Social Protection & Jobs DISCUSSION PAPER NO. 1802 Human(itarian) Capital? Lessons on Better Connecting Humanitarian Assistance and Social Protection Ugo Gentilini, Sarah Laughton and Clare O’Brien November 2018 Human(itarian) Capital? Lessons on Better Connecting Humanitarian Assistance and Social Protection Ugo Gentilini, Sarah Laughton, and Clare O’Brien November 2018 Abstract. Governments in low- and middle-income countries are increasingly invest- ing in social protection, and also address many of their own people’s “humanitarian” needs themselves. For their international partners, who may have an important role in filling gaps when household needs exceed national capacity to meet them, sup- port for the strengthening of national systems—combined with a shift from short-run to more durable approaches—is becoming a unifying framework for assistance. Some aspects of social protection and humanitarian assistance therefore seem to be on a con- verging trajectory. “Human(itarian) Capital?” discusses findings from 12 country case studies exploring the linkages between humanitarian assistance—in its various interpre- tations—and national social protection systems. Specifically, the paper distills lessons on how humanitarian assistance and social protection systems might better coexist, the possible challenges and trade-offs emerging from practical experiences, and how to facilitate, inform, and accelerate future concerted action. Keywords: social protection systems; social assistance; humanitarian assistance; cash transfers; shocks. JEL codes: P33, P36, I38, H84 C ontents Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . v Part I: Conceptual framework . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 2 Defining humanitarian assistance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 2.1 Actor-oriented definition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 2.2 Purpose-oriented definition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 2.3 Conflict-centered definition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 2.4 A framework for case studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10 PART II: Insights from practical experiences . . . . . . . . . . . . . . . . . . . . . . 13 3 Transfer of international assistance onto national systems . . . . . . . . . . . . . . . . . . . . . . 14 3.1 Liberia: strengthening social protection systems for pandemics and acute crises . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14 3.2 Palestine: a common programmatic framework for social protection and humanitarian response . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15 3.3 Philippines: channeling donor response to natural disasters via national structures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18 3.4 Takeaways from transferring international assistance onto national systems . . . . . 19 4 Improving links between emergency response and longer-term programming . . . . . 20 4.1 Guinea: school feeding programs in a crisis context . . . . . . . . . . . . . . . . . . . . . . . . . . . 20 iii 4.2 Mauritania: crisis response as a catalyst for building social protection . . . . . . . . . . . 21 4.3 Kenya: integrated information management for social protection and drought response . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22 Human(itarian) Capital? 4.4 Mozambique: enhancing connections between emergency response and social protection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24 4.5 Zimbabwe: improving the response to seasonal food insecurity and re-establishing effective social protection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25 4.6 Fiji: sequence of government and donor responses to natural disasters . . . . . . . . . . 27 4.7 Lebanon: supporting refugees alongside the national population . . . . . . . . . . . . . . . 29 4.8 Takeaways from linking emergency response and longer-term programming . . . . . 31 5 Provision of social protection in conflict settings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33 5.1 Syria: protecting lives and human capital during conflict . . . . . . . . . . . . . . . . . . . . . . 33 5.2 Yemen: addressing immediate needs and preserving social protection . . . . . . . . . 34 5.3 Takeaways from conflict settings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36 PART III: Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37 6 Emerging lessons and broader reflections . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43 Figures 1 Share of international humanitarian assistance directed to international and national actors, 2017 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 2 Concepts of bridging humanitarian assistance and social protection . . . . . . . . . . . . . 9 3 Links between humanitarian assistance and social protection in Palestine . . . . . . . . 17 4 Humanitarian–social protection milestones in Mauritania . . . . . . . . . . . . . . . . . . . . . . . 21 5 Adapting social protection to conflict in Yemen . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35 Table 1 Situating case studies within a stylized framework . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11 iv A cknowledgments W e are grateful to Amir Abdulla, Michal Rutkowski, Steen Jorgensen, Anush Bezhanyan, Jehan Arulpragasam, Stanlake Samkange, Colin Bruce, Kenn Crossley, Jon Brause, Ruslan Yemtsov, Denise Brown, Jean-Pierre de Margerie, and Stefano Paternostro for their guidance and support. As authors we warmly acknowledge the insights provided by the case studies and the draft synthesis report which, alongside the broader literature, formed the backbone for this paper. The studies were authored by Aisha Mansur, Christabel Sefa, Fernando Aldaba, Flavia Lorenzon, Haneen Ismail Sayed, Henri Leturque, Jennifer Nyanjui, Khalid Al-Qudsi, Kudzai Chatiza, Louis Rovira, Nicholas Crawford, Pantaleo Creti, Rita Neves, Sandra Uwantege Hart, Shadiyana Begum, Suleiman Namara, and S. Sarsou. These mate- rials are available upon request (ugentilini@worldbank.org, sarah.laughton@wfp.org). We are indebted to many colleagues for their technical advice on the case studies or the theme more widely, including Afrah Alawi Al-Ahmadi, Aline Coudouel, Andrea Vermehren, Asaka Nyangara, Carlo Del Ninno, Colin Andrews, David Kamau, Edmundo Murrugarra, Edouard Nizeyimana, Emma Mistiaen, Eric Zapatero, Giuseppe Zampaglione, Jean-Noël Gentile, Jesse Doyle, Jordi Renart, Julie Macdonald, Maria Vannari, Massimo La Rosa, Nicolas Babu, Niels Balzer, Pablo Ariel Acosta, Peter Holtsberg, Phillippe Leite, Praveen Agrawal, Rene Antonio Leon Solano, Robert Chase, Sarah Coll-Black, Tanuja Rastogi, Thomas Vaughan Bowen, Tina George, and Yukimi Ogaki. The paper is meant for discussion purposes and only reflects the views of the authors. Ugo Gentilini is with the World Bank’s Social Protection and Jobs Global Practice, while Sarah Laughton and Clare O’Brien are with WFP’s Safety Nets and Social Protection Unit. v Part I Conceptual framework 1  Introduction O ver the decades, the central themes of this paper have taken many names: catchy nomenclatures like the “humanitarian-development nexus,” “linking relief and devel- opment,” “development relief,” “doing development differently,” and many others are evolving variants of a core concept.1 Governments in developing countries address many of their own people’s “humanitarian” needs themselves, including through the provision of cash or in-kind transfer programs. But such needs may frequently exceed national capacities to meet them. Throughout history, the international humanitarian architecture has played an important role in fulfilling functions that are normally rendered by gov- ernments. 2 In filling this gap, the interaction between international and national actors can be complex. In part, this epitomizes a broader tension between the “humanitarian imperative,” which transcends borders and serves the individual in his own right, and that of national sovereignty of the state and its capabilities. Recent experiences offer brighter prospects. Indeed, on balance humanitarian assis- tance and social protection seem to be on a partly converging trajectory. As this paper illustrates, building national systems—combined with a shift from short-run to more durable approaches—is increasingly becoming a unifying framework for assistance pro- vided by different actors. To some extent, however, a distinctive feature of humanitarian assistance revolves around the degree of involvement and engagement of national authorities (Konyndyk 2018; Stoddard 2017). For instance, it is estimated that out of the $27.2 billion of international humanitarian assistance in 2017, only 2.5 percent was channeled through host governments (figure 1). The rest was directed outside govern- ment structures, including via some 4,480 actors comprising international and national nongovernmental organizations (NGOs), civil society, and others. Meanwhile, a number of national governments are, of course, also allocating domestic resources to address 2 1  For early contributions, see for example the seminal work by Buchanan-Smith and Maxwell (1994), as well as the discussion in TANGO (2004) and Harmer and Macrae (2004). 2  See Bennett et al. (2016). 1 Introduction Figure 1  Share of international humanitarian assistance directed to international and national actors, 2017 Southern international RCRC national societies – 0.03% 96+3+1F NGOs – 0.3% Local NGOs – 0.04% National NGOs – 0.4% Local and International national responders – 97% responders – 2.9% National governments – 2.5% Internationally affiliated NGOs – 0.08% National foundations – 0.004% National research institutions – 0.006% National private sector corporations – 0.004% Source: Adapted from Development Initiatives (2018). emergencies within their own borders and without recourse to the international com- munity, e.g., Mexico (Beazley et al. 2016). 3 For international humanitarian actors, the model of operating outside of or with limited engagement with government systems may be dictated by several constraints, includ- ing because of the risky operating environment and pressure for ensuring life-saving goals; possible lack of government sovereignty over a territory in full or part; legisla- tion preventing some types of domestic assistance to refugees; or concerns about the impartiality of governments in conflicts. These are no rare circumstances among cur- rent crisis-affected countries. Operating through parallel systems may also present an element of institutional expediency, which may enable international aid agencies to maintain more direct contact with beneficiaries. Efforts by humanitarian actors to increase national ownership as well as the political and economic sustainability of state-provided services are aiming to resolve these constraints. Our case studies pro- vide concrete examples of such efforts. In contrast, social assistance or safety net programs are operated by and through gov- ernments. By “social assistance” or “safety nets” we refer to noncontributory transfers to poor or vulnerable households or individuals. These may be in cash (e.g., child grants, social pensions) or in kind (e.g., school meals). The term also includes entitlements to reduced expenditure for poor or vulnerable households, such as targeted subsidies or It is noteworthy that data on national allocations are not necessarily compiled in the same 3  manner as humanitarian assistance, hence often not appearing in global reporting. 3 Human(itarian) Capital? fee waivers for essential social services. These programs are components of broader permanent social protection systems, with the latter also including social insurance (i.e., contributory transfers), various labor market interventions, and select social services. Social protection is provided regularly and on a multiyear basis; it is often enshrined in legislation, integrated in sectoral policies, financed from domestic budgets, and represents a cornerstone in government-citizens’ social contracts. In addition, social protection programs may be planned and coordinated via interministerial committees, while international humanitarian assistance often rests on the United Nations (UN) cluster system, which may not systematically involve national authorities. Operating through governments entails engaging in the political economy of redistribution, and this can be especially challenging among countries with a long history of sectarian and ethnic violence. Given these differences, it can be challenging to reconcile the international nature of much humanitarian assistance with the demand for increased national ownership— and short-term planning horizons with the desire for longer-term, more sustainable approaches. However, emerging experiences show that, for instance, many fragile states display high levels of ownership despite social protection systems largely resting on external financing (e.g., Ethiopia and Palestine), including often by the same donors that finance humanitarian operations. For example, in 17 Sub-Saharan African countries, over half of spending in social assistance is donor funded (Beegle et al. 2018).4 Several factors—both political and operational—suggest that there is currently an appetite among governments and their partners for exploring and resolving these chal- lenges. The literature on this question is extensive (Ulrichs and Sabates-Wheeler 2018; Winder-Rossi et al. 2017; EC 2017; Kukrety 2016; World Bank 2016; Bastagli 2014; Cherrier 2014; Kuriakose et al. 2013; McCord 2013; Davies et al. 2009; Harvey 2009). The core rationales and opportunities include the following: „„ There is growing collaboration in crisis-affected countries. The long-term plan- ning of social protection provides a degree of insurance against shocks; but, as mentioned, these often exceed the capacity of countries to manage them—hence requiring an additional line of international support. The case studies point to an increasing practice of dialogue and collaboration among both national and inter- national actors. Working together through crisis responses and in non-emergency settings should facilitate better outcomes over time. „„ Crises are often protracted. About half of the countries submitting a UN appeal in 2016 had done so every year over the past decade (Development Initiatives 2018). 4  In a similar, although less extensive way, social protection programs often make extensive use of “project implementation units” (PIUs): often proposed by donors, these are structures 4 are formally placed within governments to provide a direct channel for program reporting and management to external partners. As such, PIUs may enhance project accountability, but also introduce an element of a “parallel system” that potentially duplicates a mechanism—even an imperfect one—the country may already have in place. 1 Introduction „„ Acting early saves costs and human capital. Longer term investments may reduce the cost of response when a crisis arises. For example, it is estimated that every $1 spent on social protection and resilience programming in Ethiopia, Kenya and Somalia, when combined with early intervention around the time of the crisis, results in net benefits (savings) in the order of $2.30–$3.30 compared to an international humanitarian response delivered after a crisis has occurred (Cabot Venton 2018). In a similar vein, the title of the paper, “Human(itarian) Capital?,” taps into the emerg- ing narrative on human capital. This posits that early investments in preserving and enhancing people’s human capital (e.g., food security, nutrition, health and educa- tion) are central to development. 