Abstract Building household resilience in the face of repeated shocks is crucial to achieving the development goals of low‑income countries. In this study, we examine the role of social protection programmes in enhancing household resilience in Malawi, using four rounds of the integrated household panel survey data. Using a resilience indicator based on the Resilience Index Measurement and Analysis methodology developed by the Food and Agriculture Organization of the United Nations, we show that shocks, such as the high costs of agricultural inputs, floods, and irregular rains, have a negative relationship with resilience. Meanwhile, improvements in education, engagement in self-employment activities, and household savings increase household resilience. However, we found that coupon subsidies and cash transfers have an insignificant relationship with resilience, suggesting that they should be strengthened to improve their effectiveness through better targeting.
Keywords social protection, repeated shocks, household resilience, panel data, Malawi.
The frequency and intensity of climatic shocks and extreme weather events such as floods, heat waves, droughts, and cyclones have increased across Africa over the years (Diallo and Tapsoba 2022; Falco, Kis and Viarengo 2022). These shocks negatively affect communities by reducing the agricultural production and income of smallholder households, leading to reduced food consumption and less diverse diets (Ha et al. 2022; McCarthy et al. 2021; Vo, Mizunoya and Nguyen 2021). Research has also shown that the increased incidences of shocks are associated with reduced resilience (Chishimba and Wilson 2021; Ngoma, Finn and Kabisa 2024).
Social protection programmes have been widely utilised as policy instruments to improve the livelihoods of households and protect against shocks (Abay et al. 2022; Otchere and Handa 2022; Premand and Stoeffler 2020). A substantial body of research has shown that cash transfer programmes have strong effects on improving food security, livelihoods, human capital development (especially education), and children’s wellbeing (d’Errico et al. 2020; Handa et al. 2022; Stoeffler and Premand 2021; UNICEF 2021). However, evidence of the impacts on longer-term resilience is mixed. In Kenya, Matata, Ngigi and Bett (2023) found that cash transfers help build household resilience. Similar results were observed in Malawi, where Otchere and Handa (2022) found that unconditional cash transfers significantly boost household resilience. In contrast, Beegle, Galasso and Goldberg (2017) found no evidence of a relationship between public works programmes and food security in Malawi.
Recent synthesis evidence suggests that although cash transfers may be effective in improving short-term welfare and strengthening absorptive capacity, they are often insufficient on their own to generate sustained adaptive resilience (Pruce, Price and Sabates-Wheeler 2025). Moreover, strengthening resilience through social protection may require sustained and continuous programme participation rather than short-term engagement. This growing body of evidence points to the importance of combining cash transfers with complementary interventions – such as asset building, livelihoods support, or skills development – to achieve stronger and lasting resilience outcomes (Abay et al. 2022).
This study analyses the relationships between shocks and social protection programmes and household resilience in Malawi. The country serves as an ideal context for this analysis due to its high vulnerability to climate-related disasters, particularly droughts and floods, and other market shocks. Over the past five decades, the country has experienced more than 19 major floods and seven droughts. These shocks have increased in frequency, magnitude, and scope (Government of Malawi 2019a, 2019b). The increased frequency of shocks has affected households’ livelihoods, and Malawi may be particularly vulnerable because over half (50.7 per cent) of the population live below the poverty line, while 20.5 per cent are extremely poor (NSO 2020).
The study seeks to fill a number of literature gaps. First, most of the existing studies have used cross-sectional data to understand the short-term effects of individual programmes (such as cash transfers or public works) on household welfare without assessing their roles in building households’ absorptive, adaptive, and anticipatory capacities (Beegle et al. 2017; Otchere and Handa 2022; Premand and Stoeffler 2020; Ulrichs, Slater and Costella 2019). The use of cross-sectional data and a focus on programme-specific evaluations limits the understanding of resilience dynamics over time and comparison of the relative effectiveness of the various social protection programmes. Second, most studies use outcomes such as food security, consumption, or child wellbeing (Handa et al. 2022; Stoeffler and Premand 2021) without aligning the outcomes to multidimensional resilience frameworks such as the Resilience Index Measurement and Analysis (RIMA) methodology developed by the Food and Agriculture Organization of the United Nations (FAO). Third, available evidence for Malawi is geographically fragmented with some evaluations focused on specific programme sites (e.g. Chiwaula and Waibel 2011) as opposed to nationally representative analysis. Finally, there is limited research in Malawi that considers exposure to multiple shocks, which includes climatic and economic shocks. Abay et al. (2022) further note that the linkages between social protection and asset-building or income-generation activities remain underexplored, especially in shock-prone and low-income settings such as Malawi.
