third, it focuses on the household scale and that different individuals or households within a community have different levels of livelihood resilience. Fourth, it outlines methods for analyzing, visualizing, and interpreting the results, and identifying specific actions to build resilience, thus providing an efficient way to identify potential resilience building interventions. And lastly, the HLRA highlights the importance of human agency and power relationships in livelihood resilience and access to assets with which households build their livelihood resilience. Using the HLRA, organizations and researchers may have the ability to improve the lives and livelihoods of the world’s most vulnerable people by better understanding livelihood resilience and identifying specific livelihood resilience building activities that could be promoted with specific groups of people. Acknowledgments Most importantly I would like to acknowledge the communities of Burat and Kinna that so graciously welcomed me and to the households that spent sometimes hours answering our endless questions. Second, this research would not have been possible without the assistance of the World Agroforestry Centre in Nairobi, and the Kenya Red Cross Society – Isiolo Branch, in Isiolo. Red Cross Volunteers played an integral role in all stages of data collection, and a special thanks goes to Noor Hussein. I would also like to thank J. Terrence McCabe, Henry Neufeldt, Joel Harrter, Lisa Dilling, Myles Osborne, Kanmani Venkateshwaran, and Max Boykoff for their feedback. Funding This work was supported by a US Borlaug Fellows in Global Food Security Graduate Research Grant (grant number 206766) which supported field and research costs for Quandt. Conflict of interest The author declares no conflict of interests. Compliance of ethical standards The authors report no potential conflict of interest. Research participants provided informed consent to take part in this research. This research was approved by the Institutional Review Board at the University of Colorado under protocol # 14-0059. References Adato, M., & Meizen-Dick, R. (2002). Assessing the impact of agricultural research on poverty using the sustainable livelihoods framework, FCND discussion paper 128, and EPTD discussion paper 89. International Food Policy Research Institute. Adger, N. (2003). Social capital, collective action, and adaptation to climate change. Economic Geography, 79(4), 387–404. https://doi.org/10.1111/j.19448287.2003.tb00220.x. Agrawal, A., & Gibson, C. C. (1999). Enchantment and disenchantment: The role of community in natural resource conservation. World Development, 27(4), 629–649. Allison, E. H., & Ellis, F. (2001). The livelihoods approach and management of smallscale fisheries. Marine Policy, 25, 377–388. Babulo, B., Muys, B., Nega, F., Tollens, E., Nyssen, J., Deckers, J., et al. (2008). Household livelihood strategies and forest dependence in the highlands of Tigray, Northern Ethiopia. Agricultural Systems, 98, 147–155. https://doi.org/ 10.1016/j.agsy.2008.06.001. Bahadur, A., Lovell, E., Wilkinson, E., & Tanner, T. (2015). Resilience in the SDGs: developing an indicator for Target 1.5 that is fit for purpose. Retrieved from the Overseas Development Institute Website: https://www.odi.org/sites/odi.org. uk/files/odi-assets/publications-opinion-files/9780.pdf

