Bridging Humanitarian Responses and Long-Term Development through Transformative Changes — Some Initial Reflections from the World Bank ’ s Adaptive Social Protection Program in the Sahel

In the context of increasing climate-related extreme events and other crises, the concept of adaptive social protection (ASP) has been recognized as a potentially effective policy response to reduce the impacts of these shocks and stressors on vulnerable households. The concept is currently being tested at scale by the World Bank in six countries in the Sahel region. Based on conceptual considerations, this paper aims to address three questions: How and to what extent can adaptive social protection be considered transformative? Where does this concept sit along the humanitarian–development continuum? And, how does it relate to resilience? To answer these questions the paper draws on the authors’ exposure to the on-going World Bank ASP program, as well as documents derived from the emerging body of literature on climateand shock-responsive social protection. Drawing on these different materials the paper first demonstrates that ASP can effectively be considered as a transformative intervention at two different levels: at the system level and at the beneficiaries’ level. The paper also shows how, through its activities designed to strengthen households’ adaptive capacity, an ASP program can contribute to building resilience beyond the short-term coping strategies which humanitarian interventions generally focus on. As such ASP covers a larger spectrum along the humanitarian–development continuum than most other interventions proposed in the context of shock-responsive interventions.


Introduction
It is now well established that climate change will increase hazards, risk, and people vulnerability-thus raising significant challenges to poverty reduction and development [1].It is estimated, for instance, that climate change could result in an additional 100 million people living in extreme poverty by 2030, the majority of whom are from low-income countries [2].Climate extremes impacts mainly populations who lack the resources and capacity to anticipate, respond to and recover from shocks [3].In many cases, the impacts of these climate-related events will be exacerbated by demographic drivers such as urbanization and migration.Thus, efforts to address poverty and vulnerability will need to acknowledge not only the emerging challenges affecting the livelihoods of the poor in in the face of climate change, but also take into account the socioeconomic factors that make people vulnerable in the first place [4].
In that context, policy responses that aim to address the underlying causes of poverty and vulnerability and are able to reduce the increasing risk of climate shocks and the impoverishing effects of disasters are gaining more traction within both donors and practitioners' communities.Amongst these, social protection is one policy tool that has proven to effectively protect people's livelihoods from major shocks.Social protection describes "all public and private initiatives that provide income or consumption transfers to the poor, protect the vulnerable against livelihood risks, and enhance the social status and rights of the marginalised; with the overall objective of reducing the economic and social vulnerability of poor, vulnerable and marginalised groups" [5] (p.9).
Scholars have emphasized the importance of social protection as a means of protecting the most vulnerable members of society from the impacts of climate change [6].Those propositions derive essentially from advances in academic research made in recent years which demonstrate the promise of certain conceptual frameworks and approaches that explore the synergies between social protection, climate change adaptation, and disaster risk reduction [7][8][9][10][11].Those include the Adaptive Social Protection framework [7,12]; the Climate-Responsive Social Protection framework [13]; or more recently, the Shock-Responsive Social Protection approach [14], which all have advanced the understanding of the commonalities and differences between social protection, climate change adaptation, humanitarian interventions, and disaster risk reduction, and their respective and synergetic contribution to strengthen the resilience of people to shocks and disasters.Some of those conceptual frameworks are now being tested at scale.This is the case, in particular, of the concept of Adaptive Social Protection (ASP) which is currently piloted in six countries in the Sahel region (Burkina Faso, Chad, Mali, Mauritania, Niger, and Senegal).This ASP pilot component, which is a 5-year UK£50 million initiative funded by the UK Department for International Development (DfID), is part of a larger IDA (International Development Association) project on Social Protection implemented by the World Bank in these six countries.The ASP component started in 2014 and is to continue until 2019.Although it is too early to draw definitive conclusions on its effective impact on the beneficiaries of the IDA program, its current evaluation implemented by an independent team offers some initial reflections regarding the role that such a program can play as a vehicle to integrate development and humanitarian response and help tackle disaster and climate-related impacts.More specifically, and in line with the special issue on "Transforming Development and Disaster Risk" of this journal, we are interested in this paper to address the following three questions: How and to what extent can ASP be considered transformative?Where does this concept sit along the humanitarian-development continuum, and how does it relate to resilience?
To answer these questions, we will draw essentially on our own exposure to the on-going ASP program (as members of the team responsible for its evaluation), as well as other documents, reports and articles published on the concept of ASP.Additional insights derived from the emerging body of literature on climate-and shock-responsive social protection will also be included in the discussion.Beyond this specific literature we will also need to review and clarify the meaning of a couple of concepts, including transformation, resilience, coping strategies and adaptive strategies.The review of these concepts will constitute the initial steps of this analysis.
The rest of this paper is structured as follows.Section 2 presents the methodology applied to conduct this analysis.In Section 3, both the origin of the concept of ASP and the current on-going ASP pilot implemented by the World Bank are presented.Section 4 discusses the different dimensions of the distinction between coping strategies and adaptive responses in a way that provides useful element for the subsequent discussion on transformation.Section 5 reviews the main interpretations of transformation found in the different parts of the relevant literature and highlights the two definitions which will then be used in the next section to discuss the extent to which ASP can be considered as a transformative intervention.Section 6 starts with the more detailed presentation of the five design principles which are recognized to be necessary for a social protection system to become effectively adaptive, and then aims to demonstrate why ASP can effectively be considered as a transformative intervention at two different levels: at the system level and at the beneficiary level.Section 7 completes the analysis by exploring the relation between ASP and resilience, using a conceptual framework and builds on that framework to determine where ASP lies along the humanitarian-development continuum.Section 8 concludes.