This study contributes to addressing the above-identified research gaps by examining the relationship between shocks, participation in social protection programmes, and household resilience, using longitudinal and nationally representative data and applying a multidimensional resilience framework. Specifically, the article provides insights into how social protection programmes can mitigate the adverse effects of shocks faced by households, offering valuable evidence for policymakers and stakeholders aiming to strengthen resilience. Furthermore, the study analyses resilience dynamics over time and the factors associated with the changes in resilience status.
The next section summarises the conceptual framework for the study. Section 3 discusses the data sources and methods used in this study. Section 4 presents and discusses the findings. Section 5 concludes the study and provides policy recommendations.
The concept of resilience characterises the capacity to resist and recover from shocks (Premand and Stoeffler 2020). Following Ulrichs et al. (2019), resilience can be disaggregated into three key capacities of absorptive, anticipatory, and adaptive, each of which is enhanced by different types of social protection. Absorptive capacity, which denotes a household’s capability to manage and mitigate the immediate effects of shocks, can be strengthened using social protection instruments such as unconditional cash transfers, food assistance, and school feeding programmes. Anticipatory capacity relates to households’ ability to prepare for and lessen their vulnerability to future shocks. This can be strengthened through social protection programmes such as public works programmes and shock-responsible cash transfers. Adaptive capacity encompasses interventions that help households build more resilient livelihoods, including long‑term investments such as human capital accumulation, income diversification, savings mobilisation, and productive asset accumulation.
Social protection programmes seek to reduce the impact of shocks, improve coping, strengthen interventions that prevent shocks, and build the long-term resilience of households (Abay et al. 2022; Adato, Ahmed and Lund 2004; Otchere and Handa 2022; Premand and Stoeffler 2020). The programmes promote the resilience of households by enhancing their ability to prepare and protect themselves against shocks, as well as promoting the recovery process after experiencing a shock (Hoddinott et al. 2012). Social protection programmes that directly improve welfare outcomes for a household, such as food security, play a critical role in strengthening a household’s ability to withstand the immediate impact of shocks. In contrast, programmes that support accumulation and restoration help to build long‑term resilience by building both adaptive and absorptive capacities. Considering that households face varying levels and forms of vulnerability, resilience-building interventions must be appropriately tailored. For example, households that have an adequate productive asset base require social protection programmes that aim to help them absorb a shock. However, households with limited productive assets, or those whose asset base has been destroyed by the shock, require social protection programmes that focus on rebuilding the asset base (Chiwaula and Waibel 2011).
3.1 Data and measures
The study uses publicly available panel data from the Living Standards Measurement Study – Integrated Surveys on Agriculture, implemented by the National Statistical Office with technical support from the World Bank. Specifically, four rounds of the Integrated Household Panel Survey data (IHPS) collected in 2010, 2013, 2016, and 2019 are utilised. The panel data consists of 1,619 households interviewed in 2010, 1,990 households in 2013, 2,508 households in 2016, and 3,178 households in 2019. Our analysis is based on a balanced panel of 1,017 households that were successfully interviewed in all four survey rounds.
Using this data set, the study employs the RIMA approach to measure household resilience capacity in Malawi. The RIMA methodology was developed by FAO and first used in 2008 as a tool for measuring resilience to food insecurity. It was later improved in 2016 by addressing a number of limitations identified in previous applications (FAO 2016; Otchere and Handa 2022). The improved and updated methodology (RIMA-II) has four fundamental resilience pillars, namely Access to Basic Services (ABS), Adaptive Capacity (AC), Assets (AST), and Social Safety Nets (SSN).
While there are other methods to measure resilience (e.g. the Cissé and Barrett and the Technical Assistance to NGOs (TANGO) methodology), the RIMA approach is the most widely used method for measuring resilience (Upton, Constenla-Villoslada and Barrett 2022). However, one challenge with RIMA is that some variables used to calculate the Resilience Capacity Index (RCI) are also potential outcomes of resilience. This creates an overlap between inputs and results. Nonetheless, the RIMA approach has been widely applied and tested in many contexts (e.g. Otchere and Handa 2022; d’Errico et al. 2020; Alinovi et al. 2010) and is now the recommended approach under the African Union’s Comprehensive Africa Agricultural Development Programme and of various United Nations organisations (Otchere and Handa 2022; Upton et al. 2022).
