TABLE OF CONTENTS
STATEMENT OF THE AUTHOR
ACRONYMS AND ABBREVIATIONS
TABLE OF CONTENTS
LIST OF TABLES
1.1. Background of the Study
1.2. Statement of the Problem
1.3. Research Questions
1.4. Objectives of the Study
1.5. Significance of the Study
1.6. Scope and Limitation of the Study
2. LITERATURE REVIEW
2.1. Concepts and Definitions of Terminologies
2.2. Theoretical Framework
2.2.1. Social risk: sociological perspectives
2.2.2. Theoretical perspectives of social risk management
2.2.3. Models of poverty traps
2.2.4. Smallholders’ Commercialization
2.2.5. Agricultural household model
2.2.6. Transaction Cost Theory
2.2.7. Theory of Change
2.3. Analytical Framework
2.3.1. Institutional analysis paradigm
2.3.2. Impact assessment method
2.4. An Empirical Literature Review
2.4.1. Institutional Challenges facing Smallholders’ Commercialization
2.4.2. Risk perception and institutional response to social risk management
2.4.3. Choice of Institutional Arrangement for Social Risks Management
2.4.4. Impacts of social risk management Arrangement on Smallholder Commercialization
2.5. Conceptual Framework
3. RESEARCH METHODOLOGY
3.1. Description of the Study Area
3.2. Sample Design and Techniques
3.3. Data Type, Source, and Collection Methods
3.4. Methods of Data Analysis
3.4.1. Descriptive statistics
3.4.2. Econometric models
126.96.36.199. Multivariate probit regression
188.8.131.52. Propensity Score Matching (PSM)
4. RESULTS AND DISCUSSION
4.1. Description of demographic, socio-economic, and institutional characteristics
4.1.1. Demographic and social-economic characteristics of household
4.1.2. Social networks /organization, Social cohesion, and trust
184.108.40.206. Helpfulness of participation in different networks /organization
220.127.116.11. Social cohesion
4.2. Institutional Challenges in Smallholder s’ Commercialization
4.3. Types and Major Institutional Causes of Social Risks
4.3.1. Types of social risks
4.3.2. Institutional traps: as major causes of social risks
18.104.22.168. Governance traps of social risk
22.214.171.124. Institutional disincentives
126.96.36.199. Elites and security forces exploitation
188.8.131.52. Inefficient of social protection arrangements
4.4. Determinants of Institutional Arrangements for Social Risk Management
4.4.1. Types of institutional arrangements for social risk management
4.4.2. Determinants of Institutional Arrangement social risk management
4.5. Impact of Institutional Arrangements on Smallholder Commercialization
4.5.1. Estimation of the propensity scores
4.5.2. Average Treatment Effect on the Treated (ATT)
5. CONCLUSIONS AND RECOMMENDATIONS
5.2. Conclusion and Recommendations
7.1. Appendix Tables
7.2. Appendix Figure
Appendix Figure A: Extraction Method: PCA for Institutional Challenges
Appendix Figure B: Extraction Method: PCA for Social Risks
Appendix Figure D: Propensity Score and Balancing Property of Propensity
7.3. Questionnaire for Sample Households
The author was born in West Arsi zone of Oromia National Regional State, Kore district, Kore town, on February 20, 1984, from his mother Tiya Leta, and father Kiniso Begujie. He attended his primary and junior secondary school education at Kore Junior Secondary School, high school education at Kofele High school, and preparatory education at Shashemene preparatory school. After completing his preparatory school education in 2003 he joined the then Haramaya University in September 2004 and graduated with B.Sc. Degree in Agricultural Extension in 2006. Immediately after graduation, he was employed by Chiro Agricultural Technical Vocational Education and Training (ATVET) College, Oromia Bureau of Agriculture and Rural Development, as a junior instructor. Chiro ATVET upgraded to Haramaya University, Chiro College of Agro-industry, and Land Resource. After serving for four years, the author has got an opportunity to pursue his further education in 2011 at Haramaya University and obtained a M.Sc. degree in Rural Development and Agricultural Extension in 2013. After the completion of his M.Sc. degree study, he served as a lecturer for about two years at Haramaya University, Chiro College of Agro-industry, and Land Resource. Following the promotion of Chiro College of Agro-industry and Land Resource to Oda Bultum University in 2015, the author was employed and served Oda Bultum university as a lecturer until he joined the Directorate for Postgraduate Program of Haramaya University to pursue his Ph.D. degree in Commercialization and Rural Institutions in the Department of Rural Development and Agricultural Extension in 2017.
First, I am highly grateful to Allah that bestowed me courage and patience throughout the study period.
I would like to express my sincere appreciation and gratitude to my advisors Professor Fekadu Beyene, Professor Jema Haji, and Dr. Jemal Yusuf for their vital contribution to this dissertation through giving me inspiring guidance, encouragement, and comments. In my journey towards this degree, I have found inspiration and critical support from my advisors, they have given me their heartfelt guidance, and invaluable suggestions starting from the development proposal to the final write up of this dissertation. Indeed, it was a privilege and wonderful knowledge to acquire the opportunity to work with them.
I am thankful to Oda Bultum University for giving me the scholarship through a fund procured from the Minister of science and higher education. I acknowledge the help provided by the Mieso and Gumbi-Bordode Office of Agricultural and Natural Resource Office and my sampled farmers who patiently responded to the lengthy interview schedule. My thankfulness goes to the efforts of my data collectors and fieldwork facilitators because this dissertation achievement would not have been possible without their unreserved efforts. I am most grateful to staff members of West Hararghe Zonal Agriculture and Natural Resource office for their cooperation in getting the secondary data.
May attainment at this point was by the important role played by many people; I would also like to express my profound gratitude to my mother Tiya Lata and my brother Safi Kiniso for their all rounds support. I especially want to thank my beloved wife Zemzem Ibrahim for her moral support, encouragement, and kindness and my sons’ Kena Aman and Hilan Aman for their love and jocks when I am in hard condition. Also, my special thanks go to my friends Chalchisa Fana, Gemmachu Husen, Gashew Jifara, and Husein Mohammed, Alemayehu Beyene, and Abdurahman Usman for their moral, financial and material support during my research work.
My special thanks go to my family members. In particular, I would like to thank my brothers Mustafa Kiniso, Bonsemo Kiniso, Abdi Mohamed, Bekam Mohammed, Tokkumma Mohammed, and Rahima (Yahya) Abdela; my sisters Misra Kiniso, Dureti Mohamed, and Sifan Safi for all their support, encouragement, and prayers for my success. I would also like to thank Ibrahim Umer, Hawa Ahmed, Shukur Ibrahim, and Amriya Ibro and other family members unmentioned here, for giving me support and encouragement during this study.