Research Design and Methodology
This work draws significantly on insights derived from the evaluation of the World Bank ASP pilot program.The evaluation exercise was carried out over a 20-month period (from March 2016 to October 2017) in four out of the six countries where the ASP program is operating (Mali, Mauritania, Niger, and Senegal).In each of these four countries several series of face-to-face interviews complemented by on-line semi-quantitative surveys and reviews of project reports were used to gather information about the ASP program and the general institutional context within which it operates.In addition, working papers, reports and peer-reviewed articles on the concept of ASP and shock-responsive social protection were collected and critically reviewed.For the specific purpose of this paper, the literature review was then extended to the concepts of 'transformation' (when can an intervention be said to be transformative?)and 'resilience' (how is resilience conceptualized and understood in the specific context of humanitarian and/or food security interventions-as opposed to the much wider literature discussing resilience in other domains such as engineering, ecology, socio-ecological system, or psychology).Finally, two other concepts which are central to our discussion in relation to the question of transformation will also be reviewed-those of coping strategies and adaptive strategies.In clarifying and discussing all these different elements, the discussion will essentially be conceptual, even if, as mentioned above, it initially derives from the on-going ASP program being implemented by the World Bank in the Sahel.

Origin of the Concept
The term Adaptive Social Protection was first cast at the end of the 2000s by researchers from the Institute of Development Studies in UK who realized that social protection, disaster risk reduction and climate change adaptation were three communities of practices that had evolved from different origins but were all linked by the same fundamental concern of reducing vulnerability and building resilience-be it to chronic poverty (social protection), disasters and extreme events (disaster risk reduction) or long-term changes in climate conditions and their distribution over time and space (climate change adaptation).Until then, however, little cross-fertilization had been observed between these three communities which seemed to work essentially in silos, ignoring or being unaware of their commonalities and overlapping agendas, or being unable to overcome institutional constraints or poor communication that prevented them from working together [15].
The idea of adaptive social protection (ASP) was born there.That idea was then further pursued in an effort to elaborate a more rigorous concept that could support and facilitate the blend of social protection programs, climate change projects, and disaster risk reduction interventions, at both policy and practice levels [7,12,16].By bringing together the objectives of these three streams of work, the ambition was to provide a framework that would help social protection interventions become more resilient to risks from disaster hazards and climate change, and at the same time help understand how social protection, through its vulnerability reduction focus, could play a critical role in reducing/buffering the negative impact of climate change and disaster [17].DfID, which had followed closely and funded some of these initial reflections (see [18]), then become keen to push further the experience and test the concept at scale.They subsequently approached the World Bank and contracted the Washington-based institution to implement an Adaptive Social Protection program in the Sahel where the World Bank was already supporting the governments of six countries in their effort to develop national-scale safety net programs (In this paper the term social protection program and social safety net program are used interchangeably even if some social protection experts would argue that the two are somewhat different (see e.g., [5])).The idea was to add an ASP component to these safety net programs to ensure that they would become truly adaptive.