The RIMA-II approach estimation of the RCI is based on a two-stage procedure. In the first stage, the resilience pillars are estimated from observed variables through factor analysis, while in the second stage, a single summary index of resilience capacity, namely the RCI, is computed from the pillars using the Multiple Indicators Multiple Causes model. The RCI, which reflects a household’s resilience capacities, can be used to rank households from the least to the most resilient and to analyse the determinants of household resilience (FAO 2016; Otchere and Handa 2022). The RIMA model is mathematically presented in equation 1:
(1) RCI = f[(β1,β2 ... βn), (ABS, AC, AST, SSN)]
where the RCI is a function of the four RIMA pillars as defined earlier; namely, ABS, AC, AST, and SSN, and the coefficients from calculating the pillars β1,β2…βn. Thus, both the pillars and RCI are latent variables calculated from a set of observable variables. The set of variables is the same over the years so that the changes in the RCI are due to changes in the variable values.
FAO (2016) makes some recommendations regarding the variables to be included in each of the pillars. Following this guidance, Table 1 provides a description of the RIMA pillars and the corresponding set of variables used to construct them and, subsequently, the RCI.
FAO’s RIMA framework is well aligned with contemporary theoretical understandings of resilience. In particular, RIMA’s core pillars capture the multifaceted nature of vulnerability. For example, recent work by Sengupta and Costella (2023) conceptualises adaptive capacity outcomes across five dimensions of vulnerability: social, economic, physical, ecological, and institutional. These dimensions align closely with the RIMA pillars. Access to basic services is closely associated with physical and institutional vulnerability, highlighting spatial disparities in access to infrastructure and services. Economic vulnerability is linked to assets while adaptive capacity is linked to the dimensions of social and human capital vulnerability, particularly with respect to education and livelihood diversification. The social safety nets pillar is related to institutional and social vulnerability, reflecting the extent to which households can rely on formal and informal support systems.
While RIMA remains highly relevant to contemporary resilience discourse, one of the recognised limitations of the RIMA framework is that the RCI does not distinguish between types or intensities of shocks (Barrett and Constas 2014; d’Errico and Di Giuseppe 2018). While the index captures a household’s underlying absorptive, adaptive, and anticipatory capacities by construction, it does not differentiate as to whether households were exposed to mild, moderate, or severe shocks during the reference period. This may affect comparability between households with identical RCI scores but who may have faced very different shock environments, with one being exposed to more severe events such as acute drought or unusually high price increases. While we acknowledge this limitation, we differentiate between the different types of shocks on resilience and analyse their associations with resilience. Furthermore, through the use of panel data, we analyse resilience over time and thus observe changes over time under different shock environments.
3.2 Analysis
We begin with a temporal descriptive analysis of the patterns of shocks, social safety nets, and resilience. To inform geographical targeting, the study includes spatial comparisons of resilience across Malawi’s three administrative regions, namely Northern, Central, and Southern.
Following the descriptive analysis, we turn to econometric analysis to better understand the relationships between shocks, social protection, and resilience. To do this, we use the following regression model, which specifies the effects of shocks and social protection programmes on resilience (equation 2),
(2) RCIit = β0 + β1Sit + β2SPit + β31Xit + ci + θit
where RCIit represents the outcome variables, namely the RIMA Resilience Capacity Index for household i at time t for T = 1,2,3,4; Sit represents adverse shocks that households reported to have faced in the past 12 months; SPit represents household participation in various social protection programmes (such as cash transfers, school feeding programmes, and public works programmes) in the past 12 months; Xit represents a set of household socioeconomic factors (age of household head, education, assets, access to basic services, etc.); ci is the unobserved time-invariant fixed effect; and θit is the error term.
The study also analyses resilience dynamics between 2010 and 2019 to see how resilience status has changed over time and the factors associated with the observed changes in the resilience status of households. First, we generate a categorical variable distinguishing the various resilience states, namely equal to 1 if a household became less resilient, equal to 2 if a household stayed in the same resilience group, and equal to 3 if a household became more resilient. Second, we analyse the determinants of household resilience trajectories by estimating an ordered probit model (equation 3) as follows:
(3) Yit = ∂i Zit + μit
The dependent variable, Y, captures our three outcomes: (1) became less resilient, (2) maintained the same resilience status, and (3) became more resilient. The dependent variable is regressed on a set of variables, including shocks and participation in social protection programmes (the same variables in equation 2 above). ∂ is a vector of parameters to be estimated, Zit represents household socioeconomic and demographic variables including shocks, and μit is the error term. From the ordered probit estimation, we obtain coefficients and average marginal effects.
