Through the confluent impacts of the Covid-19 pandemic and the 2021 military coup, Myanmar has become a failed state.3 Deliberate targeting of humanitarian actors by the military junta has severely constrained the activities of international non-governmental organisations and United Nations agencies. In the first three post-coup years, welfare provision to the distressed and displaced were mainly undertaken by local actors who adapted to new conditions of both insecurity and broader economic turmoil. This localised welfare is intertwined with emergent local governance mechanisms, using technology and data analysis to deliver transparent, accountable, and inclusive public service. In the absence of a coherent central state, the social contract of welfare develops at more local levels, generating smaller islands of citizenship from which to build future political communities.
Social protection, local governance, emergence, failed state, complex humanitarian emergency, community-led welfare, parahita.
Whilst Myanmar has a long and rich history of both community and government-led welfare (Trager 1958), the term ‘social protection’ began to be used only in the aftermath of the 2008 Cyclone Nargis, the largest natural disaster in Myanmar’s history. Humanitarian clusters led to specific social protection initiatives for women, children, and persons with disabilities (Cheesman, Skidmore and Wilson 2010: 323–48). However, activities remained fragmented and lacked significant support from government until the advent of the National Social Protection Strategic Plan in 2014 (Infante-Villarroel 2015).4 The plan identified a number of flagship programmes including support for pregnant women, older persons, and persons with disabilities. This provided a framework for subsequent rollouts of maternal and child cash transfers, social pensions, and cash support for persons with disabilities,5 operated as a top-down, centrally managed system, and funded largely through foreign aid contributions (Niño-Zarazúa and Tarp 2021).6 By 2020, the programme was estimated to have reached nearly 40 per cent of Myanmar’s population (Maffioli et al. 2023), including Covid-19 pandemic response measures.
However, on 1 February 2021, weeks after securing a landslide electoral win to secure a second term of civilian rule, the National League for Democracy party, led by Aung San Suu Kyi, was again ousted in a military coup.7 As senior leaders were quickly detained, the public responded with widespread demonstrations. A Civil Disobedience Movement resulted in nearly half of civil service personnel from health, education, transportation, and other government ministries refusing to work under military rule, bringing government services to a grinding halt. The military response to both demonstrations and strike action became incrementally violent, including snipers instructed to kill demonstrators with head shots (Human Rights Watch 2021). The rapid escalation of force by the military forces deployed under the junta government (self-styled as the State Administration Council (SAC)), employed the notorious ‘four cuts’ counterinsurgency strategy,8 against mainly civilian targets. This resulted in the expansion into the previously conflict-free central heartlands of a civil war previously confined to border areas. At the time of writing, over 2 million people have been displaced, with over 35,000 reported conflict fatalities. The wider consequences are a rapid increase in poverty levels, and a sprawling, complex humanitarian crisis.
Numerous Myanmar scholars consider Myanmar a failed state9 (Aung Naing and Wells 2023; Boughton et al. 2023). Aside from ever-shrinking territorial control (Lee 2023), the junta administration is unable to deliver public goods and services such as health, education, and security. This includes the ‘complete collapse of social protection in 2021’ where ‘government social protection disappeared… only 5% of households report receiving assistance from any institution, as compared to a 50% poverty rate’ (Boughton et al. 2023: 14). However, state ‘failure’ is not simply lack of capacity but is increasingly characterised by malevolence. The state (in this case, the junta-led government) is not simply incapable of providing social protection but is directly implicated in the creation of the conditions of precarity which then necessitate intervention for social protection:
The rapid and widespread militarisation of the administrative and justice systems has enabled the process of detention, trial, sentencing and, in some cases, extrajudicial killing… which points not simply to a failure to provide such public goods, but a deliberate subversion of them.
(Aung Naing and Wells 2023: 282–3)
At its heart, social protection, defined as ‘public and private initiatives that provide income or consumption transfers to the poor, protect the vulnerable against livelihood risks, and enhance the social status and rights of the marginalised’ (Devereux and Sabates-Wheeler 2004: 9), represents a social contract, largely framed between the state and its citizens (Hickey 2011). If, however, the state becomes incompetent, or even hostile, what then happens to ‘social protection’ (Harvey 2009: 188)? In failed or fragile states where territorial control is contested by multiple actors, the process of social protection may pivot towards a more heterogenous form of state-building, with a focus on enabling more localised expressions of statehood to protect citizens in a sustainable way. This embraces some of what Harvey terms ‘shadow systems alignment’ whereby assistance is delivered in ways which are ‘compatible with existing or future state structures’ (Harvey 2009: 193) but remains aware of the possibility that future iterations of statehood may also be more fragmented.