The World Bank ASP Pilot Program
The objective of the World Bank ASP program is "to increase access to effective adaptive social protection systems for poor and vulnerable populations in the Sahel" [19] (p.5).Across the six countries where the ASP pilot is being implemented (Burkina Faso, Chad, Mali, Mauritania, Niger, and Senegal), governments have put in place national safety net systems that are providing support to poor and vulnerable households (except for Chad, where such safety net program is currently been established).Funds from the ASP Program are then used to support activities to strengthen or expand these national systems.In particular the ASP Program is structured around four main inputs-(1) knowledge activities; (2) technical assistance; (3) support to government programs; and (4) regional activities and strategic staff deployment-with the objective to create evidence, experience, and learning on how social protection programs can be most effectively designed to enable poor and vulnerable households to anticipate, absorb, and recover from shocks.For example, the ASP program is building a knowledge base on productive activities which are expected to help move poor people out of poverty.Technical assistance is helping governments strengthen the design and delivery of public works programs in a manner that is expected to build resilience to shocks, while grants provided to governments are enabling the six countries to put in place adaptive safety net systems and build experience in providing cash transfers to poor and vulnerable households.Finally, World Bank staff deployment in countries is expected to enhance technical support provided to governments [19].
As mentioned earlier, the World Bank ASP program is still ongoing for at least another year and it is too early to be able to determine whether these different activities will effectively lead to the creation of a truly adaptive social protection system in the countries where the ASP program is being implemented.What can be discussed on the other hand is whether the principles underlying the ASP program are of transformative nature.For this we need first to clarify the meaning of several concepts which are sometimes used and understood differently in the literature.Those include coping and adaptive strategies, as well as transformation.

Distinguishing Coping Strategies from Adaptive Strategies
Within both the literature on disaster risk management and climate change, there is already a large body of work that has discussed the meaning and definitions of both coping and adaptive strategies.The objective of this section is not to review this very rich literature (see e.g., [20,21] for some early reviews) but rather to summarize it in a way that provides useful element for our following discussion on transformation.We are indeed intended to show that the shift from coping to adaptation can in itself be considered as a form of transformation.
While a distinction was made early in the literature between short-term coping and longer term adaptation (see, e.g., [22][23][24][25]), this distinction was based on a time criterion (short-term versus longer-term) left some other scholars-especially those interested in the role of institutions in the adoption of coping or adaptive behaviours) unsatisfied.This led them to propose another possible dimension in this distinction, which this time considers coping as those responses individuals, households or communities can adopt to respond to an event within the existing (social, economic, technological) structures of the system, while adaption has to do with the change/evolution of the structure, functioning and organisation of the system in question [26,27].More recently, attempts have been made to further integrate these two temporal and institutional aspects.For [28] for instance, coping includes short-term strategies and actions undertaken within existing institutional settings, whereas adaptive capacity is associated with longer term strategic actions that may require institutional change.
Interpreting this discussion by claiming that coping is essentially the concern of short-term disaster risk management interventions or even humanitarian interventions while adaptive strategies is mainly related to longer-term climate change adaptation would be an oversimplification.Yet it is also correct to point out that adverse events such as floods and droughts have traditionally been addressed through disaster risk management and humanitarian interventions, while longer-term changes such as increase in frequency of adverse events or in seal level changes have been mainly addressed through adaptation [29].Paradoxically the need to differentiate between the two concepts emerge in the literature because of the increasing inclusion of disaster risk reduction considerations into climate change and vice versa [30,31].
This last remark does not mean however that the two concepts are totally exclusive.It remains unclear for instance whether rural communities seeking to reduce their vulnerability to current climate variability through coping strategies are-or are not-also fostering their capacity to adapt to future climate change.Some have indeed argued (e.g., [32]) that, in order to take place, adaptation needs to be able to rely on strong institutions and some level of assets accumulation, which implies individual and collective abilities to maintain stability and therefore to engage in coping strategies to absorb shocks in the first place.
Overall, however the different communities of experts involved in the discussion (geographers, social scientists, etc.) seem to agree that even if the two concepts are linked and their respective scope do overlap in some cases, the conceptual distinction between coping and adaptive strategies is legitimate and valid.This conceptual distinction will certainly become useful in our case to show why adaptive social protection-in contrast to other related concepts such as shock-responsive social protection-can be considered as promoting transformative changes.