4.1 Descriptive results
This section provides descriptive analysis results for our study, focusing on key household demographic and socioeconomic characteristics (Tables A1 and A2 in the Appendix).
4.1.1 Patterns of shocks and social safety nets
Data shows that the average number of shocks faced by households increased significantly from 1.8 shocks in 2010 to 3.4 shocks in 2019. Similarly, the highest number of shocks reported to have been faced by a single household increased from 7 in 2010 to 21 in 2019 (Table A3 in the Appendix). The shocks include unusually high costs of agricultural inputs, unusually high food prices, illness or injury in the household, floods, drought, irregular rain patterns, landslides, death in the household, and end of regular assistance received, among others. The Southern region consistently recorded the highest participation in social protection programmes, given the high social support it receives from the government and development partners due to its greater vulnerability to climatic shocks (Table A4 in the Appendix). Patterns of social protection programmes by durable asset wealth quintile show that the poorest quintile, while receiving more support than the richest, consistently receives less than the middle and the poor quintiles, suggesting targeting inefficiencies (Table A5 in the Appendix).
4.1.2 Correlations between RIMA and its components
All pillars except SSN show highly significant correlations with the Resilience Capacity Index (RCI) (Table 2). The Assets pillar is the strongest contributor to the RCI followed by the Adaptive Capacity pillar and Access to Basic Services. The Social Safety Nets pillar shows no significant correlations with the RCI or any other pillar. This may be due to poor targeting of the social safety nets, and low coverage, or the benefits may be too small to meaningfully impact resilience.
4.1.3 Resilience capacity: temporal and spatial variations
The RIMA resilience scores are rescaled from a minimum of 0 to a maximum of 100. Results show that the RIMA resilience capacity of households increased between 2010 and 2019 (Figure 1), suggesting a general strengthening of household resilience capacities to absorb and adapt to shocks during the reference period.
RCI comparisons by gender of the household head reveal that male-headed households exhibit a higher median resilience score than female-headed households, indicating that they are, on average, better able to withstand and recover from shocks. Across years, the RCI has registered a slight improvement for both male- and female-headed households. Despite the marginal increase, the median RCI values for both male- and female-headed households fall below 50 on a scale of 0–100, highlighting that overall resilience levels remain low in Malawi (Figure A1 in the Appendix). Furthermore, the resilience capacity improved across all regions between 2010 and 2019, but with notable geographic differences, with the Central region consistently lagging behind and recording the slowest growth among the three regions (Figure A2 in the Appendix). Similarly, the improvements in the RCI were unequal across economic status, with richer households experiencing larger increases than poorer households (Figure A3 in the Appendix).
Figure 2 shows the RCI, which measures overall household resilience by combining all RIMA components. The Southern region records the highest resilience capacity (darkest shade), followed by the Northern region (medium shade).
Figure 3 presents the disaggregated RIMA components, revealing highest access to basic services and social safety nets in the Southern region, while adaptability is highest in the Northern region. Shorter distances to basic services such as roads and agricultural markets largely drive higher values of access to basic services as observed in the Southern region. Stronger patterns in social safety nets observed in the Southern region reflect the concentration of the social protection programmes in those locations. Greater resilience capacity with respect to adaptability observed in the Northern region is driven by better historical outcomes in education and engagement on non-farm employment opportunities.
4.1.4 Resilience dynamics
We employ transition matrices to understand how the resilience capacity of households in Malawi has changed over time between 2010 and 2019 (Table 3). This was done in two steps. First, we categorised households into three tercile groups of the Resilience Capacity Index, namely, least resilient, medium resilient, and most resilient. Second, we computed shares of households that remained, moved out of, or moved into the three classes of resilience capacity identified in the first step.
The study finds that 61 per cent of the households classified as least resilient in 2010 remained least resilient in 2019, while 32.62 per cent of the households that were least resilient in 2010 improved to a medium resilience status by 2019. For the households that were moderately resilient in 2010, 53.14 per cent stayed medium resilient in 2019, 23.83 per cent became least resilient in 2019, and 23.03 per cent became most resilient in 2019. Of the households that were the most resilient in 2010, 77.95 per cent remained the most resilient in 2019, 15.22 per cent became medium resilient in 2019, and 6.83 per cent of the households became least resilient in 2019.