5 Fulfilling those functions during crises thereby becomes a cogent priority for countries’ future potential. „„ Reducing duplication of efforts. In Lebanon, for instance, a 2014 review showed that more than 30 different aid agencies managed cash transfers and vouchers for 14 different objectives, ranging from food to legal assistance (CGD and ODI 2015). „„ Changing modalities and technologies, including a wider adoption of cash-based transfers for emergency response. While food assistance has historically played an important role in crises (Alderman et al. 2017; Gentilini 2016a), the widespread use of vouchers and cash transfers for international and national emergency responses, often delivered digitally, is generating more incentives for integrating humanitarian and government-led social protection delivery systems. Estimates show that cash transfers in humanitarian settings are estimated at $2.8 billion, a 40 percent increase from 2015 (CaLP 2018).6 „„ “Meeting halfway.” Wider engagement and investment flexibility are observed in crisis-affected countries on all sides. Development agencies are expressing greater willingness to engage in crisis-affected countries, as exemplified by a new commit- ment in this regard from the World Bank, including for increased grant financing and expanded concessionary lending (including in middle-income countries). This could translate into more sustained support for national social protection systems, even in conflict settings. Meanwhile, international humanitarian actors are increasingly sup- porting investment in social protection. But how would enhanced interactions look in practice? This note summarizes main findings from 12 country case studies exploring the linkages between humanitar- ian assistance—in its various interpretations—and national social protection systems (Aldaba 2017; Al-Qudsi 2017; Chatiza 2017; Creti 2017; Ismail Sayed 2017; Leturque 2017; Lorenzon 2017; Mansur et al. 2017; Namara and Sefa 2017; Neves 2017; Nyanjui 2017; Sarsou 2017). The studies point to the conditions under which humanitarian 5  See http://www.worldbank.org/en/publication/human-capital 5 6  According to an independent evaluation, cash transfers are indeed one of the areas where humanitarian community has achieved most progress among Grand Bargain commitments (Metcalfe-Hough and Poole 2018). Human(itarian) Capital? assistance, with support from international humanitarian agencies such as the World Food Programme (WFP), and national social protection systems, with support from the World Bank, have coexisted and converged, how they might or might not reinforce each other, and their possible limits and trade-offs. The studies do not strongly focus on national disaster management in emergency preparedness and response, though these are, of course, central to any consideration of effective crisis response. Nor is this work about the World Bank and WFP per se, with findings and discussion being meant to illustrate broader points transcending those institutions. The purpose of this paper, which draws on global literature, Crawford (2017) as well as each case study report, is to clarify the ways in which humanitarian assistance and social protection programmes might better coexist, the circumstances under which conver- gence might be appropriate, and successes and challenges in doing so. The remainder of the report is structured as follows: the next section sets out core concepts, definitions, and a stylized framework situating the individual country experiences. These constitute Part I of the paper. Part II encompasses sections 3–5, which discuss main country find- ings complemented by a summary of key “take-aways.” The concluding section 6, which is Part III of the paper, distills overall lessons and reflects on the social protection– humanitarian agenda moving forward. 6 2  Defining humanitarian assistance T he concept of “humanitarian assistance,” and its potential connections to social protection, is subject to interpretation (O’Brien, forthcoming; Roelen et al. 2018; Gentilini 2016b). The Principles of Good Humanitarian Donorship refer to assistance that is provided to, “…save lives, alleviate suffering and maintain human dignity during and after man-made crises and disasters caused by natural hazards, disasters, as well as to prevent and strengthen preparedness for the occurrence of such situations.” Three contrasting interpretations of this concept stand out: one that emphasizes the inter- national or national nature of the actor, one based on the purpose and a subset of the two, which often emerges from situations of conflict (O’Brien et al. 2018). This varia- tion reflects the multisectoral nature of the concept and is not inherently problematic; but it is useful to be aware of the multiple ways in which terms are utilized. The next subsections review those interpretations to elucidate some key perspectives. These are followed by a discussion on a set of emerging cross-cutting issues. 2.1 Actor-oriented definition In a number of instances, humanitarian assistance is synonymous with support funded by the international community in response to disasters classified as requiring a human- itarian response. These funds are classified as Official Development Assistance. For the purposes of setting a convenient boundary between humanitarian assistance, in this sense, and development assistance, the former is commonly taken to mean the expen- diture that is reported on the Financial Tracking Service (FTS) of the UN Office for the Coordination of Humanitarian Affairs. This comprises the expenditure of national donor governments (to countries other than themselves), the European Civil Protection and Humanitarian Aid Operations (ECHO), UN agencies, international NGOs, civil society organizations, the International Red Cross and Red Crescent movement, multi- lateral pooled funds and private donors. This information is reported voluntarily so 7 the data may be incomplete. The crucial point for the purposes of this paper is that Human(itarian) Capital? the FTS—and therefore this first definition of humanitarian assistance—specifically excludes “a government’s expenditure on crises within its own borders.” For those adopting this interpretation, the important policy challenge is the one expressed in the Grand Bargain, namely to provide 25 percent of international human- itarian funding to local and national responders by 2020 (United Nations 2016). Social protection might offer one such channel, though of course not the only one: funds might also be directed, for example, through a national disaster management authority or a ministry of health or education. Another challenge might be to explore the feasibil- ity of sharing delivery platforms between international humanitarian actors and national social protection programs where they are distributing assistance that has similar fea- tures, such as cash transfer schemes. 2.2 Purpose-oriented definition A different interpretation of humanitarian assistance posits that the key question is not the origin of the support, but what it is intended for, i.e., efforts to prevent and/ or address disasters. Under this definition, which is also widely adopted by the interna- tional humanitarian community but not confined to it, humanitarian assistance is more or less synonymous with emergency response. The school of thought comfortable with this definition is open to the idea that national governments can therefore deliver humanitarian assistance themselves. In using this interpretation, the important policy challenge is about finding better ways of assisting households who are affected by short-term or protracted crises. For instance, in many countries, “emergency” assistance is delivered on a short-term basis, and repeatedly, to households who are in fact chronically poor, and who might benefit more from sustained and predictable assistance of the sort delivered by social protec- tion programs. The agenda of “linking humanitarian assistance to social protection” may be about finding a way to transition this caseload from short-term to more system- atic assistance. Alternatively, for households who are affected by a disaster, it may be about exploring whether social protection programs offer a better vehicle for deliver- ing assistance during a response than alternatives (which may or may not be the case), or whether those households should be integrated into a social protection system after their emergency assistance ends. It may also refer to opportunities for either the social protection sector or the disaster risk management sector to integrate systems and pro- cedures from the other sector, or to learn from them. These definitions are illustrated in figure 2: the horizontal arrow, “interpretation no. 1,” shows the concept of the definition where “bridging humanitarian assistance and social protection” refers to efforts to move from internationally to domestically led support. The vertical arrow, “interpretation no. 2,” shows the concept of purpose-based defi- 8 nitions, i.e., where the term refers to efforts to move from short-term or emergency responses—be they nationally or internationally led—to more longer term, sustainable 2  Defining humanitarian assistance Figure 2  Concepts of bridging humanitarian assistance and social protection INTERNATIONAL NATIONAL DEVELOPMENT Longer term/ DEVELOPMENT PROGRAMMING development ASSISTANCE (INCLUDING SOCIAL PROTECTION) DURATION/SUSTAINABILITY 2 Interpretation no. 2 1& Interpretation no. 2 s. no n tio ta re rp te In Interpretation no. 1 Short term/ INTERNATIONAL NATIONAL emergency HUMANITARIAN EMERGENCY ASSISTANCE RESPONSE Internationally led Nationally led LEADERSHIP Source: O’Brien (forthcoming). programming. Many interventions are aiming to move in both directions at once (the diagonal arrow). 2.3 Conflict-centered definition According to a third vision, humanitarian assistance refers to assistance provided in sit- uations of armed conflict where international humanitarian law applies. The objective may be the same as the one described above—to save lives and alleviate suffer- ing—and the assistance is likely to be funded by the international community, so this interpretation is a subset of interpretations no. 1 and 2, but more narrowly focused on conflict. It was in these situations that the humanitarian principles emerged. They pro- mote the ideals of humanity, impartiality, neutrality and independence as being central to the delivery of assistance. The confusion between this interpretation and no. 2 above has led to a misunderstanding in some quarters that the humanitarian principles must apply in all situations of crisis, and that therefore governments ought not to be respon- sible for supporting their own populations to save lives in times of natural shocks because that would not be “independent.” Evidently this is not the case: the role of states is clearly recognized in law, and most statements of principle start with a reaffir- 9 mation of the primary responsibility of states for the welfare of victims of humanitarian emergencies within their own borders (O’Brien et al. 2018). Human(itarian) Capital? In these contexts, “linking humanitarian assistance to social protection” might entail considering how, for example, the systems developed by international aid agencies to support households in a crisis context can provide inspiration for, or be handed over to, longer-term support for the poor and vulnerable once the crisis is over. However, blurred definitions and models emerge also beyond conflict situations, such as in Ethiopia and Kenya. These may include the different ways in which govern- ment-led needs assessment interface with response planning, e.g., unified response plan implemented by various actors (Ethiopia), or individual agency-led initiatives (U.K. Department for International Development [DFID] financing of scaled-up safety net transfers under the Hunger Safety Net Program in Kenya). It may also include blends of management of food distributions (different roles of governments, agencies and NGOs in functions such as monitoring). In Kenya, there are ad hoc emergency food distribu- tions financed and administered by devolved county governments. Furthermore, recent cash transfer pilots in Ethiopia financed by ECHO and supported by WFP channel fund- ing through the government (Sandford 2018; World Bank 2018). 2.4 A framework for case studies Taken together, the three definitions laid out in previous sections form an organizing framework for the 12 case studies. The first interpretation, that refers to the objective of transferring international assistance onto national systems, is reflected in the practi- cal experiences documented in case such as Liberia, Palestine, and the Philippines. The purpose-based notion of humanitarian assistance, geared to improving the links between emergency response and long-term programming, is particularly relevant to Lebanon, Guinea, Kenya, Mauritania, Zimbabwe, Fiji, and Mozambique. Finally, the third objective speaks to the provision of social protection in conflict settings, like in Yemen and Syria. We subsequently plot these observed functions along stylized levels of maturity of social protection systems (table 1). While the quantification of social protection matu- rity is a complex and multifaceted analytical question, we simply refer to three broad buckets drawing from other typologies of national social protection capacities (World Bank 2018; Gentilini 2016b; OPM 2015). Clearly, the typology is illustrative and there are multiple dimensions to consider when locating countries in the framework, with some countries that could potentially fit more than one model. Broad clusters may include the following: „„ Countries with limited national social protection systems and large-scale inter- national humanitarian assistance. These would include Syria and Yemen, which are amid complex emergencies and conflict. Yemen may be considered a case where national social protection systems are limited “temporarily,” since they are still lev- eraged for implementation. Liberia and Guinea are low capacity countries in which 10 a large-scale acute crisis—in this case, the Ebola epidemic—can quickly overwhelm 2  Defining humanitarian assistance national structures. Mauritania is an interesting example of a country with low levels of initial capacities, which were enhanced building on crisis response. „„ Contexts with emerging national social protection systems. Palestine and Lebanon are facing protracted displacement crises, punctuated by regular political crises and occasional violent conflicts. Both have relatively high national capacities and strong donor support. Other case studies refer to governments in relatively stable settings Table 1  Situating case studies within a stylized framework Significant national Limited national social Emerging national social social protection Interpretation protection systems protection systems systems Liberia (Ebola response Palestine (largely donor- Philippines (use of social funding and coordination funded assistance, but protection infrastruc- 1: Transfer of between development delivered based on ture as a vehicle for international agencies, international common programmatic international emergency assistance humanitarian actors and framework between response) onto national government in an acute development agencies, systems crisis) international humanitar- ian actors and national authorities) Guinea (introduction and Kenya (use of connected Fiji (sequence and use of effects of a social protec- administrative infrastruc- social protection system tion program in a crisis ture to inform social as a vehicle for national context) protection and emer- and international emer- gency response) gency response) Mauritania (from small-scale pilots to pro- Mozambique (poten- gressive scale up and tial for improving social institutionalization of protection programs to 2: Improving social protection) enhance disaster risk man- links between agement and resilience to emergency shocks) response and Zimbabwe (improving the longer-term response to seasonal food programming insecurity while re-estab- lishing social protection systems) Lebanon (convergence in elements of program design and implemen- tation for the national population and for refugees) Yemen (humanitar- ian and development actors leveraging parts 3: Provision of pre-existing social pro- of services tection institutions and in conflict practices) settings 11 Syria (international humanitarian assistance as foundation for future social protection system) Human(itarian) Capital? addressing frequent climate-related crises. We look at the interface of social protec- tion and humanitarian assistance to address seasonal food insecurity and drought in Zimbabwe and Kenya, and to address rapid-onset shocks in Mozambique. These countries vary widely in the volume and duration of international humanitarian assistance that they receive. Palestine, for example, was the world’s fourth largest recipient of international humanitarian assistance by volume in 2016, at over $1 bil- lion, while other countries such as Mozambique issue only occasional humanitarian appeals and on a much smaller scale. „„ Settings with significant national social protection systems. This includes coun- tries like the Philippines and Fiji, which represent some of the international cases that best lend themselves to alignment between international humanitarian assis- tance and existing national social protection structures. Overall, case studies were selected based on an initial scoping exercise based on World Bank and WFP analysis of practical cases of humanitarian–social protection interactions. The selection should not be considered exhaustive, and a range of countries has since emerged offering further insights (Madagascar is cited in the paper). Ethiopia has been subject to extensive documentation and is also referred to on specific occasions. Future research in a number of other “live” case studies, e.g., Central African Republic, Chad, Mali, Sudan, Northern Uganda, and the Rohingya response in Bangladesh, would also be enlightening. 