Definition of Transformation
Transformation is perhaps even more problematic to define than coping strategy or adaption.In some cases, transformation is understood as physical and/or qualitative changes in form, structure or properties of a system [33,34].In some part of the socio-ecological literature, 'transformability' is defined for instance as the "capacity to create a fundamentally new system when ecological, economic or social structures make the existing system untenable" [35] (p.5).In that case, changes are said to be 'transformative' because they result in modifications of the system's structure and function and often lead to some shifts in the nature of the system, either through the introduction of new state variables or the loss of others, such as when a household embarks in a totally new livelihood strategy (e.g., giving up its pastoralist life to embrace a farming one).
In other instances, research on transformation (for instance within human geography) appears to be largely embedded in adaptation research, particularly within studies that refer to the concept of resilience [30,33,36].This part of the literature distinguishes between incremental adjustments and transformational responses, implying that transformation is not just about shifts in the nature or properties of the system, but also in the way those changes are happening (from incremental to transformational).
Transformation is also sometimes understood as a more profound social process where changes in attitudes, norms and values are involved.For instance, the end of slavery or the right of women to vote would be good examples of transformational changes understood in this sense.In that case the main challenges related to transformation are generally not technical or technological.Instead, the challenges lie in the fact that these shifts may include some form of institutional reorganizations, behavioral adjustments or cultural changes; they would then often led to changes in values, beliefs, identities and stereotypes.In other words, they challenge the status quo (e.g., [30,37,38]).
One puzzling thing about transformation is that, although there is agreement that coping, adapting and transforming are somehow placed at different loci along a gradient of increasing scope, scale, and nature of change (e.g., [39,40]), we can also find propositions where transformation is used to characterize or distinguish coping from adaptive strategies.Berman and her colleagues for instance talk about "the transformation of coping capacity into adaptive capacity" [41] (p.91).In another instance, Kates et al. refer to "Transformational adaptation when incremental adaptations to climate change are insufficient" [42].Their argument is that so far adaptation to climate change has mainly been envisioned as in an incremental way in order to avoid disruptions of systems at their current locations.These authors argue however that in some cases, vulnerabilities and risks may be so sizeable that they require transformational rather than incremental adaptations [40,42].
In sum, transformation as a concept means different things to different people and is used for different purposes (to describe a type of change or as a way to characterize the nature of changes), and it is not always clear what exactly needs to be transformed and why, whose interest these transformations serve, and what will be the consequences.
For the purpose of this paper, and for the sake of remaining conceptually focused, we will refer to transformation in the rest of this paper in only two ways: (1) Changes in the system that require more than just incremental adjustment, in particular transformation that refers to change in the institutional and operational structure of the system.In our case this will concern the social protection system itself (when it becomes adaptive); and (2) Transformations from short-term to longer-term aspiration (from the reactive coping strategies to the longer-term adaptive responses).This second type of transformation will be useful when we discuss the place of ASP along the humanitarian-development continuum.
In adopting these two interpretations of transformation we are not putting any special emphasis on the political element of the concept and the idea that transformation may refer to changes that challenge the status quo.The empirical experience we had while evaluating the World Bank on-going ASP program suggests however that acknowledging the 'political economy' of change is also important.

The Five Key Principles of ASP
The literature identifies five key-principles which are deemed to be necessary for a social protection system to become effectively adaptive.Those are: (i) Promote institutional coordination of social protection agencies with relevant climate change and disaster risk management actors; (ii) Ensure the scalability of the program in the face of an extreme event; (iii) Target households that are most vulnerable to natural hazards and climate change related risks; (iv) Design the program so that it increases the adaptive capacity of households and communities; (v) Ensure the swiftness of the response by e.g., operating through an existing SP program [43] (pp.[11][12]. Principle 1. Promote institutional coordination.Because adaptive social protection brings together social protection, climate change adaptation, humanitarian response and disaster risk reduction thematic areas from various ministries, donors and NGOs that have traditionally operated as separate technical disciplines, focused on different sets of risks and target groups and (within government) have usually reported to different and uncoordinated line ministries, there is a wide consensus that one of the most important elements to establish an effective ASP program is to foster collaboration and cooperation between sectors [12,[44][45][46][47].
Principle 2. Ensure the scalability of the program.The 'scalability' of a program refers here to their ability to expand either horizontally (in relation to the number of households to be included in the program) and/or vertically (increase of the value or duration of an intervention to meet beneficiaries' additional needs) in the face of adverse events.In the literature on ASP (e.g., [12,16]) but also on shock responsive social protection (e.g., [11]), the question of scalability is presented as a central requirement of the system.OPM [48] (p.17) for instance remarks: "A key feature of shock-responsive social protection programs is their 'scalability"'.
Principle 3. Target households vulnerable to shocks and disasters.The third ASP principle is about the ability of the program to specifically target households that are vulnerable to climate change and natural hazards (e.g., [14,49]).The rational for the inclusion of this third principle as an ASP design principle is the recognition that, while the usual social protection programs are conventionally expected to focus on chronically poor and vulnerable groups such as elderlies, orphans, handicapped people or households recognized to be chronically poor, an ASP program needs to expand the pool of potential beneficiaries to those who are affected by adverse events such as natural disasters or climate change [11,45,46].Those can but are not necessarily the same household than those originally enrolled in the safety net program.
Principle 4. Foster adaptive capacity.Principle 4 refers to the need for a social protection program, in order to become adaptive, to include activities and interventions that foster or build the adaptive capacity of its beneficiaries.The rationale behind this principle is the assumption that increasing the resilience of poor households and strengthening their adaptive capacities through appropriate interventions should be a core component of any ASP intervention.In particular adaptive capacity is critical to move from recurring shock response and safety nets to longer-term sustainable livelihoods [9,50].
Principle 5. Ensure the swiftness of the program.In an ASP context, Principle 5 on swiftness refers to the capacity of an ASP program to disburse transfers (cash or other forms of support, e.g., food) in a relatively short period of time just after the occurrence of the shock the program is intended to respond.The main justification for this is the well-documented fact that an early response generally produces a better outcome than the same response delivered later (see, e.g., [51]).Swiftness-or also referred to as 'timeliness' in the shock-responsive literature-is therefore a critical element of an effective ASP program and the consensus around the importance of this swiftness is strong [11,14,45,52].