4.2 Econometric analysis
This section presents results from an analysis of the determinants of resilience. The estimations are based on fixed effects regression, whose selection was supported by the Hausman test (Hausman 1978). A variance inflation factor analysis was carried out to check for the level of collinearity in the variables, with results well below the most commonly used threshold of 10, above which multi-collinearity is considered worrisome (O’Brien 2007).
4.2.1 Shocks, social protection, and household resilience
Table 4 presents the results of our analysis, where the dependent variable is the RIMA Resilience Capacity Index. The findings reveal that shocks, such as high agricultural input costs, floods, and irregular rainfall, have a negative relationship with household resilience.
Our results further show that engagement in non-farm employment activities, savings, improvements in education,7 and the age of household head are positively associated with household resilience. This suggests that households with these characteristics are likely to be more resilient. For example, as noted in the Malawi 2063 (the country’s long-term development agenda), education has the likelihood of pulling people out of vulnerabilities through the creation of access to opportunities (National Planning Commission 2020). With respect to year, our analysis shows higher resilience capacity in 2016 and 2019 compared to the baseline year (2010).
4.2.2 Determinants of household resilience trajectories
To analyse the determinants of household resilience trajectories, we estimate an ordered probit model using the set of controls discussed in Section 3. The dependent variable, resilience trajectory, consists of three outcomes, namely (1) if households became less resilient, (2) if they maintained the same resilience status, and (3) if they became more resilient. Table 5 presents the results.
The results show that self-employment and attainment of PSLCE and JCE are positively associated with higher resilience trajectories, indicating that households with these characteristics are more likely to become more resilient. These results highlight the importance of economic opportunities in enhancing resilience and underscore the role of education in building adaptive capacity. The Coping Strategy Index (CSI) is also significantly positively associated with resilience, suggesting that targeted support for coping mechanisms can enhance resilience. In contrast, social protection programmes do not show statistically significant effects. The coefficients and marginal effects for insignificant variables are omitted for brevity. These include high agricultural costs, high food prices, illness or injury, floods, drought, irregular rains, landslides, death in the household, cash assistance, food assistance, the Malawi Social Action Fund (MASAF) (a public works programme), tertiary scholarship, secondary school scholarship, household size, and age of household head.
4.2.3 Discussion of results
Our findings reveal positive associations between education, savings, self-employment activities, and household resilience, consistent with some studies from the sub-Saharan Africa region (Crookston et al. 2018; Demisse, Bazezew and Bantigegen 2024; Giwa‑Daramola and James 2023). Households that save tend to have uninterrupted consumption in times of shocks. Further, income diversification is also key in improving resilience. In an agrarian economy such as Malawi’s, participation in non-farm employment is one way of diversifying income sources and cushioning against shocks. Studies show that households that have multiple income sources tend to be resilient (Demisse et al. 2024; Lwanga-Ntale and Owino 2020; Wan et al. 2016). The positive effects of self-employment and education also align with prior studies emphasising the role of human capital and economic opportunities in building resilience (Ali et al. 2023; Demisse et al. 2024; Muriuki, Hudson and Fuad 2024).
The Government of Malawi, in partnership with development partners, is implementing social protection programmes to assist vulnerable households by enhancing their production, productivity, incomes, and engagement in employment opportunities (Government of Malawi 2019b). Furthermore, to mitigate against the high cost of production, the Government of Malawi has been implementing a nationwide farm input subsidy programme aimed at providing smallholder farmers with access to cheaper seed and fertiliser. However, this research shows that while the programmes have, in the short term, helped to cushion households against adverse economic shocks and the high costs of food and agricultural costs, their impact on long-term resilience remains limited. Moreover, other research has shown that the effectiveness of the programmes should be strengthened by addressing targeting issues and other implementation challenges, such as delays in reaching beneficiaries (Gondwe, Nankwenya and Goeb 2023; Nyondo et al. 2022).
While social protection programmes do not show statistically significant associations with resilience, education and engagement in self-employment activities show a positive association with resilience. This highlights an important distinction between short-term absorptive effects of social protection programmes in helping households cushion the immediate impacts of shocks by stabilising consumption and reducing negative coping strategies and long-term resilience capacities driven by education and non-farm employment opportunities. Furthermore, these results seem to support growing evidence that Cash-plus (financial assistance supplemented with targeted services) and integrated livelihood approaches are more likely to generate durable resilience gains than standalone social safety nets such as cash transfers.