Key Informant Interview
LIST OF TABLES
Table 1: Sample Kebele administrations and the number of households
Table 2: Description of independent variables used in the multivariate probit model
Table 3: Description of variables and Measurement unit for impacts analysis
Table 4: Demographic characteristics of the sample households (dummy variables)
Table 5: Summary of descriptive statistics for continues variables
Table 6: Helpfulness of participation in different networks /organization
Table 7: Independent sample test for social cohesion
Table 8: Independent sample test for social cohesion
Table 9: Mean, standard deviation and varimax rotated for institutional challenges
Table 10: Mean, deviation and results of varimax rotated factor analysis for risks
Table 11: Indicators and risks induced by governance and administrative injustices
Table 12: Mean, standard deviation and varimax factor of Institutional Arrangement
Table 13: Multivariate regression for Determinants of Institutional Arrangement
Table 14: Logit estimates of the propensity scores
Table 15: Propensity score and covariate balance
Table 16: Chi-square test for the joint significance of variables
Table 17: ATT of institutional arrangement on outcome variables
LIST OF FIGURES
Figure 1: Conceptual framework
Figure 2: Map of study area
Figure 3: Kernel density of propensity scores
Figure 4: Common Support for Propensity Score Estimation
INSTITUTIONAL ARRANGEMENT FOR SOCIAL RISK MANAGEMENT AND ITS IMPACT ON SMALLHOLDERS’ COMMERCIALIZATION: THE CASE OF MIESO AND GUMBI- BORDODE DISTRICTS OF WEST HARARGHE ZONE, OROMIA NATIONAL REGIONAL STATE, ETHIOPIA
Social risks are a global phenomenon, having severe and direct implications for society, incurring large-scale influence, depending on the cultural, political, and economic context in which this phenomenon manifests itself. In recent years, a more holistic approach and society-wide arrangements of social risk management that go beyond social protection have become increasingly important to assist individuals and households to better manage multiple social risk factors. Given this, in the West Hararghe zone, pastoralists and agropastoralists’ often deal with social risks through the market, informal and public arrangements in the case of risks due to different reasons. However, the impact of the institutional arrangements for social risk management has not been well studied and documented. Hence, the major interest of this study was to identify institutional challenges in smallholders’ commercialization, to assess perceived sources of social risks, to analyze factors affecting the choice of institutional arrangements for social risk management; and to evaluate the impact of institutional arrangements for social risk management on smallholder’s commercialization the case of Mieso and Gumbi-Bordode districts of West Hararghe zone of Oromia Regional State. To meet these objectives a total of 382 sampled households were randomly selected from ten kebeles. Descriptive and inferential statistics of principal component analysis, multivariate regression, and propensity score matching (PSM) were employed to achieve these objectives. The result from principal component analysis reveals that service delivery, tenure security, price changes, insufficient market facilities, information, and channels was identified as institutional challenges faced smallholder’s commercialization having high factor loadings. The result of the principal component analysis indicates social risks faced by households include loss of life and property, politico-social instability, displacement, and violent conflict. The result of multivariate regression inferred that family size, crop income, livestock income, and extension service, organizational experience, social cohesion, and trust significantly affected households choice of institutional arrangements. Estimates of the propensity score matching shows that households’ decision to participate in the output market decreased by 14%; while an increased in households’ income in crop and livestock production value of birr 214.80 and 2952.61 respectively among participants than nonparticipants of institutional arrangements for social risk management. The findings imply that institutional arrangements are crucial interventions in managing social risks and enhancing farm income. Thus, this study suggests a well-designed institution and institutional arraignments for ensuring tenure security and removal of the underlying sources of social risks and challenges thereby help the creation of consistent livestock commercialization and sustained social protection in the area.
Keywords: Institutional Challenge, Social Risk, Institutional Arrangements for Social Risk Management, smallholders’ Commercialization, Mieso, and Gumbi-Bordode districts.
1.1. Background of the Study
People everywhere face risks, but poor people are typically more exposed to risks and least protected from them (Hoogeveen et al., 2005), because they have limited assets and less able to deal with risks and absorb shocks (Ludi, 2007). Risk exposure has a direct bearing on wellbeing and causes poverty. Risks and shocks are arguably related to all dimensions of poverty and the direction of causation can be both ways. Put differently, poverty causes exposure to risks and risks can cause poverty (Hoogeveen et al., 2005). Globally, in 2015 about 736 million people were living in extreme poverty. Nearly 40% of the world’s extreme poor live in Low-income countries (FAO, 2019). Most of them are found in sub-Saharan Africa (World Bank, 2018). Statistics show high levels of poverty in (agro-) pastoral areas of sub-Saharan Africa (SSA) including Ethiopia (SNV, 2012). In SSA, almost 300 million people rely mainly on pastoralism for their livelihoods (IFAD, 2020). To reduce this, many national governments in SSA have prioritized livestock-based commercialization and smallholders’ market participation in PAPs area.
Transforming the livestock sector from its current subsistence to a more commercialized system is an important pathway to ensure household food security and poverty reduction. In other words, livestock-based commercialization and smallholders’ market participation were the pathways to transform the livestock industry. According to FAO smallholder refer to farmers with limited resources, i.e., resource-poor in terms of farm and pasture or grazing land, labor, and capital that describe pastoral and agro-pastoral systems of the study area (Dixon et al., 2003). Commercialized forms of pastoralism have evolved over decades and pastoralists are major suppliers of livestock to domestic and export markets. This trend is likely to continue as demand for livestock products is predicted to more than double (Delgado et.al ., 1999 as cited in Henriksen, 2014). Urbanization, income growth, and globalization were also seen as common drivers for change in the agri-food system and dietary transformation (Kilelu et al., 2016). Smallholders, on contrary, continued to face challenges in grasping these chances. As a result, the majority of smallholder producers remain unable to make the transition from subsistence to commercialized production.
Ethiopia is one of the major exporters of livestock, with most of these animals are sourced from pastoralist and agro-pastoralist (PAPs) areas (Catley, 2017). However, PAPs in Ethiopia faced challenges related to lack of suitable infrastructure, lack of access to local and regional markets, and lack of access to credit and cooperatives (Kenton, 2018). In support of this, studies mentioned high transaction costs, low rate of technology adoption, limited extension services, transportation, and finance challenged output market participation of smallholders’ (Kilelu et al., 2016). Whilst these challenges are worsened by lack of political voice, market power, and political marginalization of pastoralists’ from long-standing governance failures, weak institutions, lack of will, and incentives from policy-makers to include pastorals’ interests in national policy debates (Pavanello, 2009). The collective consequences of these complex institutional challenges have led to the persistence of subsistence farming and increased risks induced by competition over resources and violent conflicts that further enhances food insecurity and traps smallholders in chronic poverty.
Increased pastorals and agro pastorals destitution and disparities in wealth driven by multiple conflicts and frequent droughts (Catley, 2017) as well as population growth make the situations extremely difficult for poor households to ‘move up’ to medium wealth status (Catley, 2010). To begin with, as a result of conflict over the border that aggravated by drought, with increased competition for scarce resources, and inter-communal violence about three million people displaced in Ethiopia (GRID, 2019). This has significantly contributed to the extensive social risks faced by 2.9 million people in Ethiopia in 2018, thereby requiring social risk management arrangements. This implies that prolonged underperformance, the failures of institutions, political and resource-based conflict, and overstretched government capacity created a high-risk environment which further causes social risk in the country. This diverse evidence helped to draw the conceptual framework of this study.
Recently, concern about social risk has become apparent in Eastern Ethiopia and one of the greatest human tragedies. For example, as indicators of social risk, conflict and mass displacement has steadily raised in PAPs areas. Consequently, Ethiopia took the largest percentage of internally displaced peoples in the world displacement figures (YIGZAW, 2019). West Hararghe was once again the region most affected by conflict and displacement since 2016. These were due to widespread poverty, longstanding economic stagnation, and competition over diminishing resources (Vicente, 2020). The situation is particularly evident in Mieso and Gumbi-Bordode districts, where the most severe humanitarian crisis and the human tragedy currently function with increasing vulnerability, driven by mounting uncertainties and risks.
Since 2015, in the study area, social risk is progressively more complicated matter and it was compounded by political unrest, social disruption, and instability. Social risk in the present study used to refer risks that negatively impact the individuals/households and societies in the form of conflict, governance, livelihood, and social instability and insecurity, driven by institutional trap. The social risk landscape is changing fast due to ethnic-based violence and a breakdown in social cohesion (OCHA, 2019). These have significantly challenged the humanitarian landscape and aid organizations in responses to mass displacement, protection concerns, and access limitations. Yet, pastoralists and agro-pastoralists in the study area were facing many social risks that threaten their lives and livelihoods. Studies carried out in Eastern Ethiopia including Mieso and Gumbi-Bordode districts, focused mainly on risks related to food insecurity (Gemechu et al., 2015), livestock production, and marketing (Kedija, 2008), climate change, and conflict (Stark, 2011).
