Social Safety Nets Social Protection & Labor Policy Note May 2016 | Number 19 Highlight Sorting through the Hype: The objective of this note is Exploring the Interface to map out some key issues relevant to humanitarian and social protection work, that between Humanitarian is, to basically explore their interface. Yet as this note Assistance and Safety Nets argues, they tend to differ in some fundamental aspects. Ugo Gentilini It then briefly reviews the rationale for public interventions, examines how the debate is panning out in different contexts, and Introduction identifies a set of strategic Global interest around humanitarian and social protection issues is and practical questions for remarkable. These feature prominently in some of today’s most complex further work in the area. development challenges – whether supporting poor people in times of crises or providing them with a steady hand while pursuing opportunity. To some extent, there is a common thread of themes that underpin both humanitarian and social protection matters, and which, at its core, is about state formation and its strengthening. In many developing countries, current policy debates seem to have historical roots. These may go back even to pre- and post-colonial institutions, how these interplayed with pre-existing and non-state support mechanisms, and how the ensuing trajectories by newly-formed states were molded by both country and global developments (Ravallion 2016; Fukuyama 2015; Davey et al. 2013). While an in-depth examination of those patterns goes beyond the scope of this note, it is important to underscore that current debates around humanitarian assistance – and those around the formalization of non-contributory transfers by the state – are the result of longstanding and organic historical processes. 1 Policy Note: Social Safety Nets May 2016 | Number 19 Figure 1. Burgeoning terms and concepts   #  !$ #! *!$ '   &'   !       !    !  '  *$ '   ! #!*# $    *++   #  '  # ' # * #!  '     $   ' &# # ! $   &  (  #   %*  ' ' Source: author’s illustration   Against this background, the increasing attention to net programs provide non-contributory transfers in cash or humanitarian and social protection matters are accompanied in-kind to beneficiaries. As such, programs like cash transfers in no small measure by an explosion of terms, definitions, and public works are part of the toolbox in both spaces. Yet as and frameworks (figure 1). On one hand, this is a positive this note argues in the next section, they tend to differ in some development as it signals the diverse nature of actors and fundamental aspects. We then briefly review the rationale for themes at the interface of those two agendas; on the other public interventions, examine how the debate is panning out in hand, it can make the issues at hand somewhat elusive and different contexts, and identify a set of strategic and practical fragmented. questions for further work in the area. The objective of this short note is to map out some key issues relevant to humanitarian and social protection work, that What’s different? is, to basically explore their interface. In doing so, we will focus on international humanitarian aid (as distinct from While programs may look similar in practice, humanitarian national humanitarian assistance) and social safety net or and safety net approaches differ in two core aspects. One key social assistance programs (as opposed to the wider social difference revolves around objectives. The overall environment protection system). Operationally, humanitarian and safety in which humanitarian actors operate is often volatile, Figure 2. Recipients of humanitarian assistance, 2010-14 3%          Host governments International organizations, NGOs, others  Source: data from World Bank ASPIRE database, ALNAP (2015), and GHA (2015) 2 Policy Note: Social Safety Nets May 2016 | Number 19 chaotic, and posing immediate threats to the lives of affected embedded in national budgets. Such contrast may create populations. As such, programs tend to be designed with an an important bifurcation in approach, including in terms emphasis on life-saving purposes, including delivering under of engagement, ownership and role of host governments short time horizons, in line with humanitarian principles, and (when they exist) in shaping, managing, and programming putting a premium on safety and timeliness of assistance as key interventions. operational principles. Yet safety net approaches are not homogeneous. In a stylized Conversely, social assistance programs tend to adopt a way the agenda in some contexts includes a particular emphasis longer-term