While the resilience capacity of households has been improving since 2010, the changes are largely correlated with positive changes in the adaptive component of RIMA, comprising self-employment, literacy, and education. Furthermore, despite improving, resilience generally remains low overall, due to the adverse effects of shocks. Resilience is uneven across regions and wealth status. For example, while some households move upwards in resilience status over time, improvements are less common among households in the poorest wealth quintile. This finding suggests that resilience tends to occur only after households reach a minimum level of assets, corroborating other emerging evidence (Shepherd et al., this IDS Bulletin). Social protection programmes that have been implemented to help households mitigate the shocks have no significant relationship with resilience, reflecting the need for more targeted interventions. Engaging in self-employment activities, having access to savings, and improving education also have a positive relationship on the resilience of households.
The findings highlight valuable lessons for programme interventions. Firstly, the current disconnect between social safety nets and resilience outcomes presents an opportunity to redesign social protection programmes through better targeting, increasing the amount of benefits and programme coverage, and increasing the period of implementation. Secondly, strategies for improvements in education (e.g. enrolment, retention, etc.) could have long-term human capital development benefits that complement household resilience. Thirdly, the susceptibility and vulnerability of the agricultural sector to climatic and weather-related shocks such as floods, drought, and irregular rains suggests the need for continued investments and the promotion of more sustainable farming and water management practices. Finally, policy responses should be differentiated across regions, population subgroups, and wealth status, recognising that resilience constraints vary across Malawi’s agro-ecological and socioeconomic contexts.
1 This issue of the IDS Bulletin was supported by the UK Foreign, Commonwealth & Development Office. The opinions expressed are the authors’ own and do not reflect the views of the funder.
2 This research was made possible by the generous support of the Ministry of Foreign Affairs of Ireland.
3 Anderson Gondwe, Research Fellow, MwAPATA Institute, Malawi.
4 Bonface Nankwenya, Research Analyst, MwAPATA Institute, Malawi.
5 Levison Chiwaula, Director of Research, MwAPATA Institute, Malawi.
6 Joseph Goeb, Senior Research Fellow, MwAPATA Institute, Malawi.
7 PSLCE is Primary School Leaving Certificate of Education, corresponding to eight years of schooling; JCE is Junior Certificate of Education, corresponding to ten years of schooling; and MSCE is Malawi School Certificate of Education, corresponding to 12 years of schooling.
Abay, K.A.; Abay, M.H.; Berhane, G. and Chamberlin, J. (2022) ‘Social Protection and Resilience: The Case of the Productive Safety Net Program in Ethiopia’, Food Policy 112: 102367, DOI: 10.1016/j.foodpol.2022.102367
Adato, M.; Ahmed, A.U. and Lund, F. (2004) Linking Safety Nets, Social Protection, and Poverty Reduction – Directions for Africa 2020, Africa Conference Brief 12, Washington DC: International Food Policy Research Institute
Ali, H.; Menza, M.; Hagos, F. and Haileslassie, A. (2023) ‘Impact of Climate-Smart Agriculture on Households’ Resilience and Vulnerability: An Example from Central Rift Valley, Ethiopia’, Climate Resilience and Sustainability 2.2: e254, DOI: 10.1002/cli2.54
Alinovi, L.; d’Errico, M.; Mane, E. and Romano, D. (2010) ‘Livelihoods Strategies and Household Resilience to Food Insecurity: An Empirical Analysis to Kenya’, paper prepared for the Conference on Promoting Resilience through Social Protection in Sub-Saharan Africa, Dakar, 28–30 June
Barrett, C.B. and Constas, M.A. (2014) ‘Toward a Theory of Resilience for International Development Applications’, Proceedings of the National Academy of Sciences 111.40: 14625–30