This article makes three points:
Myanmar’s experience of the two-pronged ‘Coupvid’ (see Lwin, San Wai and Win 2022) crisis represents what scholars term a ‘complex humanitarian emergency’ (Albala-Bertrand 2000), where the interaction between natural, external elements and systemic weaknesses, inequalities, and pre-existing tensions acts to expose the shaky foundations of a society. One consequence of the crisis has been increased difficulty in conducting much-needed research on the fluid and volatile context. However, in late 2020, in the midst of the Covid-19 pandemic, a small volunteer organisation in Myanmar’s second largest city, Mandalay, launched a pilot project, funded by local donations, to provide cash transfers to 100 peri-urban poor households. Local volunteer networks used targeting criteria based on government social protection programmes to identify eligible households who were excluded from government programmes due to lack of official registration documents. Using phone surveys and bespoke apps, they developed a model to enable safe and consistent registration for vulnerable households, whilst integrating simple, but effective post-distribution monitoring to track effectiveness and emerging needs.
The focus of the data collection and analysis was the tracking of trends in five domains: exposure to risks and hazards (Covid-19, forced displacement); food security; livelihood participation; coping strategies (including borrowing, asset liquidation, and reducing food consumption); and assistance (what help they got, from whom). Post-coup, the methodology was adopted by numerous volunteer groups and civil society organisations in other parts of the country. By the end of 2021, a cohort of over 1,000 households in eight locations had been recruited, receiving regular cash or in-kind assistance, with quarterly phone calls to collect tracking data. By 2022, this had grown to over 2,000 households, and to nearly 5,000 by mid-2023, managed by more than 20 local volunteer groups. The majority are located in rural areas with a high incidence of conflict. The author has provided technical support to these groups, who use their own funding sources to provide relief and cover human resource costs. The use of phone surveys, cloud-based applications, and mobile money transfers has enabled a longitudinal study of emergent needs, coping mechanisms, and crucially, the provision of assistance by voluntary groups in conflict areas. This article presents findings largely based on analysis of the quantitative data collected in 2023 through quarterly post-distribution monitoring of 2,822 households.
Analysis from areas with the highest intensity of conflict10 showed that 40–80 per cent of households in different rural cohorts reported being displaced at least once in the previous month as a result of raids by military troops. Displacement was often temporary, but repeated, and results in interrupted livelihood cycles, food insecurity, adverse health events, and negative coping strategies, including crisis borrowing and reducing food consumption.
In July 2023, more than eight in ten households reported that the amount they had to spend on food had increased, and most had to borrow to meet costs. Households in areas with the highest levels of reported violence and conflict were far less likely to report any form of livelihood activity, both a cause and a consequence of a shrinking of commercial and social networks, with the virtual disappearance of once-ubiquitous formal lenders and microfinance institutions.
The exposure to violence and insecurity has had a significant impact on mental wellbeing and social cohesion; one in five respondents from households in areas with higher intensity of conflict reported symptoms consistent with severe anxiety or depression; 50 per cent of households in conflict-affected areas in northwest Myanmar listed psychosocial support as a priority for assistance. Social cohesion was inevitably undermined by the process of displacement, as junta tactics specifically targeted religious and charitable institutions and organisations, which were already under strain by the combined impact of the economic shock of conflict, out-migration of young people either to join militias or to find work, and the increased demand for aid. Added to this has been the weaponising of pre-existing social divisions by both junta and anti-junta forces, resulting in whole family groups known to be military supporters moving from one village to another out of fear of reprisal from the anti-junta militias, and similar relocation of anti-junta families living in military-supporting villages. Despite these challenges, local self-organised aid continues to adapt and flourish.
For those living in conflict areas, and particularly in contexts of repeated displacement, access to timely assistance is both critical and challenging. Whilst discussions on establishing humanitarian corridors continue in international forums, the ground reality represents a multitude of ‘small streams’ of aid, mainly operationalised by local actors. The recent Humanitarian Access SCORE Report (Harvey et al. 2023), drawing data from both conflict and non-conflict-affected areas showed that over a third cited local volunteers as the main source of aid; only 6 per cent reported INGOs, and 5 per cent reported a UN agency. UN agencies and INGOs operated in a highly restrictive context, with the twin threats of actual violence against field staff, and operational sanctions by the SAC against head office personnel and assets, such as denial of visas, travel permits, and freezing of bank accounts.11 As a consequence, INGO and UN presence is relatively limited in the areas of greatest need.