12 PART II Insights from practical experiences 3  Transfer of international assistance onto national systems 3.1 Liberia: strengthening social protection systems for pandemics and acute crises Liberia suffered severe socioeconomic impacts from the Ebola crisis, including around food access and availability among affected populations. The World Bank’s regional Ebola Emergency Response Project (EERP), with some $167 million eventually allocated to Liberia, aimed to contribute to controlling the outbreak, making health services avail- able and mitigating the socioeconomic impact. WFP’s operational experience and reach in Liberia—and the fact that WFP’s Emergency Operation objectives overlapped with those of the Bank—made it a sensible Bank partner under two of the project’s five com- ponents, by which WFP delivered food assistance to meet the caloric and micronutrient requirements of affected households.1 Six key findings arise from the Liberia case study. „„ Funding arrangements. Usually the World Bank aims to transfer funds through sov- ereign governments to manage project implementation where possible. Under the EERP, the World Bank transferred WFP’s first tranche directly to WFP with the authorization of government, but without going via government systems. This was agreed because of the need for a speedy response and because government capac- ity was stretched to its limits. For the second tranche the Bank reverted to its traditional modality, moving funds through the government to WFP. Delays in the amendment of the agreement meant that WFP had to prefinance the whole second tranche. It is expected that, as government systems become stronger, the neces- sity of channeling funds directly to international agencies will be reduced. However, 1  Under the arrangement, the Government authorized WFP to receive part of the EERP funds, delivered in two tranches. The first tranche of $5.6 million was spent on the purchase of food commodities. The second, for $2.4 million, had been intended for the purchase of food; but the 14 crisis context was changing rapidly and it was decided to enable households to pay for labor so they could invest in the planting season and stimulate recovery of local markets, meeting food security needs through locally produced food. The grant agreement was revised to allow for the distribution of a cash transfer. 3  Transfer of international assistance onto national systems in the absence of alternative national systems, in some emergency contexts direct transfers of funds to international agencies may be appropriate. „„ Flexibility in grant agreements also played an important role. The original terms of the second tranche agreement referred only to the delivery of food, not cash. The need to amend project documents and financial agreements, once it was decided that cash was preferable, took time and resources. Greater flexibility in grant agree- ments for emergency contexts, compared with those that are used in regular development programs in stable conditions, may prove valuable for reducing delays. „„ Timeliness. The channeling of funds to WFP, an organization that has standard pro- cedures to enable rapid emergency response, was beneficial in that funds were quickly mobilized for use. „„ Procedures of humanitarian operations by external agencies may not necessarily be aligned with national systems of reporting, procurement, etc. This could inhibit the process of connecting to national social protection systems as they develop or resume in functionality. The Liberia case calls for more standardized and planned ways to devise bilateral or tripartite arrangements, a process that since has gathered steam and has been rolled out by the World Bank and UN agencies. „„ Leadership also mattered. The collaboration on the Ebola response was steered by the commitment of individuals in the Government of Liberia, World Bank and WFP. It could be valuable to take time now, in a noncrisis situation, to elaborate an opera- tional framework to guide triggers for future collaboration, so that opportunities are less dependent on the decisions made by individuals. „„ There are opportunities for knowledge transfer between agencies and government. As the acute emergency transitioned to more sustained development support for survivors or affected populations, there were opportunities to increase knowledge sharing and the transfer of experiences. For example, WFP undertook many vulnera- bility assessments in counties supported by the government’s cash transfer program and household-level information collected might be relevant for both types of inter- ventions. Drawing on this experience, discussions are already underway to link the development of a national social registry of vulnerable households with data gath- ered by WFP for its proprietary beneficiary registration system. 3.2 Palestine: a common programmatic framework for social protection and humanitarian response The cornerstone of the national social protection system in Palestine is the Cash Transfer Program (CTP). This embodies how collaboration among national governments, 15 development partners and international aid agencies can result not only in streamlined delivery systems for all partners, but also in greater trust and strengthened national Human(itarian) Capital? ownership of the provision of social assistance. It also reveals the trade-offs and some of the challenges to be addressed. The positive story of the national ownership is par- ticularly notable given the CTP’s uncertain fiscal sustainability and reliance on donor funding. The European Union (EU) and the World Bank finance nearly 50 percent of its cost, while the remainder is financed by the Palestinian Authority. However, the latter also receives substantial budget support, part of which covers the CTP. The CTP was launched at the time of the formulation of Palestine’s Social Protection Sector Strategy in 2010, under the leadership of the Ministry of Social Development with World Bank assistance. It was the successor to two earlier cash transfer pro- grams, the EU-funded Social Hardship Case and the World Bank–funded Social Safety Net Reform Project. The CTP merged the two in terms of program management and implementation, and uniform payments; it also built on the capacity-building activi- ties, targeting mechanism and a management information system (MIS) from the earlier interventions. Under the CTP the targeting mechanism and MIS were upgraded: the tar- geting instrument was shifted to poverty-based criteria (proxy means testing) and the MIS was developed to encompass a targeting database, an accounting system, payment lists, and a module for reporting. Around the same time, in 2009, WFP had introduced a voucher program for food assis- tance as a life-saving measure to respond to high food prices (Galluzzi and Natsheh 2010). This was intended to improve food security as well as to have positive second- ary impacts on the local economy by restricting some of the foods redeemable on the voucher to those produced locally. WFP then made significant efforts to build the capacity of the Ministry of Social Development to introduce the voucher scheme as a complement to the CTP, to enhance the effectiveness of the CTP and to support the ministry to manage the use of vouchers as an emergency response. The close partnership with the ministry has successfully built a sense of ownership by the Palestinian Authority over the assistance mechanisms, with the trade-off accepted as being a slower rate of expansion than if WFP had implemented the modality on its own. As of July 2018, the voucher reaches over 75,000 CTP beneficiaries, supplementing the CTP. Meanwhile, the electronic system that is used to manage the financial transactions under the voucher modality was designed to be easily adaptable for other crises. Indeed, it was made possible for other organizations to deliver assistance through the same card, including assistance from the United Nations Relief and Works Agency, UNICEF, and HelpAge for providing WASH items and school uniforms (Gentilini 2015). In all these arrangements we find numerous mutually beneficial synergies between development interventions and those for crisis response. Figure 3 highlights some of the links between the CTP, the food voucher program, other interventions and the sector more broadly. For example, both the World Bank and WFP worked with Palestinian authorities through a common registry system (the poverty-targeting database); they 16 also share a targeting framework and include some of the same beneficiaries. Some ele- ments of program design, such as support to national capacity building, are intended to 3  Transfer of international assistance onto national systems Figure 3  Links between humanitarian assistance and social protection in Palestine CTP poverty-targeting database used by other agencies to provide complementary assistance or avoid duplication CTP targeting formula (PMT) used CASH by WFP for non-CTP interventions OTHER TRANSFER FOOD DIRECT LINKS Food consumption questions Card with EMERGENCY PROGRAMS BETWEEN PROGRAM added to PMT formula VOUCHER e-wallet made RESPONSES Implemented Implemented by available to other Some CTP beneficiaries Including by by MoSD with MoSD with WFP agencies enrolled in voucher scheme UNRWA, UNICEF, World Bank assistance HelpAge, etc. assistance Voucher scheme partly integrated into CTP OUTCOMES/SYSTEM STRENGTHENING INDIRECT LINKS Support to ministry strengthens Voucher restricted to TO IMPROVED national capacity for analysis and purchase of locally decision making on food security, and produced food where management of social assistance possible, to improve livelihoods of producers Contribution of food voucher Contribution of CTP Note: The upper half of the diagram shows the links that have been developed between spe- cific programs—the CTP, the food voucher program and other responses to crises. The lower half shows some of the links between those programs and the broader social protection system, including activities that reduce vulnerability or promote general capacity building in social protection. have longer term positive impacts on the social protection sector as a whole. Meanwhile some potential areas for closer integration remain underexplored: for instance, at the time of the study, while international humanitarian agencies could request the MoSD to cross-check the names of their beneficiaries against the CTP’s beneficiary list, they could not feed information that they have collected back into the CTP system. More recently, there have been efforts to address this issue. The experience of the CTP also offers lessons for understanding the potential con- sequences of forging links between programs that may have different objectives or trajectories. For example, reliance on a single targeting tool may render the CTP and other programs less effective in addressing vulnerability emerging from factors not incorporated in the proxy means test formula. Hence it will be important to supplement those programs with mechanisms to protect these groups before they become further excluded. Overall, there are strong indications that the two types of interventions—govern- ment-led social protection schemes and the emergency-style responses of international aid agencies—can learn from one another’s experiences, and even to develop common practices or systems where their objectives share similarities. The likelihood of this occurring and having a positive impact depends not only on the relevance of their respective programs to one another, but also on issues ranging from political econ- 17 omy—such as international actors holding on to their space, or the preference of some donors to work through international agencies rather than the government—to the sus- tainability of funding for national systems and the capacity of state actors. Human(itarian) Capital? 3.3 Philippines: channeling donor response to natural disasters via national structures The Department of Social Welfare and Development (DSWD) in the Philippines man- ages a flagship conditional cash transfer program, the Pantawid Pamilya Pilipino Program (4Ps) reaching about 4.5 million households. After Typhoon Haiyan in November 2013, both WFP and UNICEF funded top-up payments to some beneficia- ries of the 4Ps to help meet their immediate needs and facilitate recovery. The intention was to work through government systems to provide postdisaster response to reduce the reliance on contracting NGOs to work in parallel. WFP funded top-ups of P2,600 (about $58) to nearly 100,000 beneficiary households, delivered in two tranches in January and February 2014. UNICEF, in agreement with the DSWD, later funded extra cash support to some 5,800 households in the worst affected province. Each received P4,400 (about $100) a month for six months, from July to September 2014, and January to March 2015. The DSWD’s delivery systems and resources were central to the response. Its staff managed the delivery of this assistance, in part by working overtime. WFP and UNICEF channeled their funds through the same payment mechanisms as the 4Ps—after having conducted due diligence on their ability to deliver the additional support—making use of DSWD’s partnership with the Land Bank of the Philippines and its network of sub- contracted payment service providers such as rural banks and post offices. Monitoring and evaluation of the assistance, however, remained independent. A primary observation must be, of course, that these top-ups were, by definition, con- fined to households already enrolled on the 4Ps, i.e., poor households with children. Use of the 4Ps infrastructure meant that they were familiar with the delivery pro- cess. Other households living in the same area who were not on the program had to be reached through separate interventions. For the funders—WFP and UNICEF—the arrangement was cost-efficient as they assessed it to be cheaper for the agencies themselves, per dollar transferred, to use the DSWD’s delivery channels in comparison to setting up agreements with NGOs to deliver the cash. The comparison is possible because both agencies also contracted NGOs to provide assistance to households not enrolled on the 4Ps. WFP, in its agree- ment with the DSWD, covered the government’s additional implementation costs, though constraints in procurement and hiring meant that the DSWD was unable to avail itself of these funds at the time they were required. The establishment of the memorandums of understanding was made easier by the fact that WFP was already a partner of the ministry. 