Transformative Change at the System Level
Several of the principles presented above clearly imply some fundamental changes in the social protection system and as such relate easily to the first dimension of transformation we proposed in the earlier section of this paper.We recall that this first dimension refers to "changes in the system that require more than just incremental adjustment, in particular transformations that require change in the institutional and operational structure of the system".This aspect concerns ASP Principles 2, 3 and 5. Amongst these, Principle 2 and 5 are somehow related or comparable in the nature of their requirements, and will be discussed first, followed by Principle 3.
Becoming scalable (Principle 2) is an important change in the property of a social safety net system, which certainly qualifies it as being transformational-as it requires some structural alterations of the system to accommodate either the vertical or the horizontal (or sometimes both) types of expansions.Not too surprisingly the empirical experience we collected from the World Bank ASP program reveals that such systemic changes are not necessarily easy to implement especially when the program which is expected to acquire this property has been operating as a non-scalable (conventional) social safety net program before, with a predetermined and constant number of beneficiaries-as it is the case for the countries included in the ASP program: in all these countries the safety net program which has been chosen to embrace the ASP component were already existing and had been operating for several years prior to the beginning of the ASP program (except in Chad).This situation means that we are in scenarios where the "A" (the adaptability) is somehow added in an ad-hoc manner to an existing SP program.As one may imagine trying to inject ex-post some degree of flexibility into a system which has already pre-existing structure and procedures may not always be easy.This reality may explain partially the relatively slow progress made so far on this part of the ASP program in the Sahel.In comparison the Hunger Safety Net Program in Kenya which is routinely cited as a prime example of a program with a flexible component that allows for horizontal expansion in an emergency, was designed as a permanent safety net for chronically vulnerable households, but had procedures and contingency funds put in place from the start, which would allow it to become an effective response mechanism in case of emergency and built anticipatory capacity at the systems level.
The second principle which we claim is also of a transformational nature in an ASP intervention is Principle 5-the ability of the system to respond rapidly and in particular to reach as quickly as possibly the populations who are in need after a disaster/shock.The literature has shown that this property of swiftness is absolutely critical to ensure the success of the interventions and the potential recovery of the population affected by a shock [44,45,52,53].The transformational nature of the change for this Principle 5 is however of the same nature than for Principle 2 discussed above.It is transformational because it requires some structural and procedural alterations of the system to ensure the rapidity of the disbursement of the transfers in the aftermath (or sometimes in anticipation) of a shock/disaster.
The transformational nature of the change induced by applying Principle 3 (Target households vulnerable to shocks and disasters) may not appear obvious at first sight.The reader will need to remember however that data requirement necessary for identifying households who should be included and those should be excluded from a social safety net program (what is called the targeting process) can represent a significant challenge for the implementing agency as the information needed to identify the new households affected by shock or disaster may not be recorded in the initial database of the safety net program.Indeed, shock-responsive targeting requires information that reflects sudden changes or that has the potential to capture shock-vulnerability in advance of a shock.Yet, because they focus on chronic poor, targeting mechanisms set up for regular social protection programs do not reflect well this transitory aspect, thus often fail to capture rapid changes in circumstances.Adaptive social protection or climate-smart programs need therefore to include in their targeting procedures the ability to distinguish between the chronic poor and those that are likely to suffer transitory poverty as a result of climate/disaster shocks.That particular ability (to distinguish between chronic poor and transient disaster-affected populations) is what we consider a transformational change.In fact, we would argue that from the three ASP principles discussed in this section (Principles 2, 3 and 5), this principle 3 may be the most transformational of all.While the compliance to principles 2 and 5 may to a large extent depend on technical or procedural aspects (the setting up a contingency funds and the procedures to ensure the scalability and swiftness of the program), the change required for Principle 3 affects the nature of the interventions.From a system that was targeting a well identified population of chronic poor we now need it to also identify and support transient populations affected differently by various adverse events.The experience of the ASP program in Sahel reveals the trickiness of the task and the additional strain this could put of the existing system.In particular this would require that the governments of the respective countries where the ASP system is being established have the financial and human resources to ensure that the databases of the beneficiaries-and the potential beneficiaries (those who may in the future be affected by particular shocks)-are updated on a regular basis.The experience in the six countries included in the ASP program suggests that this transformational change may be challenging for countries with limited resources and capacities.