Hence, to address these, a more holistic approach and society-wide arrangements of social risk management that go beyond social protection have evolved and become increasingly important to assist individuals and households to better manage multiple social risk factors. Institutional arrangements for social risk management in this study used to refer an informal, market, and public arrangements that help individuals and households to better manage social risks. Given this, in the West Hararghe zone, pastoralists and agro-pastoralists’ often deal with social risks through the market, informal and public arrangements in the case of risks due to different reasons. However, the impact of the institutional arrangements for social risk management on smallholders’ commercialization has not been well studied and documented. Assessing the impact of institutional arrangements for social risk management on smallholders would be significant to get lessons for the study area. This study was, therefore, carried out in the West Hararghe zone, to assess institutional arrangement for social risk management and its impact on smallholder’s commercialization.
1.2. Statement of the Problem
A major challenge to rural and agricultural development is smallholder households remain vulnerable to shocks and risks (Tirivayi, 2016). Pastoralists and agro-pastoralists in particular vulnerable to extensive risks and shocks that are worsened by a lack of comprehensive land policy, investments, postharvest processing and storage equipment, poor marketing systems, social, and environmental constraints (Dioula et al., 2013). Another argument indicates most rural households in low-income countries are net buyers of food, which makes them susceptible to production and market-related risks (De Janvry, 2011). Moreover, natural disasters, livestock diseases, climate change, financial crises, global food price hikes, conflict, economic collapse, and devastating epidemics are major threats to the welfare of rural households (Dorward et al., 2006; Dercon 2005). Many social problems can be influenced by social risk management interventions and others.
Scholarly research on Entrapment poverty, risks, and shocks has increased in recent years, explaining certain dynamics that leave poor people less able to deal with poverty, risks and absorb shocks. For example, the poverty trap is understood as “a self-reinforcing mechanism which causes poverty to persist” and whereby “poverty begets poverty” (Azariadus, 2005). In a similar vein, the theory of development economics has been built around concepts such as the “circular and cumulative causation” (Myrdal, 1957), which stresses of poverty traps. A literature has also pointed to a “middle-income trap” that affects countries’ ability to sustain long-lasting growth (Gill, 2007; Kharas, 2011; Melguizo et al., 2017). Similarly, some scholars have explained social risk from concepts relating to social problems, social crises, conflicts, and violence.
Liu et al. (2016) conceptualized social risk as risks that may influence the whole society and lead to social turbulence and social unrest in the forms of social tension and collective conflicts, which can spawn public confrontations, public protest, and even violent conflicts. Social conflict is the struggle of people in society using non-institutionalized or illegal modes of action to pursue their interests. Yuan et al. (2018) also described social risks as having the possibility of leading to social conflicts, which threaten social stability and social order and causing further social crisis. Social problems can be transformed into social risks if disputes related to social issues cannot be dealt with carefully and properly (Shi et al., 2015). When grievances and needs are not addressed, social risk increases, and seemingly “insignificant” events can trigger violence. Social risks are risks to society, social groups, or individuals. Social risk is found within a population’s underlying tensions and its struggle to acquire basic needs.
Recently, concern about social risk in Sub-Saharan Africa is one of the human adversities. Sub- Saharan Africa and South Asia account for more than half (1.44 billion) of the world’s vulnerable employment, joblessness persists or rapidly rising youth unemployment has raised social unrest (ILO, 2015). For example, social unrest, violent conflict, mass displacement, political instability, and insecurity have gradually increased in the SSA in 2019. The situation is particularly evident in Ethiopia, where West Hararghe was one of the zone affected by social risks such as conflict, migration, death, injury, and displacement since 2016. In the west Hararghe zone, pastoralists and agro-pastoralists were extensively exposed to social risks of food insecurity and production loss (Gemechu et al., 2015), illegal and forced displacement and migration CISP (2017); price rises, illness, job loss, underemployment, deaths, and injustices. Territorial expansion, competition over scarce resources, livestock raids and counter raids triggered by prolonged drought, ethnic rivalries, and the revenge tradition are also the contributors of social risk. Resource scarcity and competition in the area arise from the natural resource base, population pressures, environmental degradation, and the resulting climate change (Zigale, 2016).
Given this backdrop, the causal instruments of social risks were seemingly related to the nation of institutional trap. An institutional trap is understood as an inefficient but stable institution (norm) according to the work of V.M.Polterovich’s (1999) as cited in by (Balatsky, 2013). Institutional traps are particularly relevant in a rapidly changing pastoral and agro pastoral landscape, which poses new and increasingly complex social risks. To reverse the multidimensional poverty, social risks and shocks, intervention such as structural transformation, social protection and disaster risk management has been suggested in the rural development policies of Ethiopia.
Structural transformation with a process of agricultural and rural transformation help poverty tends to decrease involving the shift from primarily subsistence farming to market-oriented and diversified production systems, and the emergence of the rural non-farm sector (FAO, 2017). Extreme poverty reduction through smallholders’ commercialization that focused on market and value chain development is a crucial pathway towards economic growth and development von Braun, 1994; Pingali, 1995; Timmer, 1997; Romer, 1994; and World Bank, 2008 works as cited in (Berhanu, 2010). Moreover, Sen (2014) states a precondition for the reduction of rural extreme poverty is broad-based economic growth which is inclusive of the poor and extremely poor. Such growth depends on a minimum set of investments and policies, thus fostering an enabling environment for economic and social development. This suggested that managing an extreme risk among poor people is difficult at the household and community level.
In social protection landscape of Ethiopia, cash- and food-for-work Productive Safety Net Programme (PSNP), which constitutes a key part of the government’s Food Security Programme (FSP). Nevertheless, the PSNP is not the only policy arrangement that aims to provide a form of social protection in rural Ethiopia (Lavers, 2013). The challenges in operationalization of national social protection policy in Ethiopia compounded with inadequate coordination arising from institutional capacity imbalance. This, in particularly, hampered by a lack of limited vertical interface, frequent changes of institutional arrangements, and imbalance between government agencies (Hailu (2013). Moreover, a lack of common understanding on “drought resilience”, poor governance, inadequate institutions and decentralization, and weak capacity were also contributed to slow progress of droughtrisk management in Ethiopia (Mesay et al., 2017).
Most studies carried out in eastern Ethiopia, focused on the PSNP of public arrangements of social risk management. Few studies have examined the impact of social risk management arrangements on smallholders’ commercialization. Emerging studies focused on poverty reduction impacts of social protection and agricultural intervention. Tirivayi et al., (2016) observed that the empirical literature on the agricultural outcomes of social protection is neither extensive nor well established. This study aimed to assess the impact of institutional arrangements for social risk management on smallholder’s commercialization.
1.3. Research Questions
Corresponding to the above problem of the statement, this study has attempted to provide answers to the following research questions:
1. What are the institutional challenges facing smallholders’ commercialization efforts?
2. What are the perceived sources of social risk among smallholder pastoralists’ and agro pastoralists’?
3. What are the main determinants of the choice of institutional arrangements for social risk management among smallholders in the case of risks due to different reasons?
4. Do different institutional arrangements for social risk management impact smallholders’ commercialization?
1.4. Objectives of the Study
The general objective of this study is to analyze the impacts of institutional arrangements for social risk management on smallholder’s commercialization in the study area.