262

A. Quandt / World Development 107 (2018) 253–263

Bebbington, A. (1999). Capitals and capabilities: A framework for analyzing peasant viability, rural livelihoods and poverty. World Development, 27(12), 2021–2044. S0305-705X(99)00104-7. Berkes, F., Colding, J., & Folke, C. (Eds.). (2003). Navigating social-ecological systems: building resilience for complexity and change. United Kingdom: Cambridge University Press. Berkes, F., & Ross, H. (2013). Community resilience: Toward an integrated approach. Society and Natural Resources, 26. https://doi.org/10.1080/ 08941920.2012.736605. Bernard, H. R. (2011). Research Methods in Anthropology: Qualitative and Quantitative Approaches (5th ed.). United Kingdom: AltaMira Press. Brown, K. (2014). Global environmental change 1: A social turn for resilience? Progress in Human Geography, 38(1), 107–117. https://doi.org/10.1177/ 0309132513498837. Brown, K., & Westaway, E. (2011). Agency, capacity, and resilience to environmental change: Lessons from human development, well-being, and disasters. Annual Review of Environment and Resources, 36, 321–342. https://doi.org/10.1146/ annurev-environ-052610-092905. Cabell, J. F., & Oelofse, M. (2012). An indicator framework for assessing agroecosystem resilience. Ecology and Society, 17(1), 18. https://doi.org/ 10.5751/ES-04666-170118. Campbell, B., Sayer, J. A., Frost, P., Vermeulen, S., Ruiz Perez, M., Cunningham, A., et al. (2001). Assessing the performance of natural resource systems. Conservation Ecology, 5(2), 22. Retrieved from: http://www.consecol.org/vol5/ iss2/art22. Carney, D. (1998). Implementing the sustainable rural livelihoods approach. In D. Carney (Ed.), Sustainable rural livelihoods: what contribution can we make? London, UK: DFID. Carpenter, S., Walker, B., Anderies, M., & Abel, N. (2001). From metaphor to measurement: Resilience of what to what? Ecosystems, 4, 765–781. https://doi. org/10.1007/s10021-001-0045-9. Carpenter, S. R., Westley, F., & Turner, M. G. (2005). Surrogates for resilience of social-ecological systems. Ecosystems, 8, 941–944. https://doi.org/10.1007/ s10021-005-0170-y. Chambers, R., & Conway, G. (1992). Sustainable rural livelihoods: Practical concepts for the 21st Century, IDS Discussion Paper 296. Brighton, UK: Institute of Development Studies. Chambers, R. (1994). The origins and practice of participatory rural appraisal. World Development, 22(7), 953–969. Cooper, S. J., & Wheeler, T. (2015). Adaptive governance: Livelihood innovation for climate resilience in Uganda. Geoforum, 65, 96–107. https://doi.org/10.1016/ j.geoforum.2015.07.015. Cote, M., & Nightingale, A. J. (2012). Resilience thinking meets social theory: Situation social change in socio-ecological systems (SES) research. Progress in Human Geography, 36(4), 475–489. https://doi.org/10.1177/ 0309132511425708. Cutter, S. L., Barnes, L., Berry, M., Burton, C., Evans, E., Tate, E., et al. (2008). A placebased model for understanding community resilience to natural disasters. Global Environmental Change, 18, 598–606. DFID (Department for International Development) (2011). Defining disaster resilience: a DFID approach paper. London, UK,1-20. Retrieved from: https://www.gov.uk/government/publications/defining-disaster-resilience-adfid-approach-paper. Elasha, B. O., Elhassan, N. G., Hanafi, A., & Zakieldin, S. (2005). Sustainable livelihood approach for assessing community resilience to climate change: case studies from Sudan, Working Paper No. 17. Assessments of Impacts and Adaptations of Climate Change (AIACC). Retrieved from: http://www.start.org/Projects/ AIACC_Project/working_papers/Working%20Papers/AIACC_WP_No017.pdf. Ellis, F. (2000). Rural livelihoods and diversity in developing countries. Oxford, UK: Oxford University Press. Enns, C., & Bersaglio, C. (2015). Exploring the implications of oil and gas development for livelihood resilience in Turkana, Kenya. For the PRISE small grant programme of the Overseas Development Institute. Retrieved from: http://www.academia.edu/ 13047117/ Exploring_the_implications_of_oil_and_gas_development_for_ livelihood_resilience_in_Turkana_Kenya. Erenstein, O., Hellin, J., & Chandna, P. (2007). Livelihoods, poverty and targeting in the Indo-Gangetic Plains: A spatial mapping approach Research Report. New Delhi, India: CIMMYT/RWX. Retrieved from: http://citeseerx.ist.psu.edu/ viewdoc/download?doi=10.1.1.597.5921&rep=rep1$type=pdf. Erenstein, O., Hellin, J., & Chandna, P. (2010). Poverty mapping based on livelihood assets: Ameso-level application in the Indo-Gangetic Plains, India. Applied Geography, 30, 112–125. https://doi.org/10.1016/j.apgeog_2009.05.001. Gunderson, L., & Holling, C. S. (2002). Panarchy: Understanding transformation in human and natural systems. Washington DC: Island Press. Hodbod, J., & Eakin, H. (2015). Adapting a social-ecological resilience framework for food systems. Journal of Environmental Studies and Science, 5, 474–484. https:// doi.org/10.1007/s13412-015-0280-6. Hussein, K., & Nelson, J. (1998). Sustainable livelihoods and livelihood diversification, IDS Working Paper 69. UK: Institute of Development Studies, University of Sussex. International Federation of Red Cross and Red Crescent Societies (IFRC) (2015). The One Billion Coalition for Resilience: Discussion Paper on Levels of Resilience. Dalberg Analysis. Jacobs, B., Nelson, R., Kuruppu, N., & Leith, P. (2015). An adaptive capacity guide book: Assessing, building and evaluating the capacity of communities to adapt in a changing climate.. Hobart, Tasmania: Southern Slopes Climate Change