Transformation at the Beneficiary Level
The transformation may not be only taking place within the safety net system itself; it may also be an outcome of the intervention at the beneficiaries' level.In the social protection literature, the potential for interventions to be transformative for the beneficiaries is something that had been highlighted for quite a while.Devereux and Sabates-Wheeler had proposed a categorization that emphasized this potential in their 2004 paper "Transformative Social Protection" in which they argue that social protection "can empower marginalized people and be socially "transformative"" [5] (p.3).By transformative they refer to social protection interventions that "extends social protection to arenas such as equity, empowerment and economic, social and cultural rights, rather than [simply] confining the scope ( . . . ) to targeted income and consumption transfers."[5] (p.3).To a large extent, this wording is close to the previously mentioned idea of transformation linked to the challenge of the status-quo.(e.g., Mark Pelling, Kate O'Brien).
In the case of ASP, we argue that the "A" element of an ASP intervention do have, or should have, some transformative impact in addition to -or in complement tothese potential transformative effects which Devereux and Sabates-Wheeler were referring to around empowerment or equity.Those other transformative elements are related to the idea that certain ASP interventions are expected to foster the adaptive capacities of beneficiaries.A central element of ASP (in fact directly related to Principle 4) is indeed the emphasis on developing the capacities of the beneficiaries to complement the conventional short-term absorptive/coping responses by longer-term adaptive strategies.This, we argue, constitutes another important dimension of transformation, in line with what Berman and her colleagues refers to when they talk about "the transformation of coping capacity into adaptive capacity" [41] (p.91).
So far, however, no rigorous assessments have been done that evidence the impact of such activities on beneficiaries' adaptive/transformative capacity.

ASP, Resilience and Their Relation in the Context of the Humanitarian-Development Continuum
In this last section of the discussion we aim at exploring the relation between the concept of ASP and resilience as well as discussing briefly where ASP lies along the humanitarian-development continuum.In doing so we also discuss another concept that has emerged recently in the related literature: shock-responsive social protection.

Conceptualizing Resilience
Many different definitions of resilience have been proposed in the literature.In the sphere of humanitarian and food security interventions, several of those definitions and associated frameworks are now widely referred to in both academic and practitioner communities (see e.g., the frameworks of [54][55][56][57]).Although slightly different in their wording they all fundamentally carry the same message: in the context of humanitarian and food security crises, resilience is about the capacities of people and communities to deal with shock in a way that does not affect negatively their long-term wellbeing.In this paper we use Béné et al. [58]'s resilience framework.In addition of being widely used, this framework also provides a strong conceptual foundation to clarify the link between ASP and resilience and how both are linked to the humanitarian-development continuum.
In this framework, resilience is understood as a combination of three different but complementary capacities: absorptive, adaptive and transformative capacities.Absorptive capacities refer essentially to the capacities of households to develop and adopt "coping strategies by which [their] members moderate or buffer the direct short-term impacts of shocks on their livelihoods and basic needs" [58] (p.125).In contrast adaptive capacities refers to the "various incremental changes and adaptations that people undergo in order to continue functioning in response to a shock or a growing stress, without making major qualitative changes to the way they operate" [58] (p.125); and transformative capacity refers to responses and strategies that "aim at altering permanently and drastically the household structure or functioning as a way to ensure the long-term 'survival' of the individual member/household" [58] (p.125).
This conceptualization of resilience also highlights that resilience results from the combination and interactions of all of these three capacities, each of them characterized by a specific responses: (1) absorptive capacity leading to persistence; (2) adaptive capacity leading to incremental adjustments/changes and adaptation; and (3) transformative capacity leading to transformational responses [32] (for a more rigorous mathematical formulation of these different dimensions of resilience and how they contribute to the overall theory of resilience, see [59]).These three dimensions of resilience are represented schematically on the left-hand side of Figure 1.
programmatic rapprochement between their short-term humanitarian interventions and their longerterm development programs.USAID but also DFID are two concrete examples of this attempt.USAID, for instance undertook an "operational change" in 2012 which was highlighted in its Policy and Program Guidance Document of that year.In that document resilience was identified as the "common objective" that allows an "institutional[ized] change by ensuring that integrated teams of humanitarian relief and development experts work together" [55] (p.6, our emphasis).