2 UNICEF, too, had long been a partner. Now that the memorandums of understanding exist, they can be quickly modified and 18 2  WFP was a member of an advisory committee supporting the DSWD and they had collabo- rated on delivering relief in disaster and conflict areas even before typhoon Haiyan. 3  Transfer of international assistance onto national systems adopted in the event of future emergencies (as has been done already, in the response to Typhoon Ruby in 2014). The burden was heavy for the DSWD’s own staff, who already had the task of revali- dating lists of 4Ps beneficiaries in affected areas and issuing temporary identification cards for those who had lost them. They then had to oversee the arrangements for the extra payment, that is, one of WFP’s two top-ups was paid at a separate time to the reg- ular schedule of payments for 4Ps beneficiaries. The disruption of the UNICEF-funded payments between October and December 2014 was ascribed to the overburdening of DSWD staff who were responsible for the financial management of the 4Ps while han- dling other projects. Some payment service providers witnessed their system being affected by the typhoon. Allowing beneficiaries the flexibility to obtain cash from a provider other than the one they normally used was a welcome feature in this context. Desirable options for further flexibility were also identified, such as changes in the ceiling of the maximum transfer in a day which was governed by the DSWD’s agreement with the Land Bank of the Philippines. The case study and other reviews of the response to Typhoon Haiyan conclude that it is relevant to continue to explore ways to link social protection interventions with disas- ter response, especially as the DSWD is the lead agency for both. This might include, for instance, identifying triggers to mobilize the use of social protection programs in an emergency. It will be opportune to build on previous experiences while supporting improvements to the practicalities of scaling up program operations in crisis contexts. 3.4 Takeaways from transferring international assistance onto national systems „„ In countries that are still at a very early stage in developing their national capac- ity in social protection, such as Liberia, the opportunities for the meaningful use of national social protection programs and systems in response to a major acute crisis on a national scale, such as the Ebola viral disease, may be limited. In the case of Liberia, indeed, the alternative entailed experimenting with bilateral World Bank–UN or trilateral World Bank–government–UN contracting arrangements. Ongoing work by the World Bank and UN partners is geared to further codify, adapt and institu- tionalize procedures between parties. „„ In contexts where assistance might be heavily donor funded, such as in Palestine, a shared programmatic framework adopted by different actors can help foster national ownership and coherent delivery of social protection and humanitarian assistance. „„ When using government systems, humanitarian assistance can help reduce fragmen- 19 tation. However, the Philippines shows that, especially at the height of the crisis, there are also managerial costs for the government associated with such support. Moreover, there can be extensive needs among populations not served by social protection. 4  Improving links between emergency response and longer- term programming 4.1 Guinea: school feeding programs in a crisis context In Guinea, the government has gradually been expanding its provision of social protec- tion beyond contributory social insurance for formal sector employees. Since 2011, the World Bank has been providing support to lay the foundations of a social safety net strategy, through support to both policy development and the implementation of pilot programs. The process for elaborating a national social protection policy was launched in 2014, and the policy document was submitted for the approval of the head of gov- ernment in December 2016. A key component of the World Bank’s support had been a $25 million grant for a Productive Social Safety Net Project (PSSNP). Effective in February 2013, it comprised a public works program, a pilot cash transfer, and an institu- tional capacity-building component. In late 2013, a supervision mission recommended a revision of the cash transfer component to strengthen education and health outcomes. Hence, a redesign was already underway when the Ebola crisis broke out in March 2014. The restructuring was completed in November 2014, resulting in the introduction of a school feeding subcomponent, to be delivered by WFP. A separate component was cre- ated to assist UNICEF’s logistical support and other services to respond to the epidemic. The WFP-implemented school feeding project was both affected by, and had an impact on, the Ebola crisis. Its implementation was delayed while schools were closed; how- ever, once they reopened it reached a greater number of schools than planned (54 instead of 40) but for a shorter time period (13 months rather than 18). The schools funded by the World Bank’s pilot focused on remote and poorly serviced areas. Their number was small compared with WFP’s regular national school feeding assistance pro- gram which reached about 1,600 schools in 2015–16. The case study highlights several positive features—either noted by the community or drawn from WFP’s monitoring reports—relating to the fact that the school feed- 20 ing program was functioning most of the time during the Ebola crisis. First, it was reported to have an impact on households’ overall food security, which was particularly 4  Improving links between emergency response and longer-term programming important in Ebola-affected areas where food security was severely affected. Second, it alleviated the economic burden of Ebola orphans on their hosting families. Third, the provision of support to all students (not just those affected by Ebola) contributed to rebuilding communities’ trust in the authorities. The case study affirms the pragmatism of the approach taken by the World Bank, WFP and the Government of Guinea in adjusting the design of the PSSNP. Also, it docu- mented that school feeding programs could serve as a critical safety net even in areas that are highly resource constrained and face challenging conditions of accessibility. Coordination and communication could be further improved through a memorandum of understanding between interested parties, and improved data collection and reporting, to ensure that lessons learned are better integrated into future improvements to the national school feeding program. 4.2 Mauritania: crisis response as a catalyst for building social protection In Mauritania, domestic programs have a long tradition of providing food subsidies (e.g., Emel shops). However, in times of crises these programs faced severe challenges in reaching parts of the country with low population densities. International humanitar- ian assistance has been provided on a more targeted and seasonal basis to complement existing national measures. These measures have started to gradually influence and build a social protection system. Some milestone of this process since 2011 are illus- trated in figure 4. In 2011, the EU delegation funded a program of four experiences of “cash based social safety nets,” implemented by Action Contre la Faim, Gret, the French Red Cross and WFP. The WFP pilot program was eventually implemented in nine regions, covering Nouakchott and the southern rural areas. The Ministry of Economy and Finance asked WFP and the governmental Commissariat à la Sécurité Alimentaire (CSA) to establish a register of vulnerable households in Nouakchott and develop a targeting strategy. These early cash-based experiences were followed by a “post-2012 crisis” program, implemented at larger scale by the same actors. UNICEF, the Food and Agriculture Organization of the United Nations (FAO), and WFP devised a joint approach that, while increasingly adopting cash as a core transfer modality, also included a harmonizing Figure 4  Humanitarian–social protection milestones in Mauritania 2011 2012 2013 2014 2016 Cash transfers by Scale-up and International Government adopts Expansion of ACF, Gret, French harmonization of Monetary Fund national social registry to other Red Cross, WFP approaches by recommendation protection strategy 21 programs WFP Nouakchott UNICEF, FAO, of national registry Introduction of registry and and WFP based on Nouakchott national cash transfer targeting experience program supported by World Bank Human(itarian) Capital? of targeting approaches. These revolved around the Household Economy Analysis framework, which was adopted in the Sahel in the mid-2000s. The documentation of experiences, particularly those implemented by NGOs, generated technical exchanges about transfer frequency, amount, payments, conditionality, and complementary mea- sures. A “cash group” was launched, originally led by WFP and affiliated with the Cash Learning Partnership (CaLP). The group benefited from financial support from the EU. In 2012, the International Monetary Fund (IMF) alerted the Government of Mauritania to rising public expenditures on untargeted transfers while national revenues were declin- ing. In 2013, it recommended to adopt a strategic framework for social protection as well as the development of a targeting strategy based on the WFP-CSA experience in Nouakchott. The following year, Mauritania adopted the National Social Protection Strategy (SNPS), with its governing body including UNICEF, WFP and other mem- bers; also, the government requested the World Bank to support the development of a National Social Registry (NSR) as well as a cash transfer program (Tekavoul) targeting 100,000 poor household across the country. International humanitarian actors, and in particular those in the “cash group,” pro- gressively moved from technical coordination among themselves to more actively prioritizing engagement with social protection. The Ministry of Economy and Finance joined the cash group in September 2015, which renamed itself “cash and social pro- tection working group.” This adopted an advocacy strategy to institutionalize dialogue between its members, the government institutions and donor agencies around cash transfers and social protection issues. Although the institutional integration of this group is limited and only focused on cash transfers, its overall trajectory demonstrates members’ willingness to engage in technical dialogue around the NSR and SNPS. In 2016, at the initiative of the World Bank and WFP, a group of international stakehold- ers (including UNICEF, FAO, ECHO and NGOs) initiated strategic discussions on how to strengthen the crisis prevention and management systems, including early warning sys- tems and contingency planning. This is an important development connected to the launch of the registry. Household registration began in 2016 and was not limited to Tekavoul: other programs could identify their beneficiaries using their own filter. WFP plans to test its use for targeting seasonal transfers and asset-building activities, and potentially as an instrument for rapid targeting in a crisis. 4.3 Kenya: integrated information management for social protection and drought response The consolidation and harmonization of social protection schemes in Kenya has been underway for a number of years. The creation of the Social Protection Secretariat in 2010, the elaboration of the Kenya National Social Protection Policy in 2011, and the 22 establishment of the National Safety Net Program in 2013 form some cornerstones of this approach. These are guiding the development of a common operational framework 4  Improving links between emergency response and longer-term programming for many government-led programs, four of which already existed: the Cash Transfer for Orphans and Vulnerable Children, the People with Severe Disabilities Cash Transfer, the Older Persons Cash Transfer and the Hunger Safety Net Program.1 The framework aims to streamline methods and systems for functions common to all these programs, namely targeting, registration, payment, and the handling of updates, complaints and queries. A key component of the consolidation process has been the launch of a computer- ized MIS, the Single Registry, in 2016. While each program is responsible for collecting and maintaining data on its own beneficiaries and activities in its own MIS, essential data from each program MIS is conveyed to the common platform, the Single Registry, where it can be analyzed and reports generated. 2 It also links to the national identity system, the Integrated Population Registration Services. The Single Registry enables, for example, analysis of the number of beneficiaries of different programs by area, or the identification of households receiving multiple benefits. In this way it is intended to improve the delivery of social protection programs themselves. Of particular interest to this paper is the fact that the database is also designed to facil- itate the provision of support to households at times of crisis (for whom a strengthened social protection system is already valuable). It does this in two ways: first, other imple- menters—including those providing emergency response—can set up their own MISs in a way that also links to the Single Registry; and second, it permits analysis of coverage of existing programs which may help determine if they can be used for crisis response. WFP became the first nongovernment provider to link up with the Single Registry, feeding in information about the beneficiaries of its programs that support resilient livelihoods. The vision was tested in the response to the drought that was declared a national disas- ter in February 2017 following two seasons of poor rainfall in 2016. Government and donors sought to use the Single Registry to determine the coverage of social assistance in affected regions and to project additional coverage for the response. Looking to the future, plans are in place to improve the functionality of the Single Registry in crisis con- texts through a number of measures. These include, among others, updating the field that captures the location of households to reflect the new county structure; linking to other programs delivered by humanitarian actors, as well as to other major government programs that have a social protection element, such as the National Health Insurance Fund; and embarking on the decentralization of the registry to county level, since county governments can also implement their own programs. 23 1  Kenya has also recently introduced an old-age social pension for people aged 70 and above. 2  Kenya’s Single Registry can therefore be classified as an “integrated beneficiary registry” (see e.g., Barca, 2017, for a classification of types of social protection database). Human(itarian) Capital? 