Adaptation Research Partnership (SCARP), University of Technology Sydney and University of Tasmania. Johansson, K. (2015). Barriers and bridges for introducing agroforestry and community-based forestry among food insecure households in eastern Africa (Doctoral thesis). Skinnskatteberg: Swedish University of Agricultural Sciences. Jones, L., & Tanner, T. (2015). Measuring subjective resilience: using people’s perceptions to quantify household resilience, Overseas Development Institute Working Paper 423. doi: 10.13140/RG.2.1.2438.1289. Krantz, L. (2001). The sustainable livelihood approach to poverty reduction: An introduction. Sweden: Swedish International Development Cooperation Agency. Retrieved from: http://www.sida.se/contentassets/bd474c210163447c9a7963d 77c64148a/the-sustainable-livelihood-approach-to-poverty-reduction_2656.pdf. Kusters, D., Achdiawan, R., Belcher, B., & Ruiz Perez, M. (2006). Balancing development and conservation? An assessment of livelihood and environmental outcomes of nontimber forest product trade in Asia, Africa, and Latin America. Ecology and Society, 11(2), 20. Lebel, L., Anderies, J. M., Campbell, B., Folke, C., Hatfield-Dodds, S., Hughes, T. P., & James W. (2006). Governance and the capacity to manage resilience in regional social-ecological systems, Marine Sciences Faculty Scholarship, Paper 52. Retrieved from: http://digitalcommons.library.umain.edu/sms_facpub/52. Leslie, P., & McCabe, J. T. (2013). Response diversity and resilience in socialecological systems. Current Anthropology, 54(2), 114–143. https://doi.org/ 10.1086/669563. Lindstädter, A., Kuhn, A., Naumann, C., Rasch, S., Sandhage-Hofmann, A., Amelung, W., et al. (2016). Assessing the resilience of areal-world social-ecological system: Lessons from a multidisciplinary evaluation of a South African pastoral system. Ecology and Society, 21(3), 35. https://doi.org/10.5751/ES-08737210335. Marschke, M. J., & Berkes, F. (2006). Exploring strategies that build livelihood resilience: A case from Cambodia Retrieved from:. Ecology and Society, 11(1), 42 Http://www.ecologyandsociety.org/vol11/iss1/art42/. Mosberg, M., & Eriksen, S. H. (2015). Responding to climate variability and change in dryland Kenya: The role of illicit coping strategies in the politics of adaptation. Global Environmental Change, 35, 545–557. https://doi.org/10.1016/ j.gloenvcha.2015.09.006. Nadasdy, P. (2007). Adaptive co-management and the gospel of resilience. In D. Armitage, F. Berkes, & D. Doubleday (Eds.), Adaptive co-management: collaboration, learning, and multilevel governance. Vancouver, Canada: University of British Columbia Press. Nelson, V., & Stathers, T. (2009). Resilience, power, culture, and climate: A case study from semi-arid Tanzania, and new research directions. Gender and Development, 17(1), 81–94. https://doi.org/10.1080/135520708026696946. Obrist, B., Pfeiffer, C., & Henley, R. (2010). Multi-layered social resilience: A new approach in mitigation research. Progress in Development Studies, 10(4), 283–293. Otsuki, K., Jasaw, G., & Lolig, V. (2017). Linking individual and collective agency for enhancing community resilience in Northern Ghana. Society and Natural Resources, 1–15. https://doi.org/10.1080/08941920.2017.1347971. Quandt, A., & Kimathi, A. (2016). Adapting livelihoods to climate change in arid Kenya: Local perspectives and insights. African Journal of Rural Development, 1 (1), 