The specific objectives of this study are:
1. To identify institutional challenges in smallholder’s commercialization;
2. To assess the perceived types and sources of social risks among smallholder pastoralists’ and agro pastoralists’;
3. To analyze the determinants of smallholders’ choice of institutional arrangements for social risk management; and
4. To evaluate the impact of institutional arrangements for social risk management on smallholder commercialization.
1.5. Significance of the Study
It is widely recognized that strengthening smallholder farmer’s activity is a key driver to reduce poverty, ensure food security, and enhance economic development. The viability of investing in smallholders is constrained by institutional obstacles, like a pervasive market failure in most developing countries, which include lack of access to information and high marketing and transaction costs, and problems of market coordination. Thus, smallholders have challenged both in accessing and viably staying in competitive markets and this situation will pose a potential danger on the viability of developing countries’ development strategies. As a result, the concern of supporting smallholders’ access to markets began to fascinate the attention of policymakers and development practitioners. Hence, this study is thus fascinated to act in response to why smallholder farmers remain incapable to sustain their agricultural investment to play part in a lucrative market. Analyzing the institutional arrangements for social risk management and its impacts on smallholder’s commercialization will provide beneficial information to the managers and other stakeholders to know the effectiveness of the smallholder’s operation in the marketing of produces.
1.6. Scope and Limitation of the Study
This study emphasized on social risk management and smallholders’ commercialization. Social risk management entails strategies and arrangements. Prevention, mitigation, and coping strategies are organized into informal, market-based, and public arrangements. The aspects of this study look into institutional arrangements for social risk management, and the impact of various institutional arrangements on smallholder’s commercialization. Besides, the study tried to identify the institutional challenges and root causes and consequences of social risk. Focused upon the above conceptual areas, this study helps full to understand deeply about the problem. The study is bounded to the smallholders of West Hararghe Zone in particular reference to Mieso and Gumbi-Bordode districts. This area is among those areas where the major risk of disaster, a fragile and degraded natural environment, limited communication infrastructure, inefficient markets, a border conflict between Oromo and Somali peoples, and food insecurity is widely observed (CBDRM, 2016).
In the process of dealing with this research, some challenges hindered data collection and analysis. The first challenge was a financial constraint and lack of secondary data on social risk management for in-depth analysis. This study has also faced a lack of relevant documents and information on social risk in the study area.
2. LITERATURE REVIEW
2.1. Concepts and Definitions of Terminologies
Risk: It is defined inclusively, in so far as it encompasses social, economic, political, and environmental risks, including labor market-related risks, such as unemployment and nonemployment according to the reformulated framework of social risk management (SRM).
Social risk: It represents the probability for a person to be affected by an unexpected, uncertain situation (unemployment, debts, family disintegration) associated with loss of control over one’s personal actions (Sirovatka, 2010)
Social risk management: The definition of social risk management points to the need for a society-wide perspective that examines the diverse set of interventions available to help poor and near-poor households manage risks so that they can change behaviors and exit poverty and/or lower vulnerability to poverty (Grosha, 2008).
Social protection: It is defined as the set of public and private policies and program aimed at preventing, reducing and eliminating economic and social vulnerabilities to poverty and deprivation” (UNICEF, 2013). Social protection can be defined as “a system of formal and informal interventions that aim to reduce social and economic risks, vulnerabilities and deprivations for all people and facilitate inclusive social development and equitable economic growth” (UNICEF, 2012).
Institutions: Institutions are defined as humanly devised constraints that structure political, economic, and social interactions (North, 1991).
Formal institutions: Formal institutions are defined as political constraints on government behavior enforced by legal institutions. Formal rules encompass constitutional constraints, statutory rules, and other political constraints (Williamson, C. 2009).
Informal institutions: Informal institutions are private constraints stemming from norms, culture, and customs that emerge spontaneously (Williamson, C. 2009).
Institutional arrangements: It is understood as the Institutional arrangements are the combination of formal constraints, informal rules, and their enforcement characteristics (North 2005).
Institutional arrangements for social risk management: It is defined as institutional arrangements for dealing with risk through informal (family or community-based), market based and publicly-provided mechanisms (Holzmann and Kozel, 2007). It is also explained as an informal, market, and public based arrangements that help individuals and households to help manage social risks.
Informal arrangements for social risk management: These include marriage, mutual community support, and savings in real assets such as cattle, real estate, and gold.
Market-based arrangements for social risk management: These include financial assets - cash, bank deposits, bonds, and shares - and insurance contracts.
Publicly mandated or provided arrangements: These include social insurance, transfers, and public works.
Institutional trap: It is understood as an inefficient but stable institution (norm). V.M.Polterovich’s (1999) as cited in by (Balatsky, 2013).
Smallholder: It is used to refer producers with limited resources, i.e., resource-poor in terms of farm and pasture or grazing land, labor, and capital that describe pastoral and agro-pastoral systems of the study area (Dixon J., 2003).
Commercialization: “The commercialization of agricultural households implies the increased focus on market signals and comparative advantages in households’ production decisions, as opposed to a primary focus on subsistence production and the sale of purely the surplus that remains after the household’s consumption requirements have been satisfied.”
Commercialization of smallholder production: It is defined as “a process involving the transformation from production for household subsistence to production for the market” (Sokoni, 2007).
2.2. Theoretical Framework
2.2.1. Social risk: sociological perspectives
The development literature is rich with empirical and theoretical descriptions of social risk exposure and social risk management. Social risks are a global phenomenon, having irreversible and direct implications for society, incurring large-scale prejudices, depending on the cultural, political, and economic context in which this phenomenon manifests itself. In defining the concept of social risk, several approaches are taken which reflect the multidisciplinary character and the complexity of the applicability domains. The most widely recognized perspective of the subject matter of social risk in sociological publication are the ‘risk society perspective’ (Beck 1992); cultural theory (Douglas, 1992); and Governmentality (Foucault, 1991).
Becks’ (1992), Risk Society Theory, postulates that both the nature of and responses to risk differentiate late modern societies from earlier social formations. For Beck risks are manmade, invisible and can only be identified using expert technology. The criticism is that ‘risk’ is narrowed to the responses of technical and environmental risks as unforeseen consequences of industrialization. The narrowed view on technical and statistical risk management seems to be insufficient for the given complexity concerning, for example, governmental risk-strategies and rationalities (Dean 1999).
Cultural theory of Douglas (1992), postulates that risk takes a specific form in modern society. Risk is equated with dangers that threaten individual and collective security and existence. She argues that ‘modern’ risk is functionally equivalent to the premodern category of sin (Douglas 1990). Cultural theorists suggest that the dominant economic approach to risk based on the assumption that ‘all individuals are similarly rational or self-interested’ (Douglas et al. 2003, p. 99) cannot explain why individuals and social groups vary in the way they identify and respond to risks. Even though this approach was mainly criticized for an oversimplification of the far more complex and dynamic processes of how risk is dealt with and experienced(Tulloch and Lupton 2003) it is still used as a heuristic in order to analyze the different perceptions and responses to risk, (for example in organizations, Adams 2002).
In governmentality perspective, which has developed from the work of Foucault (for example Foucault 1991), risk refers to the concept of a new style of governance in modernity. In this approach risk is mainly understood as a concept entirely socially produced. There is no outer world, which forces society to respond to risk. Instead risk is understood as a specific way how to shape and control populations. Therefore it makes no sense to ask for more or less risk or how real risks are. Risk in this perspective is just a specific way to govern societies. Governmentality focuses on the ways in which disciplinary institutions such as hospitals and schools create knowledge about risks and the ways they should be collectively and individually managed. Within contemporary neo-liberal forms of government, individuals are expected to manage their own risks through self-regulation (Andy, 2009).