view and are geared to help address chronic on ex-ante preparedness and household asset-strengthening, poverty in its many dimensions. Safety nets are a component such as in the drylands of the Sahel belt (Del Ninno et al. of broader social protection systems, with the latter not 2016). This is also the case in the Horn of Africa where this is only encompassing social assistance (i.e., non-contributory combined with investments in scalability of systems to deal transfers), but also including social insurance (i.e., contributory with recurrent shocks. The flexibility of systems, and how these transfers) and various labor market interventions. Social could be leveraged for humanitarian purposes, is a central assistance is often enshrined in legislation, integrated in feature of managing large-scale natural disasters in contexts sectoral policies, financed out of domestic budgets, and like East Asia (Cabot-Venton et al. 2015). The effects of represents a cornerstone in government-citizens social hostilities in Syria are likely to shape the debate in the Middle contracts. East for years to come, including the thorny issue of serving affected and non-affected people ‘living side-by-side’ in urban Perhaps the most relevant distinction, however, revolves areas (UNHCR 2014). Finally, economic crises are likely to around the role of national governments. In 2014, only 3 play a major role in Latin America, including how to provide percent of total humanitarian funds was channeled through support and ring-fencing safety nets during fiscal downturns. national governments (figure 2). Almost the entire volume of assistance was directed outside government structures, including an estimated 4,480 actors, organizations, and Crises and (in)ability to manage them agencies. In part, this model of operating outside or with limited engagement of government systems can be dictated by A simple taxonomy of crises may envisage three basic types, several constraints, including conflict, swiftness of response, including economic shocks, natural disasters, and violence- and legislation. related shocks. Within these broad types, shocks can be more or less predictable: for example, typhoons may be more Yet such an approach represents a sharp departure from that predictable than earthquakes; employment and fiscal shocks of safety nets. This is typically provided through governments, from economic reforms can be more predictable that sudden, including via blends of domestic resources, concessional sharp rises (or collapse) in international commodity prices; and financing, and medium/longer-term programmatic frameworks while conflicts are somewhat foreseeable, they can be difficult Figure 3. Global crises patterns1                                                          %"" %"" %"" )" )" )"    #""" #""" #"""                                        '+' '+' '+' *)" *)" *)" $'" $'" $'" (" (" (" *"" *"" *"" (*" (*" (*" '" '" '" $"" $"" $"" ("" ("" ("" #&#+ &" #&#+ &" &"    #&#+ #'" #'" #'" %" %" %" %)" %)" %)" &"" &"" &"" %#" %#" %#" #"" #"" #"" $" $" $" $"" $"" $"" '" '" '" #" #" #" " " " " " " " " " $""' $""' $""( $"") $""' $""( $""( $""* $"") $""+ $""* $"#" $"") $""+ $"## $""* $"#$ $""+ $"#" $"#" $"## $"#% $"## $"#$ $"#$ $"#& $"#% $"#% $"#& $"#& $""" $""" $""" $"'" $"'" $"'"                         Source: UNHCR (2014); CRED EM-DAT database (accessed January 2016); ICRC (2010) 1  e sample includes countries for which data exists for safety nets coverage, government revenue capacity, and humanitarian assistance. As such, the graph should Th be read in tandem with figure 6. 3 Policy Note: Social Safety Nets May 2016 | Number 19 to fully predict. Yet their consequences, like protracted to raise revenues beyond external grant assistances. displacement, are a predictable outcome of conflict. So how are countries responding to growing needs? They are In terms of natural and man-made crises, the world is set to both increasingly investing in safety nets and calling for more face two major shifts: first, over the past decade the number immediate humanitarian assistance. In general, data from of people affected by conflicts has doubled, reaching about 120 developing countries show that they allocate an average 60 million people in 2014. Conversely, over the past 5 years of 1.6% of GDP on social assistance (World Bank 2015). This the number of people affected by natural disaster has halved, is a growing trend: for 32 low and middle income countries although still affecting over 140 million people (figure 3). with panel data, spending on safety nets has grown by one Man-made crises are protracted for long spells of time percentage