51–60. Quandt, A., & Kimathi, A. (2017). Perceptions of the effects of floods and droughts on livelihoods: Lessons from arid Kenya. International Journal of Climate Change Strategies and Mitigation, 9(3). https://doi.org/10.1108/IJCCSM-112014-0132. Quandt, A., Neufeldt, H., & McCabe, J. T. (2017). The role of agroforestry in building livelihood resilience to floods and drought in semiarid Kenya. Ecology and Society, 22(3), 10. Rakodi, C. (1999). A capital assets framework for analysing household livelihood strategies: Implications for policy. Development Policy Review, 17, 315–342. https://doi.org/10.1111/1467-7679.00090. Scoones, I. (1998). Sustainable Rural Livelihoods: A Framework for Analysis, IDS Working Paper 72. Retrieved from: https://www.staff.ncl.ac.uk/david.harvey/ AEF806/Sconnes1998.pdf Scoones, I. (2009). Livelihoods perspectives and rural development. Journal of Peasant Studies, 36(1), 171–196. https://doi.org/10.1080/03066150902820503. Serrat, O. (2010). The sustainable livelihoods approach. Washington, DC: Asian Development Bank. Speranza, C. I., Wiesmann, U., & Rist, S. (2014). An indicator framework for assessing livelihood resilience in the context of social-ecological dynamics. Global Environmental Change, 28, 109–119. Szoenyi, M., Nash, D., Burer, M., Keating, A., McQuistan, C., & Campbell, K. (2016). Risk Nexus: Measuring flood resilience – our approach. Zurich, Switzerland: Zurich Insurance Group. Tacoli, C. (1999). Understanding the opportunities and constraints for low-income groups in the peri-urban interface: the contribution of livelihood frameworks. London, UK: Development Planning Unit: Strategic Environmental Planning and Management for the Peri-Urban Interface Research Project. Retrieved from: http://www.ucl.ac.uk/dpu-projects/drivers_urb_change/urb_economy/ pdf_Urban_Rural/DPU_PUI_Takodi_opportunities.pdf. Tanner, T., Lewis, D., Wrathall, D., Bronen, R., Cradock-Henry, N., Huq, S., et al. (2015). Livelihood resilience in the face of climate change. Nature Climate Change, 1, 23–26. https://doi.org/10.1038/NCLIMATE2431. Thulstrup, A. W. (2015). Livelihood resilience and adaptive capacity: Tracing changes in household access to capital and central Vietnam. World Development, 74, 352–362. https://doi.org/10.1016/j.worlddev.2015.05.019.

A. Quandt / World Development 107 (2018) 253–263 Ulrich, A., Speranza, C., Roden, R., Kiteme, B., Wiesmann, U., & Nüsser, M. (2012). Smallholder livelihood dynamics between 1997 and 2010: Evidence from Laikipia, Kenya. Journal of Rural Studies, 28, 241–251. United Nations Development Program (UNDP) (1994). Human Development Report 1994 Retrieved from:. New York, USA: United Nationals Development Programme. http://hdr.undp.org/en/content/human-development-report-1994. Walker, B., & Salt, D. (2006). Resilience thinking: Sustaining ecosystems and people in a changing world. USA: Island Press.

263

Walker, B., Carpenter, S., Anderies, J., Abel, N., Cumming, G., Janssen, M., et al. (2002). Resilience management in social-ecological systems: A working hypothesis for a participatory approach Retrieved from: http://www.consecol. org/vol16/iss1/art14/Retrieved from: http://www.consecol.org/vol16/iss1/ art14/. Conservation Ecology, 6(1), 14. Walsh-Dilley, M., Wolford, W., & McCarthy, J. (2016). Rights for resilience: Food sovereignty, power, and resilience in development practice. Ecology and Society, 21(1), 11. https://doi.org/10.5751/ES-07981-21011.