2.2.2. Theoretical perspectives of social risk management
Traditionally a large part of social risk management has been defined as either social policy or more narrowly welfare state policy. In this tradition, the rationale of the grounds for government interventions in an economy is motivated by efficiency and equity grounds (De Neubourg & Weigand, 2000). Interventions are categorized as either correcting the outcome (distribution) of market- and family-processes (usually on equity grounds) or correcting/substituting the functioning of markets (largely on efficiency grounds). One of theoretical developments that help to underpin this paradigm is document published by the World Bank that purports to establish a conceptual basis for analyzing social protection as part of a wider social risk-management framework (Draxler, 2000). The Social Risk Management (SRM) conceptual framework was used to help guide the World Bank’s 1st SP Sector Strategy in 2001 (World Bank, 2001). The original framework point out that SRM is related to how society manages income risks; where “income” is assumed to be a measurable proxy for tangible and non-tangible dimensions of household well- being.
Recently, global initiatives for addressing chronic poverty add forces to paradigm shift that required different interventions and an expanded framework for social risk management. Accordingly, the World Bank definite social protection as “private (both formal and informal) and public initiatives that connect men and women to labor markets, reduce people’s exposure to risks, and enhance their capacity to protect themselves against hazards and loss of income that threaten their present and future well-being” (World Bank 2011). Given the changing world, the evolution of Social Protection, the experience in using the original SRM approach, and responding to its critics; in the discussion paper No. 1930 of 2019 the World Bank, defined social risk management (SRM 2.0) as “how society manages income/consumption variability, manages the risks of poverty and vulnerability to poverty, and builds resilience to poverty over the lifecycle” (Jorgensen and Siegel, 2019).
In this context, policy documents have begun to look at the wider picture of welfare provision (actors and modes of delivery involved - public or market, informal or formal) and also to stress social protection as a productive factor. This also implies that the analysis of public interventions cannot be limited to social policy in the narrow sense, but should take into consideration public interference with families and markets as well. Also, investment of domestic resources in social protection is considered as a key issue. The allocation of domestic resources can not only contribute to institutionalize social protection as a state policy, but also signals political commitment to poverty reduction, food security and inclusive growth (FAO, 2017). The social risk management approach also gives more scope in the design of public policies to the type of risks and the diversity of strategies and arrangements to address them (Robert Holzmann, 2007).
One of the emphases of this approach is the role of institution in risk management. Based on the formality, institutional arrangements that help individuals, households and communities to better manage social risks falls into three categories. These are informal arrangements; market based arrangements; and publicly mandated or provided arrangements. They are derived from the notion of asymmetric information (institutional economics) in a world of diverse risks.
Informal arrangements: These arrangements have existed since the dawn of mankind and still constitute the main source of risk management for the majority of the world’s population (Robert H. et.al. 2003). With the lack of market institutions and public provisions, the response by individual households is self-protection through informal/personal arrangements (Robert and Jorgensen, 1999). Although they sidestep most of the information and coordination problems that cause market failure, they may not be very effective in helping the household weather adverse events. Lacking formal (anonymous) insurance markets households may also engage in personalized insurance, i.e., informal risk sharing . They build on direct information (which avoids moral hazard and adverse selection) and relationships developed over years or generations (trust). Examples include: marriage and the extended family; remittances between friends and neighbors.
Market-based arrangements: With the existence of market-based institutions, individual households use market instruments for managing risks (Robert and Jorgensen, 1999). Individual households take advantage of market-based institutions such as money, banks, and insurance companies when they are available. However, in view of these instruments’ limitations due to market failure or asymmetric information, their usage will be initially restricted but will rise with financial market development (Robert H. et.al. 2003). Because formal market institutions are reluctant to lend to households without secured earnings, microfinance is also an important instrument of social risk management. Moreover, marketbased arrangements may not be able to cope with the consequences of asymmetric information (moral hazard and adverse selection) and provide unemployment insurance or pension annuities only at grossly actuarially unfair prices.
Public arrangements: Public arrangements for dealing with risk came into being with the development of the modern welfare state but are relatively scarce and have very limited coverage in the developing world for fiscal and other reasons (Robert H. et.al. 2003). When informal or market based risk management arrangements do not exist, break down, or are dysfunctional, the government can provide or mandate (social)insurance programs for risks such as unemployment, old-age, work injury, disability, widowhood, and sickness. The mandatory participation in a risk pool can circumvent issues of adverse selection, in which individuals with low risk profiles avoid participation in insurance pools due to premiums while individuals with high risk profiles join in order to gain access to payouts. To overcome the effects of adverse selection, governments can mandate insurance of all unemployed (pooling), pursue meritorious goals (income redistributing and coping with myopia) or safeguard the government against strategic behavior of individuals as a result of minimum benefits (Robert and Jorgensen, 1999).
2.2.3. Models of poverty traps
The poverty traps literature highlights the crucial distinction between asset risk and income risk. Models of poverty traps emphasize the role of asset accumulation in shaping welfare dynamics, as asset ownership determines income generation processes. When income is endogenous, asset risk can have a more permanent impact than income risk. In particular, uninsured asset risk has the potential to drive a household onto a path of sustained asset loss, as a result of which the household falls below a critical asset threshold from which it is unable to recover, suffering persistent income poverty thereafter. As a result, households that face asset risk may be more vulnerable to falling into persistent poverty. The classic models of such phenomena are nutritional poverty traps, wherein those with insufficient income to sustain their energy levels become unemployable, leading to a low-level equilibrium trap (Dasgupta and Ray 1986, 1987; Dasgupta 1997).
Furthermore, households must consciously trade-off asset smoothing and consumption smoothing objectives when faced with correlated asset and income risk, as McPeak (2004) demonstrates in the case of rainfall shocks that affect both herd mortality (i.e., asset risk) and productivity (i.e., income risk) among livestock producers in northern Kenya. Shocks that deplete a household’s asset stock can lead to long-term poverty, as found in studies in Ethiopia, Indonesia, Kenya, South Africa (Adato et al., 2006; Barrett et al., 2006; Gertler and Gruber 2002; Lybbert et al., 2004). Beyond the intra-generational effects of asset shocks, another important consequence of uninsured risk is its effect on inter-generational poverty transmission is mediated by the disruption of children’s accumulation of human capital. If there exist poverty traps and such phenomena could arise simply due to decreasing absolute risk aversion and uninsured risk then (asset and income) risk exposure and management become especially important. The empirical evidence suggests the widespread prevalence of uninsured risk among households in the developing world and limited access to effective risk management strategies (Devereux, 2008).
2.2.4. Smallholders’ Commercialization
Promoting agricultural transformation will require a specific focus on market and value chain development that will help smallholder farmers to become sustainably profitable and respond effectively to market demand. Interventions to promote inclusive agricultural transformation needed too. Commercial transformation of subsistence agriculture is an indispensable pathway towards economic growth and development for many agriculture-dependent developing countries von Braun, 1994; Pingali, 1995; Timmer, 1997; Romer, 1994; and World Bank, 2008 works as cited in (Berhanu, 2010). Sustainable household food security and welfare also requires a commercial transformation of subsistence agriculture. Commercial agricultural production is likely to result in welfare gains through the realization of comparative advantages, economies of scale, and from dynamic technological, organizational, and institutional change effects that arise from the flow of ideas due to exchange-based interactions. Commercialization enhances the links between the input and output sides of agricultural markets (Berhanu, 2010). Modernization and commercialization of the smallholder agricultural sector provide the stimulus and impetus to reducing food insecurity in developing countries. Thus, the commercialization of agriculture is a means that benefits marginalized groups’ in particular rural women, and youth and a crucial strategy for reducing poverty and inequality.