4.4 Mozambique: enhancing connections between emergency response and social protection Mozambique’s national disaster risk management system, led by an agency known as the National Institute of Disaster Management (INGC), has been consistently strength- ened over the last decade in terms of its policy development, contingency planning, institutional capacity and coordination function. The UN activates a Humanitarian Country Team when required in the event of crises that exceed the government’s capac- ity to respond. These systems have mainly been directed at rapid-onset shocks such as cyclones and floods; the 2015–16 drought has shifted some of this focus. Several recent studies have explored if and how social protection—led by the Ministry of Gender, Children and Social Action and its implementing arm, the National Institute of Social Action (INAS)—might contribute to the preparedness and response to natural hazards. An updated National Strategy for Basic Social Security, ENSSB II, was approved in 2016 and offers an overarching framework for those connections. The four main social protection programs are a cash transfer program (PSSB), labor-in- tensive public works (PASP), social care services (PSSAS), and assistance for vulnerable families (PASD). The social protection sector is significantly constrained at finan- cial and operational level even to deliver its current commitments to these programs. Opportunities are therefore limited to contribute strongly to disaster risk management and humanitarian action in the short term. More recently, Mozambique has been designing a program called “PASD-PE” (“PASD post-emergency”)—a social protection direct support program for post-emergen- cies. Supported by an additional finance investment program by the World Bank, the PASD-PE is designed as an unconditional cash transfer to allow a smooth transition from the humanitarian response. Beneficiaries of PASD-PE will be derived from the list of households that are beneficiaries of the emergency support provided by the INGC. Following a temporary support for 12 months, the intention is that PASD-PE benefi- ciaries are to be referred to a social pension program, labor intensive public works, or other interventions. PASD-PE is currently implemented in three districts in Southern Mozambique, supporting 18.500 households. 3 As the government has not been wholly in favor of using cash transfers as a modality for humanitarian assistance, Mozambique does not have this natural point of conver- gence between emergency and long-term cash transfers that is becoming a feature for links between emergency response and social protection elsewhere. However, efforts are being made to strengthen the social protection sector in its own right, and close linkages have been documented in other programs. For instance, under PASP the World Bank and WFP have collaborated with INAS on defining the eligibility criteria 24 3  Murragarra (2018), personal communication. 4  Improving links between emergency response and longer-term programming for potential beneficiaries, resulting in the PASP adopting a proxy means test combined with food consumption scores. The experience of humanitarian agencies in implement- ing lean-season and emergency public works led the World Bank to adopt WFP quality and monitoring protocols on asset creation programs; at the same time, WFP adopted the national public works manual, which was developed with Bank support. At a national dialogue on shock-responsive social protection in June 2018, jointly facil- itated by WFP and the World Bank, the government and its partners agreed on the need for follow-up action in six areas: „„ Preparation of collaboration protocols between INGC and INAS to clarify role and responsibilities (a joint activity with WFP). „„ Identification and prioritization of beneficiaries, including through geographic risk analysis and vulnerability assessments. District-level vulnerability to climate shocks may be a relevant criterion to consider for the geographical expansion of social pro- tection interventions which do not yet have national coverage. Drought-affected areas may be particularly relevant for greater coverage by routine social protection, not only because INGC operations have until recently focused more on rapid-onset disasters, but also because drought tends to occur in similar geographical areas and builds up over a number of months. „„ Clarification of any triggers for activation of a social protection program in an emergency. „„ Strengthening the collaboration between INAS and INGC at all levels of planning especially at district level. This could include the integration of social protection into district-level resilience plans, especially in relation to climate-sensitive assets that could be created under PASP. The government, with WFP support, has been field testing links between the creation of Local Adaptation Plans and community-level approaches for analyzing resilience. „„ Establishing integrated databases between INGC and INAS. The limited availability of high-quality data from the Ministry of Gender, Children and Social Action/INAS on vulnerable populations is a major constraint on its ability to add value to plan- ning for disaster risk management. INAS is still developing its MIS; once operational this should significantly improve its capacity. „„ Affirming the leadership of the government in this process. 4.5 Zimbabwe: improving the response to seasonal food insecurity and re-establishing effective social protection Zimbabwe is a low-income country that, like Mauritania, has high rates of food inse- curity and malnutrition. Social, economic and institutional fragility have undermined 25 Human(itarian) Capital? development in general and social protection capabilities at national and local levels. Weather-related shocks impede economic recovery, and food production is reported to be increasingly stressed by climate change. These, together with stretched state capacity, have resulted in greater visibility of nonstate programming. Yet Zimbabwe previously had one of the first and most sophisticated social protection systems in the region. In this context, the case study of WFP’s experience highlights options for enhancing responses to annual food insecurity in a way that may contribute to the re-establishment of longer term national social protection planning. This is consistent with the 2016 National Social Protection Policy Framework, which makes it clear that social assistance is core to reducing vulnerability to shocks. The prevalence of poverty, especially in rural areas, necessarily requires almost simultaneous delivery of short- and medium-to-long-term interventions. International agency interventions at the interface of emergency response and long- term social protection included various experiences. A major one is data collection, analysis and dissemination on vulnerability, food security and nutrition. WFP has led the use of three analytical tools that collect data relevant for both crisis response and social protection planning: (i) the Integrated Context Analysis tool generates an over- view of geographical areas prone to different hazards, as well as trends in food and nutrition insecurity. (ii) At subnational level, District Risk Profiles are developed on the basis of 10-year trend analyses to understand underlying causes of chronic poverty and vulnerability to shocks. A key feature is the Seasonal Livelihood Programming cal- endar, which aims to guide e.g., the location and timing of planned interventions. (iii) Community-based Participatory Planning is used to develop action plans in which com- munities identify their priorities for building resilience. Based on these experiences, WFP—alongside other UN agencies—has influenced the analytical frameworks and indicators used by the Zimbabwe Vulnerability Assessment Committee for its annual assessments. At the time of the study it was suggested that there might be an opportunity for WFP and the World Bank to collaborate on enhanc- ing the capacity of local structures for local data gathering, analysis and use for decision making. Agencies have also been supporting the design and implementation of safety nets built around food insecurity. Working with the state and other relevant agencies, WFP deliver Lean Season Assistance to households during the peak hunger season (January to March), and implements productive asset creation activities outside that season (May to November). The Lean Season Assistance, which increasingly uses a mix of in-kind food and cash to stabilize consumption, shares many features with a typical long-term social protection program. For instance, beneficiaries are issued with smartcards which they can use to spend benefits at local shops. The productive asset creation activities focus on “food assistance for assets”: they aim to provide participants with immediate support to access food, while creating assets such as small irrigation schemes or dip tanks for 26 livestock that are intended to mitigate the consequences of future shocks. Some asset 4  Improving links between emergency response and longer-term programming creation schemes are also linked to the provision of agricultural insurance for small- holder farmers to compensate for weather-related losses. 4.6 Fiji: sequence of government and donor responses to natural disasters In February 2016, Fiji was struck by Cyclone Winston, the most intense storm ever registered in the southern hemisphere. Some 540,000 people (62 percent of the pop- ulation) were affected. The National Disaster Management Office led the response, activating all government-led sectoral clusters. Its actions injected about $160 million into the economy through existing programs spanning employment, protection, food security and shelter. In addition, the country received about $33 million in international humanitarian assistance, part of which was also channeled through social protection schemes. The structures and arrangements for disaster risk management in Fiji, led by the National Disaster Management Office, are comprehensive and have undergone recent revision. They outline the roles and responsibilities of all government agencies, and require all agencies to incorporate disaster risk management practices into their plans and budgets. So, while many aspects of crisis response are unrelated to social protec- tion, and are thus dealt with by other ministries, it was nonetheless important for the Department of Social Welfare to determine how best to contribute to the relief and recovery effort. One of the most mature systems in the Pacific region, the Fiji social pro- tection system was able to be leveraged in four ways. „„ First, the government provided cash transfer top-ups. In March, the government funded additional cash to all beneficiaries of its three main social assistance pro- grams—the Poverty Benefits Scheme, the Care and Protection Allowance, and the Social Pension Scheme—reaching almost 44,000 households.4 The transfer con- sisted of a lump sum of F$600 to each beneficiary of the Poverty Benefits Scheme, and F$300 to beneficiaries of the other two, equating to roughly three months of their usual monthly transfer value. This was intended to contribute to meeting households’ needs for February to April. „„ Second, WFP’s response provided a further top-up. In May and June, at the gov- ernment’s request WFP provided further top-ups to some beneficiaries of these schemes, targeted to the most severely affected locations. It paid between F$50–150 per month per household, depending on the scheme. WFP used the 4  All three are cash transfer programs. The Poverty Benefits Scheme supports the poorest 27 10 percent of households in Fiji, identified through a proxy means test. The Care and Protection Allowance is targeted at vulnerable households with children. The Social Pension Scheme sup- ports people aged 66 and over. Some beneficiaries of these schemes are also eligible for a supplementary Food Voucher Program. Human(itarian) Capital? government’s delivery mechanisms where possible. Modalities included e-vouchers in places with supermarkets and for people familiar with an existing Food Voucher Program. Paper vouchers were provided to pensioners not previously using e-vouch- ers, while cash transfers were distributed where vouchers could not be redeemed. „„ Third, withdrawal of pension funds was made possible. Members of the Fiji National Provident Fund—the contributory social insurance scheme, mostly for formal sector workers—were authorized to make a one-off withdrawal of F$1,000 from their pension funds, and a further F$5,000 if their house was in a cyclone-affected area. Some 170,000 withdrawals were approved. This option provided by far the largest injection of cash into the economy, releasing about F$250 million, or 3 percent of gross domestic product (GDP). „„ Finally, Help for Homes was introduced as a new cash transfer program for house- holds below a certain income threshold for home repairs. It should be noted that, in the first month after the disaster, nearly all cyclone-affected households received some in-kind emergency relief in the form of food or building materials. The subsequent use of the social protection schemes was intended to com- plement, not replace, this established system for crisis response. Other small-scale cash or in-kind distribution programs were also delivered by NGOs. Program evaluations and reviews have identified some key enabling factors and con- straints arising from this use of the government’s social protection system after the cyclone. A primary advantage was that the national schemes were well established, with robust delivery systems. The cyclone struck the poorest part of Fiji where many people were already enrolled onto the poverty-targeted scheme, so it seemed logical to reach them using the same channel. A major limitation, however, was that nonprogram beneficiaries did not receive cash from the government or WFP. The Poverty Benefit Scheme had collected some data on “near-poor” families, but they were not updated electronically; meanwhile, those who were newly poor because of the cyclone may not have been on a list. While top-ups may have had positive impact on beneficiaries, non- beneficiaries were subsequently found to lag behind in their recovery. Agencies were advised to plan alternatives to reach nonprogram beneficiaries in future crises. Upgrading the Poverty Benefit Scheme database might be one means of more quickly tracing these excluded households. Improving data collection and mapping of vulnerabilities might also be valuable. Indeed, there was some inclusion of non-affected households in the government’s top-up as it did not geographically target its assistance. It was difficult to distinguish the affected areas in its database, and the need for assis- tance was too urgent to resolve this. Officials expected some redistribution among households, though this assumption was not necessarily found in practice. 28 The response generated lessons for policy and planning, including a recommendation to develop guidelines on the use of social protection under different disaster scenar- ios. It was also agreed to create a National Humanitarian Policy to guide the use of early 4  Improving links between emergency response and longer-term programming warning systems and contingency planning, and to better specify the role of interna- tional humanitarian agencies in crisis response. With the withdrawal of pension funds, households received short-term relief, although will also face reduced pensions in the future. Also, they may not be able to access further funds in future emergencies if they have already reached their maximum with- drawal limit. 