Beegle, K.; Galasso, E. and Goldberg, J. (2017) ‘Direct and Indirect Effects of Malawi’s Public Works Program on Food Security’, Journal of Development Economics 128: 1–23, DOI: 10.1016/j.jdeveco.2017.04.004
Chishimba, E.M. and Wilson, P.N. (2021) ‘Resilience to Shocks in Malawian Households’, African Journal of Agricultural and Resource Economics 16.2: 95–111, DOI: 10.22004/ag.econ.333943
Chiwaula, L. and Waibel, H. (2011) Does Seasonal Vulnerability to Poverty Matter? A Case Study from the Hadejia-Nguru Wetlands in Nigeria, Proceedings of the German Development Economics Conference, Berlin, No. 19 (accessed 23 March 2026)
Crookston, B.T.; Gray, B.; Gash, M. and Aleotti, J. (2018) ‘How Do You Know “Resilience” When You See It? Characteristics of Self-Perceived Household Resilience Among Rural Households in Burkina Faso’, Journal of International Development 30.6: 917–33, DOI: 10.1002/jid.3362
Demisse, U.; Bazezew, A. and Bantigegen, S. (2024) ‘Rural Households’ Resilience to the Adverse Impacts of Climate Variability and Food Insecurity in the North-Eastern Highlands of Ethiopia’, Heliyon 10.12: e32960
d’Errico, M. and Di Giuseppe, S. (2018) ‘Resilience Mobility in Uganda: A Dynamic Analysis’, World Development 104: 78–96, DOI: 10.1016/j.worlddev.2017.11.020
d’Errico, M.; Garbero, A.; Letta, M. and Winters, P. (2020) ‘Evaluating Program Impact on Resilience: Evidence from Lesotho’s Child Grants Programme’, Journal of Development Studies 56.12: 2212–34
Diallo, Y. and Tapsoba, R. (2022) Climate Shocks and Domestic Conflicts in Africa, IMF Working Paper 2022/250, Washington DC: International Monetary Fund
Do, M.H. (2023) ‘The Role of Savings and Income Diversification in Households’ Resilience Strategies: Evidence from Rural Vietnam’, Social Indicators Research 168: 353–88, DOI: 10.1007/s11205-023-03141-6
Falco, S.; Kis, A.B. and Viarengo, M. (2022) Cumulative Climate Shocks and Migratory Flows: Evidence from Sub-Saharan Africa, CESifo Working Paper 9582, Munich: Munich Society for the Promotion of Economic Research (accessed 23 March 2026)
FAO (2016) Resilience Index Measurement and Analysis II, Rome: Food and Agricultural Organization of the United Nations
Giwa-Daramola, D. and James, H.S. (2023) ‘COVID19 and Microeconomic Resilience in Sub-Saharan Africa: A Study on Ethiopian and Nigerian Households’, Sustainability 15.9: 1–25, DOI: 10.3390/su15097519
Gondwe, A.; Nankwenya, B. and Goeb, J. (2023) ‘Patterns of Social Safety Nets, Weather Shocks, and Household Food Security Status in Malawi’, Policy Brief 24, Lilongwe: MwAPATA Institute
Government of Malawi (2019a) Malawi 2019 Floods Post Disaster Needs Assessment Report, Lilongwe
Government of Malawi (2019b) National Resilience Strategy (2018–2030): Breaking the Cycle of Food Insecurity in Malawi, Lilongwe: Department of Disaster Management Affairs
Ha, V.H. et al. (2022) ‘Post-Flood Recovery in the Central Coastal Plain of Vietnam: Determinants and Policy Implications’, Asia‑Pacific Journal of Regional Science 6.3: 899–929
Handa, S.; Otchere, F.; Sirma, P. and Evaluation Study Team (2022) ‘More Evidence on the Impact of Government Social Protection in Sub-Saharan Africa: Ghana, Malawi, and Zimbabwe’, Development Policy Review 40.3: e12576
Hausman, J.A. (1978) ‘Specification Tests in Econometrics’, Econometrica 46.6: 1251–71, DOI: 10.2307/1913827
Hoddinott, J.; Berhane, G.; Gilligan, D.O.; Kumar, N. and Seyoum Taffesse, A. (2012) ‘The Impact of Ethiopia’s Productive Safety Net Programme and Related Transfers on Agricultural Productivity’, Journal of African Economies 21.5: 761–86, DOI: 10.1093/jae/ejs023
Lwanga-Ntale, C. and Owino, B.O. (2020) ‘Understanding Vulnerability and Resilience in Somalia’, Jàmbá: Journal of Disaster Risk Studies 12.1: 856
Matata, M.J.; Ngigi, M.W. and Bett, H.K. (2023) ‘Effects of Cash Transfers on Household Resilience to Climate Shocks in the Arid and Semi Arid Counties of Northern Kenya’, Development Studies Research 10.1: 2164031, DOI: 10.1080/21665095.2022.2164031 (accessed 17 April 2026)