point between 2000 and 2010-11. In 2014, safety (on average 17 years in the case of displaced populations). nets spending included a total volume of about US$329 billion, In those contexts, safety nets can be hampered by low- a level that dwarfs that of humanitarian aid flows; between 2008 capacities, wiped out by conflicts or, in the case of refugees, and 2014, humanitarian funding requirements almost doubled, hindered by legal constraints in serving foreigners. reaching US$19.5 billion. Yet only US$12 billion were actually met, which marked a record absolute volume received and the In terms of a second global shift, whether natural or man- highest share of unmet requirements (38%). made, crises are becoming more spatially-concentrated, including presenting an urban dimension. For example, In some situations, humanitarian action provides the core of compared to the 2000 level, the risk of major disasters will support to beneficiaries, with social assistance representing affect an additional 60-190 million urban people by 2050; at a minor fraction of assistance compared to humanitarian the same time, more displaced people reach cities (within volumes: for example, the latter is 7.5 times higher than the or outside national borders) in an attempt to escape civil former in the case of Mali – and 14 times so in Sudan. In other conflicts and sectarian wars, such as in Lebanon and DRC. contexts, the level of humanitarian and social assistance can be more levelled, like in Lebanon or Palestine, hence more clearly In all the above situations, it becomes key to understand illustrating the emergence of two parallel systems in providing whether countries have enough capacity to meet shock- assistance (figure 5). generated needs. The key dilemma is that countries with lower institutional and administrative capacity are hard-hit Taken together, the ability to cover the poor with safety nets by shocks and display lower coverage of national safety and humanitarian assistance shows the emergence of some nets. For example, based on data from 38 countries , figure broad typology of contexts. Based on countries with data on 4 shows a statistically significant correlation (R2=0.47) both issues, figure 6 illustrates the emergence of three clusters: between social assistance coverage and government’s one includes countries with relatively well-performing safety domestic fiscal capacity. The latter is measured by the non- net coverage and receiving comparatively little humanitarian grant revenue per capita, which measures countries’ ability aid. This encompasses contexts that are exposed to shocks, but that have a basic platform to address part of the generated needs. This may include the Pantawid program in the Figure 4. Safety nets coverage of the poor Philippines, the PDS in Iraq, and emerging systems in countries and fiscal capacity2 like Lesotho, Kenya, and Nepal. In those contexts, it may be possible that humanitarian assistance will play an important MNG role, although growing investments in system-enhancements RWA IDN BWA may reduce such need over time. BOL MEX UGA GTM PRY PAN Another cluster includes countries with low social assistance VNM SWZ coverage and relatively high volume of humanitarian aid. This LSO encapsulates countries mired in protracted and multiple crises, NIC PHL EGY GEO GAB such as Afghanistan, DRC, Haiti, Yemen, and Zimbabwe. A MAR LKA DOM core challenge in such contexts may be to find ways for SLE KEN ARM aligning incentives and plans between agendas in highly MRT BGD DJI CPV NAM TMP volatile contexts, including being able to operate and ZWE PAK MWI TZA TJK YEM BLZ work with ‘fragile’ governments and deeply ingrained ETH AFG MOZ SEN GHA KGZ FJI humanitarian approaches. MDG NGA FSM CAF GMB KHM COG ZMB CMR CIV Finally, there are countries that are neither covered by safety nets nor are receiving substantial humanitarian support. These include various African countries, such as Congo, Senegal, Nigeria, Gambia, and Cameroon. In those Source: data from World Bank ASPIRE database and Development Initiatives (2015) cases, there might be benefits for building systems ‘from 2 The figure includes countries with humanitarian assistance of at least $0.5 percapita and less than $40. 4 Policy Note: Social Safety Nets May 2016 | Number 19 Figure 5. Volume of resources as percentage of GDP, latest available estimates        !     $  $                                    Source: data from World Bank ASPIRE database, ALNAP (2015), and GHA (2015) scratch’, especially among middle-income countries. Yet using national administrative platforms (e.g., Philippines), that is also presents a limitation in terms of bedrock of and ‘shadow alignment’, whereby a parallel humanitarian capacity: consider, for example, that Ethiopia’s Productive system tries to converge as much as possible with safety Safety Net Program was born by transforming pre-existing net programs (e.g., Kenya). Moreover, the new ways in humanitarian assistance, which included a considerable which the spheres interact, like the Lebanese example of a level of experience that was harnessed to forging the humanitarian voucher program for Syrian refugees being safety net (Wiseman et al. 2010). extended to Lebanese citizens, speak to the dynamic nature of the humanitarian-safety net iterations. These, nonetheless, All these different scenarios suggest a variety of ways tend to occur between worlds that are still largely apart. in which the humanitarian-safety net interactions may occur. For example, from a humanitarian standpoint, CaLP (2016) laid out several types of linkages during A crowded playing field crises, like ‘piggybacking’ of humanitarian assistance by The coexistence between humanitarian assistance and safety nets emerges as a systematic feature across the vast majority Figure 6. Safety nets coverage of the poor of crises and contexts. In some respect, such interface is and humanitarian spending3 perhaps one of the few common narratives underpinning differing shocks and situations. But why is it relevant for the safety net world? It is so from a financing, strategic, and 100 MNG operational standpoint. Safety Net Coverage of the Poorest (%, MRV) RWA IRQ From a financing perspective, a growing pool of donors seem 80 UGA SWZ to promote the idea of ‘transferring’ humanitarian caseloads UKR LSO LBR onto social protection systems. This implicitly conveys the 60 TUR PHL message that some of the current humanitarian functions NPL LKA could be better served by safety nets, including doing so more effectively and efficiently. This is a highly plausible 40 KEN BFA SLE MRT DJI assumption which, as we’ll discuss in the final question, TUN PAK ZWE would need to be further strengthened with quantitative 20 ETH MWI YEM TJK KGZ AFG evidence. ZAR SEN CIV CAF SLB COG NER HTI From a strategic standpoint, the theme is likely to play a 0 GMBCMR 0 10 20 30 Humanitarian Aid ($ per capita, av. 2010-15) 40 pivotal role in the shrinking pool of countries that, in just a few years, may constitute the core of the low-income country Note: Excluded Humanitarian Aid per capita less than $0.5 and greater than $40 base. Standing ready to engage with ‘fragile’ partners would Source: UNHCR (2014); CRED EM-DAT database (accessed January 2016); be an important element in such process. This, of course, ICRC (2010) is in addition to the increasing number of middle-income countries that are exposed to disaster and displacement risks. 3 Note: Excluded Humanitarian Aid per capita less than $0.5 and greater than $40 5 Policy Note: Social Safety Nets May 2016 | Number 19  Figure 7. Trends in in-kind food aid and humanitarian cash transfers                                          Source: CaLP-Cash Atlas and WFP-FAIS online databases (accessed January 2016) Looking at the issue with an operational lens, it is clear that economic activity, and technology are creating a landscape humanitarians may operate in similar communities and that is increasingly conducive for cash as an appropriate using same instruments (e.g., cash) as deployed under a humanitarian response. This would entail a careful social safety net. Those programs, however, may be served balancing between recalibrating the overall composition by different nuts and bolts mechanisms (e.g., targeting of humanitarian assistance and providing appropriate methods, institutional arrangements, MIS, M&E), which responses (Gentilini 2016). may fuel overall fragmentation. There are emerging It is in this context that a wider use of government-led practices from the Sahel (e.g., Senegal, Mauritania) and safety nets, whether nascent or developed, emerges as a key Middle East (e.g., Lebanon) showing that part of the issue, and sometimes a bottleneck, in the humanitarian architecture underlying safety net programs could be discourse. In other words, reconciling humanitarian used – and indeed be strengthened by – the humanitarian and safety net approaches would in large part hinge on community. Yet the challenge around using existing incentivizing and strengthening national systems as the national systems remains a key constraint (World Bank ‘default’ option. 