4.7 Lebanon: supporting refugees alongside the national population Faced with an influx of 1.5 million Syrians, Lebanon and its partners have developed their support to the growing numbers of Lebanese pushed into poverty as a result of the crisis, at the same time that international agencies in the country have been address- ing needs among the refugees themselves. The introduction by external actors (United Nations Refugee Agency [UNHCR], WFP) of a comprehensive transfer system for Syrian refugees and, to a lesser extent, for Lebanese families affected by the crisis, offers an opportunity for reforming and updating longer-term national social protection systems. Some convergence between delivery system for refugee assistance and Lebanon’s exist- ing social protection programs has already taken place. Lebanon’s National Poverty Targeting Program (NPTP), supported by the World Bank, was launched in October 2011 to provide fee waivers for health and education targeted at the poorest and most vulnerable Lebanese households—coverage that continues today for 105,000 households. In May 2012, the Government of Lebanon requested WFP to return to Lebanon to address the food and nutrition needs of the growing population of Syrian refugees in the country. WFP began delivering food assistance to Syrian refugees in June 2012 using paper vouchers. WFP scaled up and expanded its assistance program in 2013, and shifted its transfer modality from the paper voucher to an electronic card providing assistance. Functioning markets, technical capacity, adequate banking services and infra- structure throughout the country allowed for this change. By December 2013, more than 500,000 refugees had received an electronic food voucher redeemable in local shops throughout the country, contributing to developing value chains and having a positive impact on the Lebanese economy. In 2017, the number of assisted beneficiaries reached 680,000 following a series of validation and targeting exercises. The Banque Libano Française is the financial service provider of the prepaid card system used for registered Syrian refugees. The system enables targeted beneficiaries to pur- chase food commodities at WFP-contracted shops only. As of today, over 480 retailers throughout Lebanon contracted by WFP have a devoted point-of-sale device to process 29 beneficiaries’ e-cards. For the successful roll-out of this innovative approach, WFP relied on the guidance and support of the Lebanese Central Bank. To receive assistance from Human(itarian) Capital? WFP, Syrian refugees must be registered with UNHCR. WFP provides e-cards to tar- geted refugees in Lebanon based upon their refugee status and eligibility for assistance, as determined by the Vulnerability Assessment of Syrian Refugees in Lebanon (a multi- sector analysis that WFP conducts annually with UNHCR and UNICEF). By early 2014, at the height of the influx of refugees into Lebanon, and with the WFP e-card food voucher program in full swing throughout the country for refugees, there was increasing evidence of growing tensions among poor Lebanese families and ref- ugees residing within the same communities. This was primarily due to the fact that Lebanese families were ineligible to receive WFP e-card assistance. These growing ten- sions were further identified by regular field visits conducted by the World Bank team, the Ministry of Social Affairs’ (MOSA’s) social workers, as well as WFP. In response, and to help mitigate the impact of the refugee influx on Lebanon (the main recommendation of the Lebanon economic and social impact assessment), the World Bank proposed to the government the introduction of the e-card food vouch- ers for poor Lebanese families, and approached WFP. Both the World Bank and WFP saw the importance of this step not only in terms of a means of reducing poverty and tension among the two communities, but also to strengthen the national system (the NPTP). If successful, the international community would have contributed to developing Lebanon’s social safety net system. In November 2014, the NPTP was scaled up through the Emergency NPTP project (E-NPTP) with the support of an $8.2 million grant from the Lebanon Syrian Crisis Trust Fund. A second phase of support was provided through a $10 million grant in 2016 (Additional Financing to the E-NPTP—World Bank). By introducing food assistance via the e-card food voucher to NPTP beneficiaries, the scale ups aimed to help allevi- ate extreme poverty experienced by vulnerable Lebanese and reduce tension between Lebanese host communities and Syrian refugees by providing a level of assistance parity received by refugees. NPTP also included financing operational support, training and capacity development assistance for MOSA from WFP to assume the overall responsibility for the imple- mentation of the food voucher program within the following areas: (i) beneficiary sensitization on the use, maintenance and operation of the e-card; (ii) distribution of e-cards to beneficiaries; (iii) assessment, monitoring (pre-assessment baseline, postdistri- bution monitoring and sampling) and reporting tools; (iv) beneficiary data management; and (iv) fraud detection techniques. Based on funding availability, the most vulnerable 5,076 households from the NPTP beneficiary database (27,209 individuals) were initially deemed eligible to receive the e-card. Since August 2016, coverage of the e-card food voucher has been expanding at a rate of approximately 1,000 households per month. As such, the number of NPTP 30 beneficiary households receiving food assistance through the e-card reached 10,008 households (52,724 individuals) in December 2016. 4  Improving links between emergency response and longer-term programming Both Syrian refugees and NPTP households benefiting from the WFP e-cards are eli- gible to use them within the 480 WFP-contracted shops across the country. To ensure not to overwhelm contracted shops, it was agreed to separate the dates of card loading for Syrian refugees and Lebanese participating in NPTP. Therefore, refugee e-cards are loaded on the fifth of every month, while NPTP e-cards are loaded on the 15th of each month. The value of the unconditional cash transfer received monthly by each group of beneficiaries through the e-card is also harmonized, so as to avoid any social tensions which could arise. As such, the value of the NPTP voucher has ranged between $27 and $30 per household member per month, capped at six household members. For refu- gees, the voucher value has remained stable at $27 since February 2016. 4.8 Takeaways from linking emergency response and longer-term programming „„ If well documented, quality pilots can have demonstration effects at national scale: the experience of Mauritania illuminates how small-scale, externally funded human- itarian programs informed subsequent large-scale initiatives, including as supported by international finance institutions. „„ Large-scale emergency responses can provide a similar “proof of concept” for national social protection programs. The Lebanon case study has shown this in rela- tion to the extension of e-voucher to citizens modeled after the parallel refugee scheme. This is particularly compelling for ensuring equity in provision in urban areas where refugees and local residents live “side by side” and tensions may escalate rapidly. „„ Guinea’s pilot of school feeding funded by the World Bank and implemented by WFP in some of the remotest areas affected by Ebola show the benefits of protect- ing human capital during crises. „„ Sharing information through a common registry does not require a common plat- form per se, but the ability of different platforms to connect to each other. Kenya is a premier example on how this process was achieved for national social protection programs and extended to international humanitarian schemes. „„ Humanitarian actors have developed a number of data collection, analytical and pro- gramming tools which feed into national country diagnostics, such as the Zimbabwe Vulnerability Assessment Committee. These would be key for future expansions in social protection. Meanwhile, data that are generated for the use of social protec- tion programs may also be relevant for emergencies, whether they are collected by national or international actors. 31 „„ Rapid-onset emergencies when a rapid response may be a priority could entail sup- porting needs that may not necessarily be covered by social protection systems, Human(itarian) Capital? whether by design or because of outdated information induced by the crisis. Hence, it is important to plan in advance how needs will be met, including through social protection schemes and other means. „„ While in some cases international humanitarian assistance is the first frontline pro- vider of support, this may not always be the case. The experience of countries like Fiji, for instance, illustrates a possible sequencing in rolling-out of assistance, includ- ing by first extending government programs, and then providing top-up resources by the humanitarian community. Such assistance can also be diversified according to spatial circumstances (e.g., availability of markets) and include vouchers. However, this can result in gaps in coverage among those not originally enrolled in govern- ment schemes. 32 5  Provision of social protection in conflict settings 5.1 Syria: protecting lives and human capital during conflict The National Social Aid Fund, created in 2011 as a semi-autonomous entity under the Ministry of Social Affairs and Labor, has become of limited relevance for the scale of needs in the light of the subsequent years of conflict. With GDP more than halved, 80 percent of the population living below the poverty line, 13.5 million people in need of humanitarian assistance, parts of the country besieged, and the regime highly con- tested, there is little scope for working to strengthen government social protection. The World Bank’s activities in Syria, including support to social protection, have been sus- pended since 2011. WFP is implementing major humanitarian programs, but apart from maintaining relationships with counterpart Ministries, the case study confirms the lim- ited prospects for WFP or other humanitarian actors either to use existing government structures or to encourage sustainability in longer-term social protection schemes. A number of activities, however, could benefit an eventual Recovery and Peace Building Assessment and, further into the future, the re-establishment of a national social pro- tection program. These include initiatives where WFP continues to work with Syrian ministries, such as those for health and education. For example, WFP regularly produces and disseminates food security and nutrition vulnerability data as part of its General Food Assistance and nutrition interventions. This also includes the identification of food insecure districts, profiling of food insecure households, and market analysis. Other activities such as conditional cash transfers to support education (out-of-school chil- dren) and nutrition (for pregnant and lactating women), as well as support to resilient livelihoods related to agricultural and livestock, could help lay the basis for social pro- tection options as peace is re-established. 33 Human(itarian) Capital? 5.2 Yemen: addressing immediate needs and preserving social protection Unlike in Syria, the World Bank has maintained engagement and operations in Yemen throughout the crisis and conflict. It deployed substantial grant resources—including through restructuring and cancelation of its regular programs which were suspended due to the conflict—both to address immediate social protection needs and to preserve long-standing Bank, donor, and government investments in national social protec- tion systems. The Bank’s operational engagement and financing flexibility has allowed it to maintain a full partnership with international humanitarian actors such as the United Nations Development Programme (UNDP) and UNICEF (in social protection) and UNICEF, WHO, and WFP (in health and nutrition). In 2016, the World Bank began a substantial re-engagement through grant fund- ing, including via partnership with the UN agencies as recipients of International Development Association (IDA) grants. Seven major grants have been approved— including grants mainly channeled through UNDP and UNICEF for social protection programs—with the dual aim of directly assisting impoverished Yemenis (with cash assistance, income and livelihood opportunities, and facilitating health and nutrition ser- vices) and maintaining the capacity of Yemen’s national social protection programs. According to a detailed account by Alawi Al-Ahmadi and De Silva (2018, 38–39), …the World Bank has shown flexibility in the interpretation and application of its oper- ational policies, especially OP 2:30. The legal interpretation of operational policy (specifically paragraph 3 of OP2:30) stipulates that if there is no government in power, World Bank assistance may be initiated by requests from the international commu- nity (for example, from UN agencies), subject in each case to the prior approval of the Executive Directors of the World Bank… A request from the UN for the funding addressed this question [and] for the first time in the World Bank’s history, the pro- posed grants were to be made out of the country’s IDA resources…without government acquiescence. This support built on the experience of the World Bank’s engagement in Yemen’s social protection agenda since 1995, including responding to the 2011 Arab Spring crisis and the 2014 fiscal crisis. The Bank played a major role in building and strengthening Yemen’s flagship social protection institutions. These include the Social Welfare Fund (SWF), providing unconditional cash transfers to poor people, and the Social Fund for Development, engaged in public works, training, the development of small and medium enterprises. As in other crisis-affected countries covered in this study, government-led social protec- tion systems in Yemen were likely to be overstretched to meet a growing and changing 34 caseload as the conflict persisted. Nevertheless, the case of Yemen provides evidence that even in the most severe complex emergencies, social protection systems built for more stable settings might be preserved for the future while also contributing to 5  Provision of social protection in conflict settings immediate emergency response purposes. Even in the present acute phase of the crisis, international humanitarian actors such as WFP, UNICEF, and international NGOs con- tinue to work with national social protection programs, coordinating activities with the authorities on the ground, and adapting targeting and cash transfer decisions based on experiences built up in partnership with the government between 2008 and 2014. UNICEF, for example, as part of a Bank-funded $200 million cash transfer program to 1.5 million households, has adapted the SWF beneficiary list and design parameters of the national program, while factoring in the conflict and child protection dimensions. UNICEF continues to engage the SWF at the technical level, and the agency has bene- fited from the government’s technical staff expertise on local communication strategies. The emergency project has also introduced improved complaints and appeals mech- anisms and alternative payment agencies and fostering partnership with local private sector. These enhanced processes are expected to be transferred to SWF postconflict. Figure 5  Adapting social protection to conflict in Yemen Source: Alawi Al-Ahmadi and De Silva (2018) Figure 5 lays out main adaptations to social protection before and after the conflict. With the SWF beneficiary lists last updated in 2013, UNICEF carried out a beneficiary identity verification exercise prior to launching the emergency cash transfer program to ensure the integrity of the SWF list. While SWF beneficiary list was not updated to accommodate the “newly vulnerable”—conflict-linked vulnerabilities—WFP has turned increasingly to its own cloud-based beneficiary and transfer management platform (SCOPE). This was apt in a context with a dearth of alternatives, and could eventu- ally help strengthen Government systems and expand the national beneficiary registry. Other international agency interventions, such as the introduction of e-vouchers and biometric capability, or the introduction of alternative payment agencies, should like- 35 wise help strengthen national social protection systems in a postconflict Yemen. Human(itarian) Capital? 