Households in areas of highest conflict intensity (%) |
Households in areas of lowest conflict intensity (%) |
|||
---|---|---|---|---|
July 2023 |
December 2023 |
July 2023 |
December 2023 |
|
Local welfare organisations (parahita) |
12 |
27 |
3 |
5 |
Family/neighbours |
62 |
57 |
57 |
55 |
INGO/local NGO |
6 |
8 |
18 |
14 |
Local government |
0 |
11 |
10 |
4 |
Source: Author’s own.
Data from this study shows households in areas experiencing the highest levels of conflict12 and displacement were more likely than households in lower conflict intensity areas to receive aid (64 per cent vs 51 per cent), and the proportion reporting receiving aid from local welfare organisations increased between July 2023 and December 2023. In the highest intensity conflict areas, there was a greater likelihood of receiving any aid, and of those who did receive some form of aid, the main providers were more likely to be local, either informal (such as family or neighbours), or semi-formal, such as local welfare organisations.
Thus, despite the protracted conflict, economic volatility, the sustained assault by junta forces on both humanitarian actors and localised social, moral, and spiritual capital, localised aid is increasing as a significant source of assistance for households in conflict areas. Local welfare organisations (parahita) have been in existence for decades, despite suppression during military rule. Usually operating locally in relation to a village or urban ward, such organisations tend to occupy a space between village traditions, religious authority, and formal governance structures (Griffiths 2019; McCarthy 2016). Their typical mode of operation involves collecting donations from community members and other donors, as well as operating revolving funds to provide low-interest credit, usually for livelihood activities. This generates funds which are then disbursed as welfare, most frequently to cover funeral expenses, but in more sophisticated operations, grants provide help for older persons, persons with disabilities, and displaced people (Griffiths 2019).
However, the nature of such groups is changing, and their activities are increasingly aligned with emergent self-governance structures, which may in turn be affiliated with the nascent National Unity Government (NUG),13 which was slowly increasing its presence in conflict-affected areas in the northwest of Myanmar. This explains the increase in the proportion of households in conflict-affected areas reporting aid from ‘local government’: in these areas, this represents self-organised administration in opposition to the military junta. An illustrative case of this intersection between emergent relief and wider political movements is described in Box 1.
In an area considered ‘semi-liberated’ in northwest Myanmar, local community welfare associations have self-organised to form a ‘Social Welfare Department’ allied to the emergent People’s Administrative Unit. This department links community volunteers in nearly 100 villages to a broader welfare effort. Volunteers use the aforementioned data tools to identify and recruit vulnerable households, such as those with people with disabilities, for participation in a monthly cash transfer programme, based on parameters used under the previous civilian government. Whilst the People’s Administration is aligned to the National Unity Government, the welfare activities draw strongly on local resources and co-ordination, linking a range of local and external funding sources with village volunteer networks. Many of the volunteers had previous experience with pre-coup parahita organisations, which have continued to adapt and change in form. Beneficiaries are increasingly likely to attribute assistance to the new local government. Significantly, though, it also highlights the way in which emergent welfare is embedded in local political expressions. Volunteers in this programme use the same post-distribution monitoring methodologies as other groups in the area (and indeed in other parts of Myanmar), enabling experience sharing, peer learning, and potential integration with other welfare programmes. Programme activities, including details of welfare measures, are regularly reported on the Facebook page of the Ministry of Home Affairs and Immigration of the National Unity Government. This points to both a transformation of localised welfare, and the potential role of common methodologies to build coherence and synergies for broader networks. This offers potential to avoid the emergence of a myriad of incompatible local systems, and instead develop a degree of methodological coherence for future collaborative iterations.
Source: Author’s own. End of Box 1.