Adaptive Social Protection and Shock Responsive Social Protection
Over the last seven years, several donors and development agencies including DFID and UNDP have been paying more attention to the concept of Adaptive Social Protection but also what is called Shock Responsive Social Protection (SRSP).At first sight, both concepts seem very close, at least as far as their names are concerned, and it seems therefore legitimate to raise the question of whether they are theoretically and empirically the same types of programs or whether they refer to two closely related but different types of interventions.As we shall see below, elucidating this distinction will also provide additional useful element to our discussion and our attempt to clarify where ASP lies along the humanitarian-development continuum.
Although neither ASP nor SRSP have explicit definitions and are relatively new concepts, both are widely described and discussed in the recent literature, so that it is possible to identify more precisely their respective scopes and aims.

Humanitarian-Development Intervention Continuum
The center part of Figure 1 shows three generic types of interventions: humanitarian, climate change and long-term development interventions and sketches their respective focus and scope in relation to the three dimensions of resilience capacities discussed above.Historically, humanitarian interventions' objective has been to respond to crises and to support populations after these have been hit by an extreme event or a disaster.In so doing, humanitarian interventions focus mainly on coping strategies by strengthening the absorptive capacities of these households, and by helping them in particular to 'bounce back' after the disaster [52,60,61].This means that these humanitarian interventions are located mainly at the lower part of Figure 1 in relation to the absorptive/coping capacities of resilience.In contrast development programs which are focusing on longer-term issues related to societal changes (e.g., promotion of gender equality, shift in power-relationship, etc.) [62], are more closely linked to the transformational dimension of resilience at the top of the diagram.Finally, the middle part of the diagram includes climate change adaptation programs, which for their large majority aim to strengthen population's adaptive capacities in the face of climate related shocks and stressors [21].Overlap between the respective scopes of these three types of generic interventions exist, and they are also represented on Figure 1.
On the right-hand side of Figure 1 the humanitarian-development continuum is represented.The concept of linking relief and development emerged in the 1990s when practitioners started to acknowledge a lack of coordination between humanitarian assistance, relief, and development activities [63].Since then, agencies, academics and practitioners have attempted to find ways of reconciling the humanitarian-development nexus to provide both effective humanitarian relief, and sustainable medium-and long-term development action.The basic premise of the humanitarian-development continuum is the need to link and create synergies between short-term relief measures, and longer-term development programs [64].It reflects the belief that humanitarian needs and poverty are inter-related and often occur concurrently [63].The framework presented in Figure 1 is useful in this regard.Although it offers what is with no doubt a simplified representation of the reality, it illustrates how conceptually resilience can help development agencies to undertake some programmatic rapprochement between their short-term humanitarian interventions and their longer-term development programs.USAID but also DFID are two concrete examples of this attempt.USAID, for instance undertook an "operational change" in 2012 which was highlighted in its Policy and Program Guidance Document of that year.In that document resilience was identified as the "common objective" that allows an "institutional[ized] change by ensuring that integrated teams of humanitarian relief and development experts work together" [55] (p.6, our emphasis).

Adaptive Social Protection and Shock Responsive Social Protection
Over the last seven years, several donors and development agencies including DFID and UNDP have been paying more attention to the concept of Adaptive Social Protection but also what is called Shock Responsive Social Protection (SRSP).At first sight, both concepts seem very close, at least as far as their names are concerned, and it seems therefore legitimate to raise the question of whether they are theoretically and empirically the same types of programs or whether they refer to two closely related but different types of interventions.As we shall see below, elucidating this distinction will also provide additional useful element to our discussion and our attempt to clarify where ASP lies along the humanitarian-development continuum.
Although neither ASP nor SRSP have explicit definitions and are relatively new concepts, both are widely described and discussed in the recent literature, so that it is possible to identify more precisely their respective scopes and aims.
The review of the literature reveals that SRSP programs generally involve three sectors: social protection, disaster risk management and humanitarian assistance (e.g., [11,14,45,48]).Within this broad context, SRSP mainly aims at determining the ways social protection or safety net programs can be useful in relation to disaster or humanitarian crisis.For illustration, the research question of the Shock-Responsive Social Protection Systems Program that was recently completed by OPM [65] was: "Under what conditions can social protection systems be responsive to shocks and deliver effective shock response?"In effect a lot of the documents that discuss SRSP aim to determine how existing social protection (datasets, tools, concepts, techniques, or programs, etc.) can be used in an effective way by emergency actors.For illustration Table 1 lists the headings of the policy recommendations that were proposed by the Shock-Responsive Social Protection Systems program at the time of its completion ( [14], pp.80-82).Out of these 12 recommendations, six of them directly address this specific objective.
Focusing on how social protection can help making responses to shocks and disaster more effective also means that, along the humanitarian-development continuum, the emphasis of SRSP is more toward the bottom part of the continuum, i.e., more toward short-term responses focusing essentially on relief interventions.Within the resilience framework presented in Figure 1, it would mean that SRSP is mainly focusing on improving the coping strategies and absorptive capacities of the populations affected by shocks.This aspect contrasts with ASP which aims to deliver activities that support coping strategies but also to provide interventions that foster adaptive and possibly transformative strategies.In that sense ASP supports a more comprehensive agenda that SRSP since it does consider the interface between social protection and disaster risk reduction (like SRSP does), but also looks at the possibility of social protection contributing to strengthen households' adaptive capacity to climate change.