5.3 Takeaways from conflict settings „„ When national delivery systems for social protection exist precrisis, they offer the possibility to respond even in the midst of an active and complex conflict. They do so by providing a starting point for targeting and delivery mechanisms for interna- tional humanitarian assistance to individuals or households. The suitability of this action depends on the extent to which those targeted by the national social pro- tection program are also affected by the complex emergency, as well as on the functioning of the mechanism in the crisis. „„ Conversely, the work carried out by humanitarian actors in complex crises like Syria presents an array of activities that, while geared to short-term relief, may help lay the basis for future re-engagement in social protection. These include geospatial information systems, market analysis and nutritional programs, among others. „„ Development agency support to maintain the capacity of national systems during crises—as in Yemen—may yield benefits for relaunching national social protection programs after the crisis and rebuilding country systems. Grants for direct trans- fers to beneficiaries, using national systems, serve a valuable humanitarian purpose. Development-focused agencies intervening directly in complex emergencies may want to take into consideration the humanitarian principles when planning their approach. „„ Channeling World Bank grants to humanitarian agencies (such as UNICEF and UNDP in Yemen) during a conflict may help forge or maintain partnerships that will be useful for postcrisis investments in national social protection systems. „„ It will be important to consider how lessons and systems generated by UN agencies during an emergency (vulnerability data, beneficiary lists, distribution systems, etc.) can be retained and, if appropriate, shared with governments in a postcrisis setting, particularly if personnel and resources are redirected elsewhere. 36 PART III Conclusions 6  Emerging lessons and broader reflections A set of 10 key lessons emerge from the case studies and broader literature. These may lay the basis for a more in-depth conversation between the humanitarian and social protection communities on how to build upon positive experiences and address identified bottlenecks. „„ Leadership and engagement of national authorities must be the norm, not the exception. The international humanitarian community plays a pivotal role in saving lives and protecting people in contexts of pervasive risk and volatility. Where national institutions collapse because of conflict, for example, a parallel system could be established, humanitarian imperative upheld, and international law applied. However, such an approach cannot be exported to contexts with an existing and functional state—even if its capacities are nascent.1 Data and country experiences show that it is not uncommon for humanitarian assistance to be delivered and coor- dinated separately from national structures. The entry point for assistance in the country should always be the national government: if it is not feasible or appropri- ate to use national structures (because they are limited or inadequate), this should be demonstrated. A coherent assessment of national systems before crises hit could help map out, quantify and “stress test” those capacities. 2 Even where humanitarian assistance runs in parallel, there may be still scope to introduce practical elements 1  If international aid has previously been delivered by an agency committed to humanitarian prin- ciples, espousing neutrality and impartiality and based on the measured “needs” of individuals, a shift towards government-led programming—and, specifically, government response through its social protection schemes rather than through its own emergency response mechanisms—may result in a change in emphasis of the principles, including based on politics and political economy considerations. 2  See, for example, the various tools developed under the ISPA multi-agency platform (https:// 38 ispatools.org/). In addition, a “stress test” of social protection by different shocks and scenarios may help identify ex ante priority areas for investment, plan support by humanitarian and devel- opment actors, as well as inform the outsourcing of risks via insurance products in ways that conform to and strengthen national systems. 6  Emerging lessons and broader reflections of alignment with pre-existing or future national social protection systems, such as the use of local domestic institutions in Yemen. „„ Crises may present a trade-off between degree of ownership and a range of dimensions, such as timeliness or accountability. Case studies show that—espe- cially in low-capacity contexts— when crises hit, the design of the response entails several core choices. 3 The first of these may be a trade-off between national engage- ment and ownership and speed of response. The channeling of international funds for emergency response through a government’s financial channel can strengthen ownership; however, this may be slower to disburse than alternative routes (e.g., the Ebola response in Liberia). In Palestine, the domestic institutionalization of WFP vouchers came at the cost of a slower rate in expansion, but this was deemed acceptable given the gains in ownership. Of course, this is not always the case, as the Fiji case study demonstrates, where government responded before the inter- national humanitarian community. A second trade-off for international actors may revolve around project accountability vs. program ownership. For international actors, working through government structures may entail different reporting pro- cedures. Parallel structures create fragmentation of overall crisis response, but may enhance accountability of specific projects. Reporting requirements, for instance, are often specified in donor contracts. Governments, too, can be affected by different reporting burdens, depending on the nature of their collaboration with partners. 4 This implies that international agencies and donors could better align with national processes, and pre-agree on special circumstances and arrangements whereby, in the interest of effectiveness, responses could be adjusted. Operations could be facilitated by ongoing work by the World Bank and UN agencies to standardize operational reporting templates as part of contractual agreements. „„ There is a need for strategic, institutional and technical brokering at the intersec- tion of humanitarian assistance—in all its definitions—and social protection. As international humanitarian agencies develop approaches more closely connected to social protection, and as development institutions move their operational frontier toward more challenging contexts, there is a need to carefully manage the inter- face of nascent national social protection systems and international humanitarian 3  O’Brien et al. (2018) identify seven factors that might be perceived to entail trade-offs in programme design and the use of government systems during crises. These are the ability to provide relevant support; coverage of the affected population; timeliness of response; sus- tainability (especially national government ownership); predictability for implementers and recipients; minimization of duplicated systems; and cost. 4  To some extent, the issues also pertain to approaches around social funds and Community Driven Development. While these have not be the subject of our analysis, it might be useful to conduct more research on their opportunities and limitations, especially in complex crises. 39 Experimental evidence from Sierra Leone, for example, points to CDDs as a powerful vehicle for delivering assistance, while presenting limitations in terms of institutional inclusiveness, e.g., tap- ping skills within the community for effectively submitting grant proposals instead of relying on village leaders (Casey et al. 2018). Human(itarian) Capital? programs. This may entail a brokering function by designated bodies tasked with navigating the technical and political economy challenges that transitions entail. „„ Shared planning and preparedness would help minimize unintended conse- quences. The prospects of effective collaboration between actors working in social protection and in emergency response may be increased if there is a unified vision on diagnostics (i.e., a common understanding of the nature of the problem), and agree- ment on the use of key mechanisms such as early warning systems. In contexts like Palestine, actors have rallied around a concerted effort for diagnostics and match- ing vulnerability profiles with interventions. In other cases, the fact that crisis-related information systems—e.g., the International Phase Classification (IPC)—are often used by the humanitarian community and less so in social protection can gener- ate unintended negative implications. Different diagnostics and early warnings have deep implications for design: if an IPC hits level five (famines), this has implications for the level of assistance provided; instead, if social protection programs are based on more chronic poverty variables and the IPC is not included in the policy frame- work, there might be tensions in areas reached by both spheres. This was recently the case for Madagascar, for example, which was not included in the case studies but may offer valuable future lessons. „„ There is not always a neat separation between humanitarian and social protec- tion systems. In most low-income countries, external financing heavily supports both humanitarian assistance and social protection. Also, our case studies point to a granular, modular division of labor between actors—sometimes governments may provide the channel for transferring resources, but with programs being imple- mented by other actors (e.g., Liberia); or external actors may channel resources, with programs being implemented by domestic institutions (e.g., Yemen). Recognizing this variety of models could help nuance the discussion on how to more flexibly, con- cretely and creatively manage humanitarian–social protection transitions along the continuum of combinations. 5 „„ The absorptive capacity of national systems needs to be considered. National social protection schemes in low and middle-income countries may be stretched financially, institutionally, and administratively. This may be the case even before con- sidering new caseloads that result from covariate shocks or from the transfer of beneficiaries as previously supported by humanitarian assistance. Compounding this, development actors may suspend capacity-building investments during crises, while humanitarian actors may interrupt capacity-building initiatives when a crisis subsides and funding declines. However, even where programs themselves are not mature enough to support an additional caseload, routine social protection programs can 40 5  We find that it can sometimes be hard to identify if an intervention is responding to a shock or to chronic poverty, especially as exacerbated by seasonal variations (Mauritania, Zimbabwe) (Bradbury 1998). The distinction may matter less for considering the most appropriate response, though it may have a considerable difference in terms of duration, funding etc. 6  Emerging lessons and broader reflections still be developed in a way that considers future shocks, such as by adjusting the gradual rollout of a program so that it prioritizes disaster-prone areas (Mozambique). The absorption of additional beneficiaries (previously served by humanitarian assis- tance) by social protection systems should be calibrated to the sector’s capacity. To do so, measuring the degree of “readiness” of national systems to expand is key. „„ Cash transfer programs are a natural point of convergence between social pro- tection and humanitarian programming—though they are not, of course, the only type of program that has a potential crossover in crisis and noncrisis contexts (Alderman et al. 2017). The impacts of providing assistance through cash—when appropriate—have been amply investigated and demonstrated elsewhere. These relate not only to increased household consumption among direct beneficiaries, but also, potentially, to improved human development outcomes, multiplier effects in the local economy and/or greater financial inclusion, depending on the design (Bastagli et al. 2018). Our case studies have highlighted the use of cash transfers in a range of contexts, including protracted crisis (Palestine), slow-onset droughts (Kenya) and rapid-onset natural hazards such as typhoons (Philippines, Fiji). As for other types of program, the case studies offered examples of food distribution, school feeding programs, public works or asset creation programs (some of which in any case use cash as their modality of payment to recipients), a subsidy and a formal contributory pension scheme all having this potential role. „„ Both social protection and humanitarian assistance intersect with disaster risk management institutions. If the interaction between international agencies and government social protection ministries is being undertaken in a crisis context, the national disaster management authority should be expected have a role in coordi- nation and engagement in the response, if the crisis falls within its remit. Points of intersection may include data and tools on disaster preparedness and response that could be leveraged by social protection, and vice versa. Where there is agreement on the value of the social protection sector being used as a vehicle for emergency response, it is advantageous for this role to be acknowledged and framed consis- tently both in a social protection sector policy (Mauritania, Mozambique) and in the national disaster risk management strategy and plans (Fiji). In addition, it may be valuable for the role of international humanitarian actors to be clearly articulated ahead of disasters (Fiji). „„ There is an important agenda for more effectively building delivery systems in partnership. Many elements of a delivery system may be developed in common, or harmonized, between the interventions of international agencies and national gov- ernments, or between those handling responses in crisis and noncrisis contexts. Low-hanging fruit could include ensuring that actors (e.g., NGOs) collecting data for a given program do so in adherence to an agreed protocol of data collection. Our case studies have highlighted instances where governments and their part- 41 ners perceive the value in striving towards closer integration or sharing of ideas in their particular context across a range of elements: these include high quality data Human(itarian) Capital? collection and analysis, approaches to targeting, the development of integrated management information systems such as social registries, communications tools, operational manuals and payment mechanisms. In complex emergencies where reg- ular government service provision is not active, direct programmatic links may not always be possible. Nonetheless, international agencies may be able to conduct inter- ventions whilst being open to the eventual future adoption of successful practices (Syria), or to maintain support to previously functioning government institutions in order not to lose them (Yemen). Where government services are functioning, there can be a mutual exchange of ideas and sharing of approaches and delivery sys- tems, ranging from assessment forms to targeting methods, databases and payment mechanisms (Lebanon, Palestine). These links become easier if the governments have stronger social protection capacity (Fiji), and/or if their programs have similar target populations or share similar objectives. „„ There is a need for better documentation of lessons, performance and knowledge sharing in humanitarian assistance. The humanitarian programs of today might be the social protection program of tomorrow. Indeed, national social protection sys- tems often originate from humanitarian assistance. This was the case of Mauritania, where the knowledge, tools and practices of NGOs, international humanitarian actors, and donors was identified, institutionalized and scaled up by the govern- ment, including with support from the World Bank and the International Monetary Fund. The widely documented case of Ethiopia points to a similar process, with decades of experience with past (pre-2005) Employment Guarantee Schemes being leveraged to form the institutional and programmatic backbone of (post-2005) public works under the Productive Safety Net Program. In many ways, the extent to which social protection can build on humanitarian assistance hinges in part on the quality of evidence available, as well as the documentation of practices and lessons. While there are clear limitations in the kind of evaluations viable in crisis situations, there is an encouraging growing trend in investing in evidence generation and knowl- edge management even in some of the most challenging settings (Peterman et al. 2018). 