The objective, mode, and framing of these welfare activities align closely with previous national-level social protection initiatives. The difference is that the social contract is developed at a much more local level. The performance of public service uses transparent criteria to distribute welfare, and data collection and reporting to demonstrate accountability. By aligning with nascent/emergent self-governance efforts, this performance of social welfare in turn serves to validate the authority of the new governing body. In the absence of a coherent state, emergent self-governance establishes a form of localised, visible ‘state’ together with those who live there – and the provision of social protection is a key element to that process. Discussions and agreements on modes of identification, redistribution, protection of rights, and the give-and-take of roles, responsibilities, and information are, in themselves, elements of what being a citizen, and being a political community, are all about.
Social protection is commonly viewed as a strategy to strengthen state legitimacy. However, where the state parties claiming to govern are weak, fragmented, failing, illegitimate, or even hostile towards citizens, where should the ‘state-building’ efforts of social protection be directed? Where territorial control is increasingly heterogenous, the provision of social protection also increasingly comes under the purview of multiple non-state actors, which include armed groups. Where standard iterations of social protection are largely directed towards the political legitimacy of the state, in a context of plural regimes, social protection increasingly functions to legitimise the claims of non-state actors.
In Myanmar, at the end of 2023, there are a multitude of emergent welfare organisations and programmes, often linked to either to Ethnic Armed Organisations15 or local People’s Administration Units (in turn often linked either to the NUG or to the various levels of Consultative Councils).16 Writing nearly a decade ago, as national social protection strategies were initially being developed, Yaw Bawm Mangshan highlighted the potential role of social protection in establishing political legitimacy in Myanmar but argues that existing social welfare organisations hold the key to building such trust from the ground up:
Building localized accountability mechanisms is a crucial element of trust-building in contexts where there is still fear of government and widespread lack of access to justice. This is where a system integrating regionalized implementation together with accountability processes, perhaps facilitated by community social organizations, can strengthen the effectiveness of social protection as a trust-building and legitimizing process.
(Mangshan and Griffiths 2018: 74)
However, even in the context of a failed or rogue state, there remains a perspective that emergent welfare is a temporary solution until ‘the state can take over’. This ignores the fact that in many cases, particularly in areas where welfare is more explicitly linked to political power, they are the state, and that their practice is expressing what the states failed to do: build trust, distribute resources, and adequately protect the most vulnerable. Emergent welfare should be rightly seen as expressions of citizenship, whereby people actively belong to an identifiable unit of political community. Rather than presuming that a new national-level state structure will supersede these local expressions, the focus instead should be on building sufficiently common ‘DNA’ so that future state structures can be built from those efforts.
International actors have a significant role to play. At the time of writing, several larger UN agencies still advocate ongoing support for junta-led social protection programmes, despite the inherent contradictions noted earlier, and the limited capacity of junta administration to deliver assistance. An alternative is to recognise and support emergent efforts, directing the state-building objectives of social protection towards multiple sites and actors, rather than a single institution. Far from compromising principles of humanitarian neutrality (Slim 2022), this represents a pragmatic approach not only to enable the effective delivery of much-needed assistance, but to do so in ways which enable the development of inclusive, transparent, and accountable local governance.
Embracing this, the capacity of international actors to strengthen the work of local actors through well-directed funding and appropriate technical support can in turn enable the development of connections between different sites of practice. What this research illustrates is the potential of combining the politics of humanitarian activism with the use of common technical tools, such as data monitoring and post-distribution frameworks, to generate peer-support networks between different local welfare associations. These communities of practice allow the exchange of ideas, tools, and experiences based on a common core of indicators. Rather than insisting on uniform practice, such co-produced knowledge enables the development of locally appropriate but peer-informed solutions. The development of such networks is crucial, given the propensity of localised welfare arrangements to be parochial and inward focused; in the context of Myanmar’s long-standing ethnonationalist tendencies (Buscemi 2022), cross-fertilisation is critical to avoid excessive fragmentation.
In conclusion, this article firstly warns of the hazards of ignoring the political underpinnings of a social protection system. Where a national system is imposed from above, with little meaningful consent from the people, the risks are high of either being hijacked as a tool of populist politics, or of being economically or politically unsustainable. Secondly, the article suggests that the source of political capital necessary for maintaining social protection systems, particularly in failed states, is often to be found in smaller units of local practice. In Myanmar’s case, these are the myriad community-based welfare organisations which have continued to operate under extreme duress, establishing and maintaining small islands of citizenship in the midst of civil war and fragmentation. Thirdly, this article challenges the assumption that there needs to be an established political settlement prior to developing a social protection system. The argument instead is that the development of social protection systems may in itself be an integral part of the development of the political settlement.