McCarthy, N.; Kilic, T.; Brubaker, J.; De La Fuente, A. and Murray, S. (2021) Recurrent Climatic Shocks and Humanitarian Aid: Impacts on Livelihood Outcomes in Malawi, Policy Research Working Paper 9666, Washington DC: World Bank (accessed 23 March 2026)
Muriuki, J.; Hudson, D. and Fuad, S. (2024) ‘Household Resilience and Mitigating Strategies to Conflicts and Shocks: Evidence from Household Data in Uganda and Malawi’, Agrekon 63.1–2: 65–81
National Planning Commission (2020) Malawi 2063: An Inclusively Wealthy and Self-Reliant Nation, Lilongwe
NSO (2020) Fifth Integrated Household Survey, Malawi 2019–2020 (IHS5): Main Report, National Statistical Office, World Bank (accessed 23 March 2026)
Ngoma, H.; Finn, A. and Kabisa, M. (2024) ‘Climate Shocks, Vulnerability, Resilience and Livelihoods in Rural Zambia’, Climate and Development 16.6: 490–501, DOI: 10.1080/17565529.2023.2246031
Nyondo, C.J.; Burke, W.J.; Muyanga, M. and Chadza, W. (2022) ‘Redesigning the Affordable Inputs Program to Diversify and Sustain Growth’, Policy Brief 14, Lilongwe: MwAPATA Institute
O’Brien, R.M. (2007) ‘A Caution Regarding Rules of Thumb for Variance Inflation Factors’, Quality and Quantity 41: 673–90
Otchere, F. and Handa, S. (2022) ‘Building Resilience through Social Protection: Evidence from Malawi’, Journal of Development Studies 58.10: 1958–80
Premand, P. and Stoeffler, Q. (2020) Do Cash Transfers Foster Resilience? Evidence from Rural Niger, Policy Research Working Paper 9473, Washington DC: World Bank
Pruce, K.; Price, R. and Sabates-Wheeler, R. (2025) Learning from Cash Plus: A Summary of Evidence, CPAN Report, Brighton: Chronic Poverty Advisory Network, Institute of Development Studies, DOI: 10.19088/CPAN.2025.005 (accessed 9 April 2026)
Sengupta, S. and Costella, C. (2023) ‘A Framework to Assess the Role of Social Cash Transfers in Building Adaptive Capacity for Climate Resilience’, Journal of Integrative Environmental Sciences 20.1: 2218472, DOI: 10.1080/1943815X.2023.2218472
Stoeffler, Q. and Premand, P. (2021) Cash Transfers, Climatic Shocks and Resilience in the Sahel, paper presented at the International Association of Agricultural Economists (IAAE) Conference 2021, online, 17–31 August, DOI: 10.22004/ag.econ.315354
Ulrichs, M.; Slater, R. and Costella, C. (2019) ‘Building Resilience to Climate Risks through Social Protection: From Individualised Models to Systemic Transformation’, Disasters 43.S3: S368–S387, DOI: 10.1111/disa.12339
Upton, J.; Constenla-Villoslada, S. and Barrett, C.B. (2022) ‘Caveat Utilitor: A Comparative Assessment of Resilience Measurement Approaches’, Journal of Development Economics 157: 102873, DOI: 10.1016/j.jdeveco.2022.102873
UNICEF (2021) ‘Protecting and Transforming Social Protection Spending During and Beyond COVID-19’, Malawi Social Protection Budget Brief 2020/21, Lilongwe: UNICEF Malawi (accessed 23 March 2026)
Vo, H.H.; Mizunoya, T. and Nguyen, C.D. (2021) ‘Determinants of Farmers’ Adaptation Decisions to Climate Change in the Central Coastal Region of Vietnam’, Asia-Pacific Journal of Regional Science 5: 327–49
Wan, J.; Li, R.; Wang, W.; Liu, Z. and Chen, B. (2016) ‘Income Diversification: A Strategy for Rural Region Risk Management’, Sustainability 8.10: 1064, DOI: 10.3390/su8101064
© 2026 The Authors. IDS Bulletin © Institute of Development Studies | DOI: 10.19088/1968-2026.174
This is an Open Access article distributed under the terms of the Creative Commons Attribution 4.0 International licence (CC BY), which permits unrestricted use, distribution, and reproduction in any medium, provided the original authors and source are credited and any modifications or adaptations are indicated.
The IDS Bulletin is published by Institute of Development Studies, Library Road, Brighton, BN1 9RE, UK. This article is part of IDS Bulletin Vol. 57 No. 2 July 2026 ‘Climate Resilience and Poverty Reduction’.