42 R eferences Alawi Al-Ahmadi, A. and De Silva, S. (2018) “Delivering Social Protection in the Midst of Conflict and Crisis: The Case of Yemen.” World Bank, Social Protection and Labor Discussion Paper 1801. Washington, DC. Aldaba, F. (2017) “Linking Social Protection and Humanitarian Assistance in the Philippines: A Case Study.” Background note for the World Bank and WFP. Rome and Washington, DC. Alderman, H., Gentilini, U., and Yemtsov, R. (2017) The 1.5 Billion People Question: Food, Vouchers or Cash Transfers? World Bank. Washington, DC. Al-Qudsi, K. (2017) “Bridging Humanitarian Assistance and Social Protection: Syria Case Study.” Background note for the World Bank and WFP. Rome and Washington, DC. Barca, V. (2017), “Integrating Data and Information Management for Social Protection: Social Registries and Integrated Beneficiary Registries”. Department of Foreign Affairs and Trade, Canberra. Bastagli, F. (2014) “Responding to a Crisis: The Design and Delivery of Social Protection” ODI, London. Bastagli, F., Hagen-Zanker, J., Harman, L., Barca, V., Sturge, G., and Schmidt, T. (2018) “The Impact of Cash Transfers: A Review of Evidence form Low and Middle-Income Countries.” Journal of Social Policy, in press. Beazley, R., Solórzano, A. and Sossouvi, K. (2016) “Study on Shock-Responsive Social Protection in Latin America and the Caribbean: Theoretical Framework and Literature Review.” OPM and WFP. Oxford. Beegle, K., Coudouel, A. and Monsalve, E. (eds.) (2018) Realizing the Full Potential of Social Safety Nets in Africa. World Bank. Washington, DC. Bennett, C., Foley, M., Krebs, H., (2016) “Learning From the Past to Shape the Future: Lessons From the History of Humanitarian Action in Africa.” ODI. London. Bradbury, M. (1998) “Normalising the Crisis in Africa.” Disasters 22(4): 328–38. 43 Buchanan-Smith, M. and Maxwell, S. (1994) “Linking Relief and Development: An Introduction” IDS Bulletin 25(4): 2–16. Human(itarian) Capital? Cabot Venton, C. (2018) “Economics of Resilience to Drought in Ethiopia, Kenya, and Somalia.” USAID. Washington, DC. Casey, K., Glennerster, R., Miguel, E., and Voors, M. (2018) “Skills versus Voice in Local Development.” NBER, Working Paper 25022. Cambridge. Cash Learning Partnership (CaLP) (2018) The State of World’s Cash Report 2018. Oxford. CGD and ODI (2015) Doing Cash Differently: How Cash Transfers Can Transform Humanitarian Aid . High Level Panel on Humanitarian Cash Transfers. London and Washington, DC. Chatiza, K. (2017), “The humanitarian assistance–social protection link: a case study of Zimbabwe.” Background note for the World Bank and WFP. Rome and Washington, DC. Cherrier, C. (2014) Cash Transfers and Resilience: Strengthening Linkages between Emergency Cash Transfers and National Social Transfer Programmes in the Sahel.” CaLP, Oxford. Crawford, N. (2017) “Linking Humanitarian Assistance and Social Protection: Summary of Case Studies.” Draft synthesis paper for the World Bank and WFP. Rome and Washington, DC. Creti, P. (2017) “School feeding as a safety net response during the Ebola crisis in Guinea.” Background note for the World Bank and WFP. Rome and Washington, DC. Davies, M., Guenther, B., Leavy, J., Mitchell, T. and Tanner, T. (2009) “Climate Change Adaptation, Disaster Risk Reduction and Social Protection: Complementary Roles in Agriculture and Rural Growth?” IDS, Working Paper 320. Brighton. Development Initiatives (2018) Global Humanitarian Assistance Report 2018. London. European Commission (EC) (2017) “Joint Communication to the European Parliament and the Council: A Strategic Approach to Resilience in the EU’s External Action.” Brussels. Galluzzi, C. and Natsheh, S. (2010) “Market-Based Food Assistance in Protracted Crisis: Vouchers in the Occupied Palestinian Territory.” In Omamo, S.W., Gentilini, U. and Sandstrom, S. (eds) Revolution: From Food Aid to Food Assistance. Innovations in Overcoming Hunger. WFP, Rome. Gentilini, U. (2016a) The Other Side of the Coin: The Comparative Evidence of Cash and In-Kind Transfers in Humanitarian Situations. World Bank. Washington, DC. Gentilini, U. (2016b) “Sorting through the Hype: Exploring the Interface between Humanitarian Assistance and Safety Nets.” World Bank, Social Protection and Labor Policy Note 19. Washington, DC. Gentilini, U. (2015) “Entering the City: Emerging Evidence and Practices with Safety Nets ibn Urban Areas.” World Bank, Social Protection and Labor Discussion Paper 1504. Washington, DC. Harmer A. and J. Macrae (2004) “Beyond the Continuum: The Changing Role of Aid Policy in Protracted Crises.” ODI, HPG Report 18. London. Harvey, P. (2009) “Social Protection in Fragile States: Lessons Learned.” In OECD (ed.) Promoting Pro-Poor Growth: Social Protection. Paris. 44 Ismail Sayed, H. (2017) “Bridging Humanitarian Assistance and Social Safety Net Systems: The Syrian Refugee Crisis in Lebanon.” Background note for the World Bank and WFP. Rome and Washington, DC. References Konyndyk, J. (2018) “Rethinking the Humanitarian Business Model.” CGD Brief. Washington, DC. Kukrety, N. (2016) “Working with Cash Based Safety Nets in Humanitarian Contexts: Guidance Note for Humanitarian Practitioners.” CaLP, Oxford. Kuriakose, A., Heltberg, R., Wiseman, W., Costella, C., Cipryk, R. and Cornelius, S. (2013) “Climate-Responsive Social Protection.” Development Policy Review 31(S2): o19–o34. Leturque, H. (2017) “Linking Humanitarian Assistance and Social Protection: Mauritania Case Study.” Background note for the World Bank and WFP. Rome and Washington, DC. Lorenzon, F. (2017) “Linking Humanitarian Assistance and Social Protection: Yemen Case Study.” Background note for the World Bank and WFP. Rome and Washington, DC. Mansur, A., Rovira, L., Uwantege Hart, S. and Begum, S. (2017) “Linking Humanitarian Assistance and Social Protection: Fiji Case study.” Background note for the World Bank and WFP. Rome and Washington, DC. McCord, A. (2013) “Review of the Literature on Social Protection Shock Responses and Readiness.” ODI, London. Metcalfe-Hough, V. and Poole, L. (2018) Grand Bargain Annual Independent Report 2018. ODI. London. Namara, S. and Sefa, C. (2017) “Linking Humanitarian Assistance and Safety Nets under Ebola: A case study of the collaboration between the World Bank and the World Food Program in Liberia.” Background note for the World Bank and WFP. Rome and Washington, DC. Neves, R. (2017) “Linking Humanitarian Assistance and Social Protection: Mozambique Case Study.” Background note for the World Bank and WFP. Rome and Washington, DC. Nyanjui, J. (2017), “Kenya’s single registry for social protection: a case study.” Background note for the World Bank and WFP. Rome and Washington, DC. O’Brien, C., Scott, Z., Smith, G., Barca V., Kardan, A., Holmes, R., Watson, C. and Congrave, J. (2018) “Shock-Responsive Social Protection Systems Research: Synthesis Report.” Oxford Policy Management. Oxford. Oxford Policy Management (OPM) (2015) “DFID Shock-Responsive Social Protection Systems Research: Inception Report.” Oxford. Peterman, A., de Hoop, J., Cuesta, J. and Yuster, A. (2018) “Evidence on Social Protection in Contexts of Fragility and Forced Displacement.” UNICEF, Innocent Research Brief 2018– 22. Florence. Roelen, K., Longhurst, D. and Sabates-Wheeler, R. (2018) “The Role of Cash Transfers in Social Protection, Humanitarian Response and Shock-Responsive Social Protection.” IDS, Working Paper 517. Brighton. Sandford, J. (2018) “Responding to Shocks: Humanitarian response through national sys- tems: Ethiopia and Kenya Case Studies.” Draft reports prepared for the World Bank. Mimeo. Sarsou, S. (2017) “Linking Humanitarian Assistance and Social Protection: Palestine Case 45 Study.” Background note for the World Bank and WFP. Rome and Washington, DC Human(itarian) Capital? Stoddard, A. (2017) “Humanitarian Funding to Sub-Saharan Africa: Tracking Flows Through and Outside Government Structures.” Background paper produced for the World Bank. Washington DC. TANGO (Technical Assistance to NGOs) (2004) “Development Relief.” Report prepared for USAID and FAM. Washington, DC. Ulrichs, M. and Sabates-Wheeler, R. (2018) “Social Protection and Humanitarian Response: What is the Scope for Integration?” IDS, Working Paper 516. Brighton. United Nations (2016) The Grand Bargain: A Shared Commitment to Better Serve People in Need . New York. Winder-Rossi, N., Spano, F., Sabates-Wheeler, R. and Kohnstamm, S. (2017) “Social Protection and Resilience: Supporting Livelihoods in Protracted Crises and in Fragile and Humanitarian Contexts.” FAO, Rome. World Bank (2018) “South-South Learning Forum 2018: Building Resilience Through Adaptive Social Protection—Summary Report.” Washington, DC World Bank (2016) Cash Transfers in Humanitarian Contexts: A Strategic Note. Report pro- duced for the United Nations Inter-Agency Standing Committee. Washington, DC. 46 Social Protection & Jobs Discussion Paper Series Titles 2015-2018 No. Title 1802 Human(itarian) Capital? Lessons on Better Connecting Humanitarian Assistance and Social Protection by Ugo Gentilini, Sarah Laughton and Clare O’Brien, November 2018 1801 Delivering Social Protection in the Midst of Conflict and Crisis: The Case of Yemen by Afrah Alawi Al-Ahmadi and Samantha de Silva, July 2018 1705 Aging and Long-Term Care Systems: A Review of Finance and Governance Arrangements in Europe, North America and Asia-Pacific by Laurie Joshua, November 2017 1704 Social Registries for Social Assistance and Beyond: A Guidance Note & Assessment Tool by Phillippe Leite, Tina George, Changqing Sun, Theresa Jones and Kathy Lindert, July 1027 1703 Social Citizenship for Older Persons? Measuring the Social Quality of Social Pensions in the Global South and Explaining Their Spread by Tobias Böger and Lutz Leisering, July 2017 1702 The Impacts of Cash Transfers on Women’s Empowerment: Learning from Pakistan’s BISP Program by Kate Ambler and Alan de Brauw, February 2017 1701 Social Protection and Humanitarian Assistance Nexus for Disaster Response: Lessons Learnt from Fiji’s Tropical Cyclone Winston by Aisha Mansur, Jesse Doyle, and Oleksiy Ivaschenko, February 2017 1614 Urban Social Assistance: Emerging Insights from Three African Countries by Vanessa Moreira and Ugo Gentilini, December 2016 1613 Issues for Civil Service Pension Reform in Sub-Saharan Africa by Anita M. Schwarz and Miglena Abels, November 2016 1612 How to Target Households in Adaptive Social Protection Systems? Relative Efficiency of Proxy Means Test and Household Economy Analysis in Niger by Pascale Schnitzer, October 2016 1611 Pensions for Public-Sector Employees: Lessons from OECD Countries’ Experience by Edward Whitehouse, October 2016 1610 Pension Systems in Sub-Saharan Africa: Brief Review of Design Parameters and Key Performance Indicators by Miglena Abels and Melis U. Guven, October 2016 1609 Household Enterprises in Fragile and Conflict-Affected States: Results from a Qualitative Toolkit Piloted in Liberia, Volume 2 – Annexes by Emily Weedon and Gwendolyn Heaner, August 2016 1608 Household Enterprises in Fragile and Conflict-Affected States: Results from a Qualitative Toolkit Piloted in Liberia, Volume 1 – Report by Emily Weedon and Gwendolyn Heaner, August 2016 1607 Benefits and Costs of Social Pensions in Sub-Saharan Africa by Melis U. Guven and Phillippe G. Leite, June 2016 1606 Assessing Benefit Portability for International Migrant Workers: A Review of the Germany- Turkey Bilateral Social Security Agreement by Robert Holzmann, Michael Fuchs, Seçil Paçacı Elitok and Pamela Dale, May 2016 1605 Do Bilateral Social Security Agreements Deliver on the Portability of Pensions and Health Care Benefits? A Summary Policy Paper on Four Migration Corridors Between EU and Non-EU Member States by Robert Holzmann, May 2016 1604 Assessing Benefit Portability for International Migrant Workers: A Review of the France-Morocco Bilateral Social Security Agreement by Robert Holzmann, Florence Legro and Pamela Dale, May 2016 1603 Assessing Benefit Portability for International Migrant Workers: A Review of the Belgium-Morocco Bilateral Social Security Agreement by Robert Holzmann, Jacques Wels and Pamela Dale, May 2016 1602 Assessing Benefit Portability for International Migrant Workers: A Review of the Austria-Turkey Bilateral Social Security Agreement by Robert Holzmann, Michael Fuchs, Seçil Paçaci Elitok and Pamela Dale, May 2016 1601 The Greek Pension Reform Strategy 2010-2016 by Georgios Symeonidis, July 2016 1507 Integrating Disaster Response and Climate Resilience in Social Protection Programs in the Pacific Island Countries by Cecilia Costella and Oleksiy Ivaschenko, September 2015 1506 Effectiveness of Targeting Mechanisms Utilized in Social Protection Programs in Bolivia by Ignacio Apella and Gastón Blanco, September 2015 1505 Kyrgyz Republic: Social Sectors at a Glance by João Pedro Azevedo, Paula Calvo, Minh Nguyen and Josefina Posadas, August 2015 1504 Entering the City: Emerging Evidence and Practices with Safety Nets in Urban Areas by Ugo Gentilini, July 2015 1503 Pension Patterns in Sub-Saharan Africa by Mark Dorfman, July 2015 1502 Social Protection in Fragile and Conflict-Affected Countries: Trends and Challenges by Mirey Ovadiya, Adea Kryeziu, Syeda Masood and Eric Zapatero, April 2015 1501 Defining, Measuring, and Benchmarking Administrative Expenditures of Mandatory Social Security Programs by Oleksiy Sluchynsky, February 2015 To view Social Protection & Jobs Discussion Papers published prior to 2013, please visit www.worldbank.org/sp. Abstract Governments in low- and middle-income countries are increasingly investing in social protection, and also address many of their own people’s “humanitarian” needs themselves. For their international partners, who may have an important role in filling gaps when household needs exceed national capacity to meet them, support for the strengthening of national systems— combined with a shift from short-run to more durable approaches—is becoming a unifying framework for assistance. Some aspects of social protection and humanitarian assistance therefore seem to be on a converging trajectory. “Human(itarian) Capital” discusses findings from 12 country case studies exploring the linkages between humanitarian assistance—in its various interpretations—and national social protection systems. Specifically, the paper distills lessons on how humanitarian assistance and social protection systems might better coexist, the possible challenges and trade-offs emerging from practical experiences, and how to facilitate, inform, and accelerate future concerted action. About this series Social Protection & Jobs Discussion Papers are published to communicate the results of The World Bank’s work to the development community with the least possible delay. This paper therefore has not been prepared in accordance with the procedures appropriate for formally edited texts. This paper is a joint paper by the World Bank and the World Food Programme. The findings, interpretations, and conclusions expressed herein are those of the author(s), and do not necessarily reflect the views of the International Bank for Reconstruction and Development/The World Bank and its affiliated organizations, or those of the Executive Directors of The World Bank or the governments they represent, and the World Food Programme. The World Bank and World Food Programme do not guarantee the accuracy of the data included in this work. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgement on the part of The World Bank or World Food Programme concerning the legal status of any territory or the endorsement or acceptance of such boundaries. For more information, please contact the Social Protection Advisory Service at socialprotection@ worldbank.org or visit us on-line at www.worldbank.org/sp. © 2018 International Bank for Reconstruction and Development / The World Bank, and the World Food Programme