The challenge, particularly in failed or fragile states, is to correctly interpret and grasp the political moment. It is not simply that engaging with emergent welfare is a preferable, more ethical approach to enabling social protection, but that the very process of emergent welfare is itself an act of state-building, albeit on multiple, as yet unconnected sites. In the current Myanmar context, social protection may represent a particular space of dialogue and practice which facilitates wider settlements around collaboration, trust-building, inclusion, conflict resolution, and reconciliation with estranged ‘others’.
Instead of viewing emergent, local welfare movements as simply stopgaps or indigenous artefacts, to be discarded once a viable national government re-emerges, this article argues instead for their potential as the sites from where the critical resources for a viable future political settlement may in fact emerge. Resisting the lure of ‘scale-ups’ and ‘rollouts’, universal, one-size-fits-all solutions, and the urge to reimpose prior frameworks based on redundant politics, international actors instead should prioritise funding the ‘imagination’ of new, inclusive, and potentially transformative settlements. Practically speaking, this involves supporting the development of shared methodologies to enable peer learning, multi-level networks, and, in the long run, the basis for a larger political community. If they can embrace the inherent messiness and lumpiness of this kind of community building, these courageous sites of practice may yet provide the basic building blocks for whatever new political communities may emerge from the current crisis.
Akpinarli, N. (2009) The Fragility of the ‘Failed State’ Paradigm: A Different International Law Perception of the Absence of Effective Government, Leiden: Martinus Nijhoff
Albala-Bertrand, J.M. (2000) ‘Complex Emergencies Versus Natural Disasters: An Analytical Comparison of Causes and Effects’, Oxford Development Studies 28.2: 187–204, DOI: 10.1080/713688308 (accessed 5 June 2024)
Aung Naing and Wells, T. (2023) ‘Relief as Resistance: (Re)emergent Humanitarianism in Post-Coup Myanmar’, in A. Ware and M. Skidmore (eds), After the Coup: Myanmar’s Political and Humanitarian Crises, Canberra: ANU Press
Blazevic, I. (2023) ‘Unpacking the UN Humanitarian Aid Chief’s Visit to Myanmar’, The Irrawaddy, 17 August (accessed 12 November 2023)
Bøås, M. and Jennings, K.M. (2007) ‘ “Failed States” and “State Failure”: Threats or Opportunities?’, Globalizations 4.4: 475–85, DOI: 10.1080/14747730701695729 (accessed 5 June 2024)
Boughton, D. et al. (2023) ‘Double Jeopardy: Covid-19, Coup D’état and Poverty in Myanmar’, Applied Economic Perspectives and Policy 45.4: 1998–2016, DOI: 10.1002/aepp.13390 (accessed 5 June 2024)
Buscemi, F. (2022) ‘Weapons and Ethnonational Geographies in the Borderlands: The Case of the Ta’ang Rebel Movements in Myanmar’, in G. Gabusi and R. Neironi (eds), Myanmar After the Coup: Resistance, Resilience and Reinvention, Turin: Torino World Affairs Institute
Cheesman, N.; Skidmore, M. and Wilson, T. (2010) Ruling Myanmar: From Cyclone Nargis to National Elections, Singapore: Institute of Southeast Asian Studies
Devereux, S. and Sabates-Wheeler, R. (2004) Transformative Social Protection, IDS Working Paper 232, Brighton: Institute of Development Studies (accessed 16 May 2024)
Dingli, S. (2013) ‘Is the Failed State Thesis Analytically Useful? The Case of Yemen’, Politics 33.2: 91–100, DOI: 10.1111/j.1467-9256.2012.01453.x (accessed 6 June 2024)
Fishbein, E.; Lusan, N.N. and Vahpual (2012) ‘What is the Myanmar Military’s “Four Cuts” Strategy?’, Al Jazeera, 5 July (accessed 3 November 2023)
Griffiths, M. (2019) Community Welfare Organisations in Rural Myanmar: Precarity and Parahita, Milton: Routledge
Harvey, P. (2009) ‘Social Protection in Fragile States: Lessons Learned’, Promoting Pro-Poor Growth, Paris: Organisation for Economic Co-operation and Development (OECD)