Commercialization of smallholder agriculture has been a subject of considerable focus among policy-makers and development specialists not only at the level of farming households but also at the level of national and international policies. However, Von Braun (1995) and Govereh et al., (1999) argued that the perception that agricultural commercialization in developing countries undermines food security is “overly simplistic” and based on “if” statements. Examples of such “if” statements include: if food crops are replaced with nonfood crops and if markets are not working well; if landless farm laborers are replaced with highly mechanized and less labor intensive production systems. As such, von Braun (1995) concludes that many adverse effects of commercialization are not because of the inherent nature of commercialization, but instead, it is because of bad policies. The answer to these bad policies is policy reforms and not reversal or deceleration of technological advancement and commercialization (Boka, 2017).
2.2.5. Agricultural household model
Behaviors of farm households were best understood in a household-firm framework, where potentially important interactions existed between external labor markets (nonfarm labor markets), the farm operation, and household consumption. The agricultural household model is often used to describe the economic decision-making of rural households (Singh et al., 1986). In this model, when markets function perfectly, production and consumption decisions can be viewed as ‘separable.’ Under perfect markets and no uncertainty, the model assumes that all prices are determined through market mechanisms and that households are price takers. Households suffer no labor, credit, or other market constraints. There is no trade-off between the consumption of agricultural commodities and products for sale since there are no transaction costs in food markets. In this context, agricultural households solve the profit maximization and utility maximization problem recursively by first maximizing profits from agricultural production based on standard economic theory, and second, given that profit, maximizing utility (Taylor and Adelman, 2003). If the agricultural household model reflects reality, cash transfers should have little effect on agricultural production and instead only impact consumption (Boone et al., 2013).
Yet, rural markets in developing countries do not often function perfectly. Liquidity and credit constraints are key factors leading poor agricultural households to lower than optimal use of types and quantities of inputs. Further, agricultural households in less developed regions often rely upon assets, such as livestock, as a form of savings or insurance, often an unsatisfactory risk-coping strategy for a variety of reasons. The lumpiness of these assets increases the difficulties of using them for savings and, in the context of covariate risk, they usually drop in value as many households try to sell them as a coping mechanism. Without access to adequate credit markets and with poor alternative risk-coping mechanisms, agricultural households often adopt low-risk low-return income generation strategies, selling more than the optimal amount of labor off-farm in casual or exploitative labor markets to secure diverse sources of income (Dercon, 2002). It is in this context that social protection instruments and agricultural interventions can help rural households alleviate some of the constraints and market failures that underlie non-separable consumption and production decisions.
2.2.6. Transaction Cost Theory
The trajectory of institutional economics changed in the 1970s when new institutional economics (NIE) began to take shape around some relatively vague intuitions which eventually developed into powerful conceptual and analytical tools. The emergence of NIE is a success story by many measures: four Nobel laureates in less than 20 years, increasing penetration of mainstream journals, and significant impacts on major policy debates. This rapid acceptance is remarkable when we consider that it was divided from birth into distinct schools of thought (Menard, 2017). Transaction cost economics, for instance, pioneered by Coase in his 1937 article where he argues that market exchange is not costless and underlines the important role of transaction costs in the organization of firms and other contracts. Following Coase’s line of thinking Williamson has combined the concepts of bounded rationality and opportunistic behavior to explain the contractual choice and the ownership structure of firms (Rees, 2001). Neoclassical economics assumes that transactions are free of cost. In other words, transactions could occur based on market mechanisms without paying any costs. This view contradicts recent institutional economics thought which has the opposite assumption. The market will not function perfectly if the economic actor does not have information about goods which will be traded.
Incompleteness information and uncertainty refer to a situation where all parties involved in the transaction process are facing the same level of information which is vague. Asymmetric information could result in opportunistic behavior Stigler, 1961 as cited in (Priyanto and Khusaini, 2014). Asymmetric information among market partners, individuals, groups, and the government has an important bearing on the form and effectiveness of risk management instruments and on governments' capacity of achieving more equality in income and assets distribution. Under symmetric information among all economic actors and complete markets, the sources and characteristics of risks have no bearing for risk management. When individuals, households, or communities hold private information some risk markets may not be established, tend to break down, or function poorly. Insurance becomes only one and often not even the best choice to address risks, and for many risks’ insurance markets do not even exist. Debt and labor contracts emerge as a device to avoid costly state verifications (Holzmann, 2000). Informal risk-sharing mechanisms substitute for the market- based instruments, in particular at the beginning of economic development since the financial systems are very vulnerable to private information.
Asymmetric information causes two problems: adverse selection and moral hazard. In the case of adverse selection, farmers have better knowledge than do the insurers about the probability distribution of losses. Moral hazard similarly affects the incentive structure of the relationship between insurer and insured (World Bank, 2005). Recent advances in the importance of institutions as the backbone on which households and individuals can coordinate their expectations and thereby create effective collective action has been repeatedly demonstrated. And market exchanges depend crucially on institutions that can reduce or eliminate problems of adverse selection, moral hazard, and in general lower different types of transaction costs (Agrawal, 2011). In principle, there is an important role for the government in helping to establish, regulate, and supervise risk markets and to provide risk instruments where markets are bound to fail. Yet asymmetric information applies also to the relation between the citizen and the government leading to government failure and political risk.
Public policy help in the development of social protection programs aims to protect chronically poor households through social assistance, social insurance, or labor market programs (Davis et al., 2016). Evidence is emerging that by alleviating credit, savings, and liquidity constraints, such transfers can stimulate agricultural production through investment in technology and productive assets, and increased own-farm household labor allocation (Asfaw et al., 2014; Davis et al., 2016; Tirivayi et al., 2016; Todd et al., 2010; Kabeer et al., 2012; Hagen- Zanker et al., 2011). Evidence from several studies in Ethiopia, (e.g. Weldegebriel, B.Z. and Prowse, 2013; Ulrichs, 2016; FAO, 2015; USAID, 2012; and Devereux et al., 2008) have highlighted the potential effect of social protection on smallholder farming activities. The support provided by the Productive Safety Net Programme was allowed beneficiaries to increase their asset base (Ulrichs, 2016). In Ethiopia, for example, evidence shows the PSNP increased off-farm income, yet a large proportion of this stemmed from the sale of natural resources, with potential long-term environmental impact (Weldegebriel and Prowse, 2013). This has resulted in social protection becoming the greatest source of income for rural households in Ethiopia, surpassing that of smallholder agriculture considerably. However, the understanding of the linkages between smallholder commercialization and social risk management has not been based on in-depth empirical analyses.
2.2.7. Theory of Change
Following the rationale of the agricultural household model, the central assumption behind our theory of change is that consumption and production decisions are not separable for rural households living in a context of missing or incomplete markets. Multiple market failures and credit constraints may lead to suboptimal human capital investments, while weak or missing credit and insurance markets, inability to smooth consumption, savings, and liquidity constraints, risk aversion, inefficient farming systems, dysfunctional markets, market failures, nutritional shortfalls, public goods investments, and inequitable formal and informal regulation hamper agricultural production. Social risk management and smallholders' commercialization can play a vital role in reducing these constraints for smallholders' households.
The underlying principles of the agricultural household model and the other models previewed earlier help us identify potential impacts of social protection interventions on smallholders’ commercialization outcomes (income from the sale of farm products). Most of the studies Boone et al., 2013; Covarrubias et al., 2012; Radel et al., 2016; Todd et al., 2010, mainly in Latin American and sub-Saharan African (SSA) countries, evaluated the direct and indirect impact of social transfers on agriculture and focused on agricultural outcomes such as asset accumulation, input use, output, and labor allocation. While literature generally showed the positive role of social transfers in improving the productive capacity of beneficiaries, little is known about whether social cash transfers have positive or negative effects on the smallholder market participation (Sinyolo et al, 2017). Understanding the linkages between social transfers and agricultural interventions is important to improve the policy coherence between social protection and smallholder, as called upon in recent literature (e.g. Boone et al., 2013; FAO 2016; and Tirivayi et al., 2016).
