Resilience Against Chronic Poverty: Some Reflections and An Agenda Christopher B. Barrett Cornell University Seminar to Harvard University Sustainability Science Program October 24, 2012 Motivation Resilience has quickly become a buzzword in the development and humanitarian communities. Two big drivers: 1) Perceived increasing risk – climate, mkts, macroeconomy, violence, etc. – in both frequency and intensity 2) Recurring crises lay bare the longstanding difficulty of reconciling humanitarian response to disasters with longer-term development efforts. Many recent calls for renewed efforts to “build resilience” quite explicitly aim to align humanitarian and development objectives. But we lack a theory-measurement-and-evidence-based understanding of what resilience is, how to measure it, and how to effectively promote it so as to reduce chronic poverty. Motivation Shocks that disrupt lives and livelihoods: the single greatest cause of descents into chronic poverty (Krishna, etc.) Uninsured risk of catastrophic loss (stressors): a key structural reason for poverty traps (Carter&Barrett; Santos&Barrett) We seek to bridge the ecological/engineering literatures on resilience with the social science literature on poverty traps to: - advance a theory of resilience against chronic poverty - tease out measurement principles appropriate to the theory - build toward a body of empirical evidence on resilience Focus Resilience of whom to what? Subject of interest – quality of life, roughly Sen’s ‘capabilities’. This implies a focus on individuals’ (and groups’) well-being within a system, not the state of a system itself. System has instrumental rather than intrinsic importance. Focus further on minimizing the human experience of chronic poverty. We therefore focus on places with high rates of chronic poverty. Do not focus on a specific source of risk b/c problem is uninsured exposure to a wide array of stressors and shocks to which resilience implies adaptability while staying non-poor. Toward a theory We need to adapt ecological/engineering theory to the development/humanitarian response context. As used in ecology or engineering – e.g., “the ability of the system to maintain its identity in the face of internal change and external shocks and disturbances” (Cumming et al. 2005 Ecosystems, p. 976) – resilience is not necessarily desirable for populations trapped in chronic poverty. Their objective may be escape from – not persistence in -- their present state of existence. To be useful for development policy, we need resilience to be a normative property, to be orderable – and preferably decomposable (in FGT sense) – in order to offer a useful metric to gauge performance and guide policy/programming. Toward a theory Figure 1: Nonlinear expected well-being dynamics with multiple stable states Humanitarian emergency zone E[future] capabilities Death Death T1 Chronic poverty zone Non-poor zone T2 Current capabilities Noncontroversially: NPZ >> CPZ >> HEZ Those in CPZ or HEZ are chronically poor in expectation The CEF reflects indiv/collective behaviors (agency/power) w/n system Toward a theory Figure 1: Nonlinear expected well-being dynamics with multiple stable states The humanitarian ambition is to keep people from falling into HEZ … offers foundation of a rightsbased approach to resilience. E[future] capabilities Humanitarian emergency zone The development ambition is to move people into the non-poor zone and keep them there. Death Death Chronic poverty zone Non-poor zone Current capabilities For the current non-poor, seek ‘resilience’ against shocks in the ecological sense: no shift to either of the lower, less desirable zones. But for the current poor, those in HEZ/CPZ, the objective is productive disruption, to shift states. Asymmetry is therefore a fundamental property of resilience against chronic poverty. Thus stability ≠ resilience. Toward a theory A Utopian, asymmetric vision of well-being dynamics: Figure 2: Desired expected well-being dynamics with multiple stable states E[future] capabilities Humanitarian emergency zone Egalitarian option Death Death Chronic poverty zone Non-poor zone Current capabilities ‘Egalitarian option’: engineering concept applies - return to initial state. ‘Random walk w/safety net option’: implies perfect downward resilience at NPZ/CPZ boundary … but zero resilience upward or w/n NPZ. Toward a theory Explicitly incorporate risk, move from CEF to CTD: Figure 3: Nonlinear well-being dynamics with conditional transition distributions Humanitarian emergency zone Future capabilities Death Death Chronic poverty zone Non-poor zone Current capabilities Note: The shape of the CTD affects the shape of the CEF Transitory shocks (- or +) can have persistent effects Risk may be partly endogenous to system state Toward a theory Feedback between sub-systems can be crucial If we represent the preceding conditional transitions as: Wt+1=g(Wt|Rt,εt) where W is welfare, R is the state of the natural resource, and ε is an exogenous stochastic driver Then simply introducing feedback between R and W (e.g., range conditions depend on herd size/stocking rate) Rt+1=h(Rt|Wt,εt) or allowing for drift in ε (e.g., due to climate change) means the underlying CTD changes over time. Then the resilience of the underlying resource base becomes instrumentally important to resilience against chronic poverty. Programming implications Objective: min likelihood people fall into HEZ/CPZ Three options: 1) Shift people’s current state – i.e., move initial state rightward. Ex: asset transfers: cash, education, land. 2) Alter CTDs directly (and thereby ∆ system too). Ex: social protection - EGS, insurance, improved police protection, drought-resistant animal/plant genetics. 3) Change the underlying system structure – institutions/ technologies – induces ∆ in behaviors and CTDs. Prob: multi-scalar reinforcement – ‘fractal poverty traps’ Systems modeling becomes important to reveal the structure – and possible intervention points – behind univariate dynamics. Programming implications The importance of social institutional arrangements “A tale of two widows” And would the widower’s dynamic = the widow’s? Toward measurement Define resilience as a – or perhaps function (e.g., discountweighted avg probability) of the – sequence of period-specific Pr(well-beingt)<poverty line Pr(well-being) < Pov. Line 1 0 Chronically poor just >T1 Marginally poor just <T2 Begins in non-poor zone just >T2 Time Big issues: - defining the poverty line? - units of observation – individuals? households? aggregates? - frequency of longitudinal observation (retrospective/prospective)? - how to estimate probabilities? Objective/subjective? - how to allow x-sectional heterogeneity in CTDs/CEFs? - how to triangulate with subjective and qualitative measures? Build empirical evidence Need to study interventions aimed at improving resilience and replicate across contexts: Candidates to discuss: i. Index-based livestock insurance for pastoralists ii. Safety nets (NREGS in India, PSNP in Ethiopia) iii. Soil health interventions (e.g., fertilizers, NRM) in African smallholder agriculture iv. Accelerated disaster response interventions (e.g., LRP of food aid vs. traditional, transoceanic deliveries) Develop longitudinal data on individuals and households integrating qualitative and quantitative measures in sentinel sites. Where ethical/feasible, use RCTs or exploit natural/policy discontinuities to identify causal effects. Use results to develop clear policy/programming guidance. Ex: HSNP vs. IBLI in Kenya; post-drought herd restocking; livestock gift programs Summary Resilience is a popular buzzword now. But little precision in its use, either theoretically or empirically. Aim to help facilitate rigorous, precise use of the concept to help identify how best to reduce chronic poverty. This will require advances in theory, measurement and empirical work in many different contexts and over time. Much to do in all of these areas … a massive research agenda. Thank you Thank you for your time, interest and comments!