Harvey, P.; Stoddard, A.; Czwarno, M.; Breckenridge, M.-J. and Aung Naing (2023) Humanitarian Access SCORE Report: Myanmar. Survey on the Coverage, Operational Reach, and Effectiveness of Humanitarian Aid, London: Humanitarian Outcomes (accessed 6 June 2024)
Hickey, S. (2011) ‘The Politics of Social Protection: What Do We Get From a “Social Contract” Approach?’, Canadian Journal of Development Studies/Revue canadienne d’études du développement 32.4: 426–38, DOI: 10.1080/02255189.2011.647447 (accessed 6 June 2024)
Human Rights Watch (2021) ‘Myanmar: Protesters Targeted in March Massacre’, 2 December (accessed 4 November 2023)
Infante-Villarroel, M. (2015) Inventory of Social Protection Programs in Myanmar, Washington DC: World Bank (accessed 12 November 2023)
International Commission of Jurists (2021) ‘Myanmar: Military Coup d’État Violates Principles of Rule of Law, International Law and Myanmar’s Constitution’, press release, 8 February (accessed 4 November 2023)
Jolliffe, K. (2014) Ethnic Conflict and Social Services in Myanmar’s Contested Regions, Yangon: Asia Foundation
Lee, R. (2023) ‘Myanmar’s Military Junta Appears to be in Terminal Decline’, The Conversation, 16 November (accessed 28 November 2023)
Lwin, S.; San Wai, K. and Win, N. (2022) ‘Capabilities, Coupvid and Cataclysm: Myanmar’s Responses to Covid-19’, paper presented at the Covid-19 in Southeast Asia, 2020–2022: Restriction, Relief, Recovery Conference, ISEAS – Yusof Ishak Institute, Singapore, 28–29 July (accessed 6 June 2024)
Maffioli, E.M.; Headey, D.; Lambrecht, I.; Oo, T.Z. and Zaw, N.T. (2023) ‘A Prepandemic Nutrition-Sensitive Social Protection Program Has Sustained Benefits for Food Security and Diet Diversity in Myanmar during a Severe Economic Crisis’, Journal of Nutrition 153.4: 1052–62, DOI: 10.1016/j.tjnut.2023.02.009 (accessed 6 June 2024)
Mangshan, Y.B. and Griffiths, M. (2018) ‘Social Protection in Myanmar: A Key Mechanism for Political Legitimacy?’, in J. Chambers et al. (eds), Myanmar Transformed? People, Places and Politics, Singapore: ISEAS – Yusof Ishak Institute
McCarthy, G. (2016) ‘Buddhist Welfare and the Limits of Big “P” Politics in Provincial Myanmar’, in N. Cheesman and N. Farelly (eds), Conflict in Myanmar: War, Politics, Religion, Singapore: ISEAS – Yusof Ishak Institute
Niño-Zarazúa, M. and Tarp, F. (2021) Poverty, Inequality and Social Protection in Myanmar, Development Economics Research Group Working Paper Series 11-2021, Copenhagen: University of Copenhagen (accessed 12 July 2023)
Slim, H. (2022) Humanitarian Resistance: Its Ethical and Operational Importance, Network Paper 87, London: Humanitarian Practice Network, ODI (accessed 8 June 2023)
Spring Rain and Aung Naing (2022) ‘Emergent Identity: “Coupvid”, Volunteerism and the New Citizen in Myanmar’, in G. Houtman and Chayan Vaddhanaphuti (eds), Triple Crisis in Myanmar: Coup, Covid and Climate Change, Chiangmai: Regional Centre for Social Science and Sustainable Development
Trager, F.N. (1958) Building a Welfare State in Burma, 1948–1956, New York NY: Institute of Pacific Relations
Yutwon (2024) Shifting Borders and Shrinking Networks: Precarity in the Post-Coup Era, Civic Monitoring Series, Yangon: Yutwon
Copyright © 2024 The Authors. IDS Bulletin © Institute of Development Studies | DOI: 10.19088/1968-2024.125
This is an Open Access article distributed under the terms of the Creative Commons Attribution Non Commercial 4.0 International licence (CC BY-NC), which permits use, distribution and reproduction in any medium, provided the original authors and source are credited, any modifications or adaptations are indicated, and the work is not used for commercial purposes.
The IDS Bulletin is published by Institute of Development Studies, Library Road, Brighton, BN1 9RE, UK. This article is part of IDS Bulletin Vol. 55 No. 2 October 2024 ‘Reimagining Social Protection’; the Introduction is also recommended reading.