The theory of change is also premised on the notion that the impacts of social risk management interventions and smallholders' commercialization policy are not parallel but are interlinked, such that they contribute to each other’s objectives of reducing risks and enhancing agricultural production. The first shared pathway of impact is the alleviation of credit, savings, and liquidity constraints. Social protection interventions like cash transfers or cash for work schemes can either improve savings or alleviate credit constraints and, if they are regular and predictable, they can improve access to credit by acting as collateral (Barrientos, 2012). Additional cash or disposable income resulting from a social protection intervention can also improve liquidity, a buffer for consumption shortfall, thus encouraging risk-taking and spending on inputs (Dercon, 1996). Agricultural interventions like microcredit, microfinance, and input subsidies may also alleviate the credit constraints of rural households. This would improve farm productivity and lead to gains in rural household welfare. The second shared pathway of impact is a certainty . The vagaries of weather and lack of insurance accentuate the risks and vulnerability that rural households face and they are often accompanied by substantial reductions in household consumption and assets. Not surprisingly, rural households are usually risk-averse (Barrientos 2012; Fenwick and Lyne 1999; Rosenzweig and Wolpin 1993; Morduch 1995). In this context, social protection instruments like cash transfers or social pensions, which are provided at regular and predictable intervals, can increase certainty and security and act as insurance against risks. Agricultural interventions can also increase certainty and security and assure rural households. Other interventions like weather-based crop insurance schemes directly address the lack of insurance and uncertainty related to weather variability.
The third pathway of impact specific to agricultural interventions is increased direct access to technology, knowledge, inputs, and factors of production ( e.g. land). As stated earlier, the lack of technology, knowledge, inputs, and factors of production limits agricultural productivity. There are several examples of productivity-enhancing agricultural interventions that can be used to address these constraints. These include input subsidies and grants; input technology (e.g. new high yield varieties, fertilizer); natural resource management techniques (e.g. soil conservation practices, irrigation); land tenure reform; marketing arrangements (e.g. producer organizations, contract farming); and macroeconomic reforms (e.g. price liberalization). Infrastructural interventions, such as roads, increased access to local markets, and market information. Other interventions like farmer field schools and extension services enhance access to agricultural knowledge and skills. Some social protection interventions like public works programs may also work through this third pathway since they facilitate access to relevant knowledge and skills and rural infrastructure.
2.3. Analytical Framework
2.3.1. Institutional analysis paradigm
Evolution of the paradigm debate and the relationship to research commenced with a challenge to the dominance of the mono method era during the 1960s and resulted in the emergence of mixed methods and later in the 1990s of mixed model eras Creswell 2003 as cited in (Armitage, 2007). The debate over the relationship between paradigm and methodology leads to create new sets of beliefs that guide new kinds of actions and thus debate leads to the emergence of the third set of beliefs (third way) the pragmatic paradigm. Pragmatism includes perspectives of post-positivism and constructivism. It avoids the use of metaphysical concepts and it presents a very practical and applied research philosophy (Tashakkori et al., 2003; and Baert 2005) as cited in (Werthmann, 2018).
The mixed methodologists believe that qualitative and quantitative methods are compatible and that the incompatibility thesis of (post-) positivism and constructivism stating that these methods are incompatible due to the incompatibility of the paradigms underlying those are rejected (Howe 1988; and Tashakkori et al., 2003 as cited in Werthmann, 2018). Mixed methods research, in turn, arose in a period when it was expected that any approach to social research would have a metaphysical paradigm explicitly based on the philosophy of knowledge. According to Moran Ellis et al., (2006), a mixed-method approach can be defined as the “use of two or more methods that draw on different meta-theoretical assumptions.”The mixed-method approach has been adopted for this study in an attempt to combine the strong point and elements of both qualitative and quantitative data collection and analysis techniques. Three reasons have taken into consideration while adopting a mixed-method as appropriate methods for this study.
First, the paradigm as a set of beliefs, illustrated above, is one of the reasons that can be argued that the mixed methods can be adopted for social research endeavors as this is going well together with the mixed quantitative and qualitative approach. The combing of quantitative and qualitative research includes the logic of triangulation, an ability to fill in the gaps left when using one dominant approach.
Secondly, selecting the right research method starts with identifying the research question and study aims. A mixed-methods design is appropriate for answering research questions that neither quantitative nor qualitative methods could answer alone Tashakkori et al ., as cited by (Shorten and Smith, 2017). Mixed methods research requires a purposeful mixing of methods in data collection, data analysis, and interpretation of the evidence.
Thirdly, the methods in empirical institutional analysis, most importantly, in the selection of methods for institutional analysis, level of analysis, research questions, time horizons, measurability and observability of institutions and the roles played by actors are, or should be, central concerns and helps to precisely distinguish between differently oriented investigations within a common theme (Beckmann, 2009). A mixed-method allows for in-depth analysis and with the choice of methods this investigation receives depth (mainly through qualitative methods) and breadth (mainly through the use of quantitative methods). The approach supports obtaining holistic understanding results from noting trends and generalizations as well as in-depth knowledge of participant’s perspectives (Creswell, 2007).
From the perspective of social analysis, this study focus on the governance level (L3) deals with organizations and the appropriate choice of contractual relations/arrangements. Markets, firms, public agencies, and contracts are located on this level as well as actors as individuals include consumers/producers, members of households and firms and as organizations households, firms, markets, and networks. Econometrically, multivariate probit model was adopted for this study to assess factors that determine households’ decisions to choose institutional arrangements for social risk management. It helps to estimate several correlated binary outcomes jointly because it simultaneously captures the influence of the set of explanatory variables on each of the different institutional arrangement choices, while allowing for the potential correlations between unobserved disturbances, and the relationships between the choices of different market outlets (Greene, 2002). Multivariate Probit model has been used widely in assessing discrete choices for example (Ondieki-Mwaura et al., 2013) determinants of participation in identified institutional arrangements, Goshu et al., (2016) assess determinants of access to credit and credit source choice by micro, small and medium enterprises, and (Msami and Ngaruko, 2014) assess determinants of the choice of institutional marketing arrangements by small poultry businesses.
2.3.2. Impact assessment method
An impact evaluation provides information about the impacts produced by an intervention - positive and negative, intended and unintended, direct and indirect. This means that an impact evaluation must establish what has been the cause of observed changes (in this case impact referred to as causal attribution (also referred to as causal inference). The intervention might be a small project, a large programme, a collection of activities, or a policy. Impact refers to the positive and negative, primary and secondary long-term effects produced by a development intervention, directly or indirectly, intended or unintended (OECD-DAC, 2010), or it is the longer term intended and unintended results of a programme: technical, economic, socio-cultural, institutional, and environmental or other (World Bank, 2007).
Impact assessment can focus on whether a policy or intervention has succeeded in terms of its original objectives, or it may be a wider assessment of overall changes caused by the policy or intervention - positive and negative, intended or unintended (Roche, 1999). The impact of the full range of development interventions can be assessed using qualitative or quantitative approaches or a mix. Roche defines impact assessment as the systematic analysis of the lasting or significant changes - positive or negative, intended or not - in people’s lives brought about by a given action or series of actions (Roche, 1999). Impact assessment may be on-going in the form of monitoring, may be occasional, in the form of mid-term reviews, or may be ex-post, in the form of ex-post impact assessments or evaluations. Ex-post evaluations are often used to learn lessons for the future and to judge whether the intervention is suitable for replication (Kate, 2002). Impact assessment methods can also be classified as “before and after” and “with and without” approaches.