2016). For instance, more and more actors are involved in the provision of humanitarian cash transfers as complements or alternatives to in-kind food. In 2014, they totaled a volume of US$1.2 billion and represent a growing line of work: as of 2012, cash reached 6.8 million beneficiaries (and 5.6 million in 2014), up from 2.4 in 2000. Over the same period, the quantity of global food aid halved, dropping from 10.9 to 4.7 million metric tons (figure 7). Key actors in the humanitarian cash arena include both private charities like Give Directly, as well as more institutionalized actors and agencies. In Lebanon, for example, cash was provided by 30 agencies for 14 objectives (ODI 2015). Yet, only 6% of total humanitarian assistance is devoted to cash (UN 2016). Although in-kind assistance will continue to be a strategically important component of humanitarian assistance in the years to come, the fundamental long-term trends in concentration of people, 6 Policy Note: Social Safety Nets May 2016 | Number 19 Deepening the agenda There has been tremendous progress around understanding when and how the humanitarian and safety nets worlds could be better bridged. Yet a number of key questions remain somewhat pending or partially addressed. A selection of these might include the following: ■■ Developing an operational definition of ‘humanitarian’ and ‘safety net’ programs; ■■ Map out the functions that are ‘purely’ humanitarian, and those that are currently performed by humanitarian actors, but could, potentially, be undertaken by safety nets; ■■ Where functions overlap, examine the comparative costs of humanitarian and safety net models for providing assistance to a comparable set of beneficiaries and interventions; ■■ Identify if and under what circumstances protracted humanitarian assistance may hinder national systems development; ■■ Examine the conditions under which humanitarian and safety net programs could better coexist and complement each other; ■■ Document cases of joint programming, including technical specifications and political economy dynamics, as they emerge from live case studies. The deepening of such a critical agenda – one at the interface of two significant areas of work such as humanitarian assistance and safety nets – would greatly benefit from a more nuanced and operationally-oriented analysis which we hope some of the above question may help pursue. 7 Policy Note: Social Safety Nets May 2016 | Number 19 References ODI (2015) Doing Cash Differently: How Cash Transfers Can Transform Humanitarian Aid. London. ALNAP (2015) The State of the Humanitarian System: 2015 OPM, ODI, CaLP, and Inasp (2015) DFID Shock-Responsive Edition. London. Social Protection Systems Research. Oxford. Cabot Venton, C., Bailey, S. and Pongracz, S. 2015. Value for Ravallion, M. (2016) The Economics of Poverty: History, Money of Cash Transfers in Emergencies. London. Measurement and Policy. Oxford University Press. New York. CaLP (2016) Working with Cash-Based Safety Nets in UN (2016) Too Important To Fail – Addressing the Humanitarian Contexts: Guidance Note for Humanitarian Humanitarian Financing Gap. Report to the Secretary-General Practitioners. Oxford. by the High-Level Panel on Humanitarian Financing. New Davey, E., Borton, J. and M. Foley (2013) A History of the York. Humanitarian System: Western Origins and Foundations. UNHCR (2014) Annual Global Trends: Forced Displacement in ODI, HPG Working Paper. London. 2014. Geneva. Del Ninno, C., Fallavier, P. and S. Coll-Black (2016) Protecting Wiseman, W., Van Domelen, J. & Coll-Black, S. (2010) the Vulnerable in the Drylands of Sub Saharan Africa: The Role “Designing and Implementing a Rural Safety Net in a Low of Social Protection. World Bank. Washington DC. Income Setting: Lessons Learned from Ethiopia’s PSNP 2005- Development Initiatives (2015) Investments to End Poverty 2009”. World Bank. Washington DC. 2015. United Kingdom. World Bank (2016) Strategic Note: Cash in Humanitarian Fukuyama, F. (2015) Political Order and Political Decay: From Contexts. Washington DC. the Industrial Revolution to the Globalization of Democracy. World Bank (2015) The State of Social Safety Nets 2015. New York. Washington DC. Gentilini, U. (2016) The Humanitarian Dividend: The Comparative Evidence of Cash and In-Kind Transfers in Humanitarian Situations. World Bank. Washington DC. Forthcoming. GHA (2015) Global Humanitarian Assistance Report 2015. Bristol. Humanitarian Outcomes (2015) Global Database of Humanitarian Organization. London. OCHA (2014) World Humanitarian Data and Trends 2014. New York. 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. The World Bank does not guarantee the accuracy of the data included in this work. 8 © 2016 International Bank for Reconstruction and Development / The World Bank