“Before and after” compares the performance of key variables during and after the program, with those prior to the program. This simple approach can give an idea of the change that occurs over the course of an intervention but should be regarded as part of a monitoring system rather than as providing evidence of the causal impact of an intervention since there is no way of knowing if an observed change should be attributed to the intervention in question or to other circumstances (ILO, 2018). This approach uses statistical methods to evaluate whether there is a significant change in some essential variables over time. It often gives biased results because it assumes that had it not been for the program, the performance indicators would have taken their pre-intervention-period values. A before-and-after comparison attempts to establish the impact of a program by tracking changes in outcomes for program participants over time (Schomburg, 2016). Performing before-and-after comparisons could make sense if there are reasons to believe that, in the absence of the treatment, outcomes would, on average, remain unchanged. These limitations persist and, unlike well- implemented experimental and quasi-experimental methods, simple before-and-after comparisons cannot be considered robust impact evaluations.
With and without comparisons compare the behavior in the key variables in a sample of program beneficiaries, with their behavior in non-program participants (a comparison group). This is an approach to the counterfactual question, using the experiences of the comparison group as a proxy for what would otherwise have happened in the program beneficiaries (AIEI, 2011). If one could observe the same individual at the same point in time, with and without the project, this would effectively account for any observed or unobserved intervening factors or contemporaneous events and the problem of endogeneity do not arise (Ravallion, 2005; Gilligan et al., 2008). Inputs into a project lead to direct outcomes or impacts through produced output, or through the impact on other variables which then impact on outcomes. There may be intervening factors on which the project has an effect that are either observed or not observed which also contribute to outcomes. In addition, there may also be other factors or events that are correlated with the outcomes, but which are not caused by the project (Kene et.al., 1999). Accurate and successful evaluations are those which are able to control for these effects. This task of “netting out” the effect of the program from other factors is facilitated if control groups are introduced. “Control groups” consist of a comparator group of individuals or households who did not receive the intervention, but have similar characteristics as those receiving the intervention, called the “treatment groups”. Identifying these groups correctly is a key to identifying what would have occurred in the absence of the intervention (Ezemenari et al., 1999; Gilligan et al., 2008).
The proper analysis of impact requires a counterfactual of what those outcomes would have been in the absence of the intervention. The counterfactual is necessary for comparing actual outputs and outcomes to what they would have been in the absence of the intervention, i.e. with versus without. Impact Evaluations require comparisons between beneficiaries that are assigned to either a treatment or a control (counterfactual) group to measure attribution between interventions and observed outcomes (USAID, 2013). However, experimental designs are based on randomly selected groups put in contrast to randomly selected comparison groups (the counterfactual), which makes it obsolete to go further into this matter. An alternative to the experimental approach using randomization is the quasi-experimental approach. In that case, other methods can be used that seek to undertake internally valid comparisons by constructing a valid counterfactual. These are called quasi-experimental impact evaluation methods (ILO, 2018). The main benefits of quasi-experimental designs are; they can draw on existing data sources and are thus often quicker and cheaper to implement, and they can be performed after a project has been implemented, given sufficient existing data (Regalia, 1999).The most commonly used ones are difference-in-differences (DID), reflexive comparison, and propensity score matching (PSM), all of which will be briefly introduced in this section.
Difference in-differences (DID: One of the most commonly used techniques is the difference in-differences (DID) approach, which compares the change in outcomes experienced by the treatment group with the change in outcomes experienced by the comparison group. It is a method in which one compares a treatment and comparison group (first difference) before and after a project (second difference) (Baker, 2000). The difference in a given outcome between recipients of the project (the treatment group) and a comparison or control group is computed before the project is implemented. This difference is called the “first difference”. The difference in outcomes between treatment and control groups is again computed some time after the project is implemented, and this is called the “second difference”. Under the difference-in-difference technique, the impact of the project is the second difference less the first difference. The logic is that the impact of the project is the difference in outcomes for treatment and control groups after the project is implemented, net of any pre-existing differences in outcomes between treatment and control groups that pre-date the project (AIEI, 2011).
The strength of DID estimator comes from its intuitive appeal and simplicity. This method provides a way to account for both observable and unobservable differences between participants and nonparticipants. More precisely, it controls for all individual effects that remain constant over time, or that share the same course of change over time (i.e. treated and comparison groups show similar trends in the outcomes of interest). Even if the method is not experimental, it allows for a (partial) check of the assumption that renders it internally valid. This implies that we can have a sense of whether our estimated impacts are valid or not. If good administrative data is available, the method can be applied fairly easily and even expost, based on before and after data from the programme. This method has two disadvantages. First, the methods produce less reliable results than randomized selection methods. Secondly, n order to test the key assumption of “ common trends ”, at least three data collections are required, so the implementation can be expensive if data are not available initially.
Reflexive comparison: It is another type of quasi-experimental design. This method requires a baseline survey of program beneficiaries before the program is implemented and a followup survey (Regalia, 1999). Here, participants who receive the intervention are compared to themselves before and after receiving the intervention. The counterfactual group is the set of participating individuals themselves (Jalan and Ravallion, 1999; Baker, 2000; World Bank, 2011). The main advantage of reflexive methods is that they make possible the evaluation of policies that cover the entire population, not just subgroups (Baker, 2000). There is, however, a major drawback with reflexive comparisons: the situation of program participants before and after the intervention may change owing to myriad of reasons independent of the program (World Bank, 2011). Likewise, this method cannot identify the impact of the program from that of other factors ( e.g. economy wide changes) that have affected the beneficiaries. For this reason results are biased, and the direction of that bias is difficult to assess (Regalia, 1999).
Propensity Score Matching (PSM): Propensity score matching (PSM) is a very commonly used approach among the quasiexperimental evaluation methods. Its basic principle is to construct a comparison group by matching participants with similar non-participants, based on their predicted probability of participating in the intervention. This is called the propensity score, which is calculated based on a range of observed characteristics. Propensity scores matching method was adopted for this study in an attempt to estimate the effect of receiving treatment when a random assignment of treatments to subjects is not feasible. Matching has become a popular approach to estimate causal treatment effects and is found in very diverse fields of study (Kopeinig, 2005).
One key use of the propensity score is to reveal when it is simply impossible to compare groups. Researchers should always plot the distributions of propensity scores in the treatment groups. If the groups have little overlap in propensity scores, they are inherently incomparable, and no statistical tricks can overcome this problem (Sainani, 2012). Propensity score matching (PSM) refers to the pairing of treatment and control units with similar values on the propensity score, and possibly other covariates, and the discarding of all unmatched units (Rubin, 2001). This method allows one to find a comparison group from a sample of non-participants closest in terms of observable characteristics to a sample of program participants (Baker, 2000). PSM is a particularly useful method when large and rich amounts of secondary data are available, as these are necessary to define a good propensity score and to match sufficient numbers of participants and non-participants with similar scores, i.e. to find a large enough region of common support.
Furthermore, PSM relies on the assumption that only observed factors influence both participation and outcomes (conditional independence assumption). Thus, PSM should only be applied if there is a good understanding of the drivers of programme participation and the outcomes of interest, and should be avoided if unobservable characteristics can be expected to affect those variables. However, in PSM the sample size of beneficiaries should be large enough to ensure the high level of the statistical significance of the estimated parameters (Heckman et al., 1997; Bernard et al., 2007). Group overlap must be substantial, and hidden bias may remain because matching only controls for observed variables (to the extent that they are perfectly measured) (Vasisht, 2007). Irrespective of its shortcomings, PSM is extensively used in the recent literature on economic impact evaluation (Jalan and Ravallion 2003). Propensity scores matching method was adopted for this study in an attempt to estimate the effect of receiving treatment when a random assignment of treatments to subjects is not feasible.