Build strategic coordination and collaboration between the social protection, humanitarian and DRM sectors.
12. We need more M&E information on the efficiency and effectiveness of shock-responsive social protection.
It is too soon to be able to assess the World Bank ASP program's contribution to this adaptive dimension.Nevertheless, a previous study reviewing 124 existing programs with similar aims showed that the inclusion of a combination of social protection, disaster risk reduction and climate change adaptation objectives in an integrated program is more likely to foster the adoption of preventive (anticipative) and longer-term adaptive and/or transformational interventions [12] than programs that focus only on the combination of social protection and disaster risk reduction.This would locate ASP along a wider spectrum along the humanitarian-development continuum than SRSP (see Figure 1 right hand side of the diagram).In the context of resilience intervention this would also mean the implementation of interventions that aim not simply at supporting short-term absorptive capacities but also at strengthening the adaptive and transformative resilience capacities of households and communities affected by shocks and stressors.Importantly this distinction has also been acknowledged by members of the World Bank ASP team.The Senegal Social ASP Program leader for instance pointed out in a conversation: "The objective is not only for the safety nets systems [supported by the World Bank ASP program] to have the capacity to react and scale up response to a variety of shocks-but more generally to build resilience of vulnerable households in the face of climate change related shocks.That means before the shocks happen and after" (S.Rougeaux pers.Communication-31 January 2018).
This emphasis on strengthening adaptive resilience capacities is the main difference between ASP and SRSP.This inclusion of activities and interventions that foster or build the adaptive capacity of its beneficiaries is also what makes ASP transformative in the sense defined by Berman and her colleagues referring to "the transformation of coping capacity into adaptive capacity" [41] (p.91).

Conclusions
This paper was an invited contribution to the Special issue of Sustainability on "Transforming Development and Disaster Risk".The paper focuses the discussion around one particular type of intervention referred to in the literature as adaptive social protection (ASP).In the current context of increased intensity and frequency of climate-related extreme events and disasters combined with global and local economic volatility, adaptive social protection has been recognized as a potentially powerful policy response to reduce the impoverishing impacts of these different types of adverse events.The implementation of ASP is currently being piloted at scale by the World Bank in six countries in the Sahel region.
Relying essentially on conceptual considerations, the paper aimed to address three questions: How and to what extent can ASP be considered transformative?Where does this concept sit along the humanitarian-development continuum?And, how does it relate to resilience?
The analysis was limited mainly on one account.The World Bank ASP program has been implemented for less than thirty months (at the time of writing this paper) which means that some of the elements on which the arguments presented in this research are based would need to be revisited at the end of the program's implementation in 2019.In particular, whether or not the ASP program is effectively successful at strengthening the resilience capacities of the beneficiaries of the program in the Sahel would still have to be determined.Nevertheless, none of the main points discussed in the paper depend for their conceptual relevance on the ASP program's final outcomes.This means that the main arguments proposed in the paper remain valid irrespective of the success (or failure) of the current ASP program.
Drawing on the authors' involvement in the evaluation of the ASP pilot program, and including reflections derived from the review of other documents published recently on the concepts of ASP and shock-responsive social protection, the paper demonstrates why ASP can be considered as a transformative intervention, in contrast to other types of programs of the same 'family' such as shock-responsive social protection.The paper shows in particular how through their activities designed to strengthen the beneficiaries' adaptive capacity ASP-type programs can (at least in theory) contribute to build the resilience capacity of households beyond the short-term coping strategies which humanitarian interventions generally focus on.As such, the paper argues, ASP-type programs cover a larger spectrum along the humanitarian-development spectrum than most of other interventions proposed in the context of shock-responsive interventions.
These reflections remain however conceptual and the pilot being implemented by the World Bank provides some important additional empirical lessons.First, in line with some part of the literature, the World Bank ASP pilot shows that transformational programs are not always welcome especially when they challenge the current institutional status quo.In these conditions it is important to acknowledge the 'political economy' of the changes that are imposed by the new intervention and ensure that the process used to introduce these changes is adequate (e.g., consultative or participatory enough) to reduce the risk of rejection.Second, beyond this political economy of change, transformative interventions-because they are usually associated with significant/systemic alterations in the structure or the functioning of the system-may also require additional skills, knowledge or resources, something which is not systematically available in low or middle-income countries.
These various conclusions highlight several potential areas for further research.One of them relates to the effectiveness of ASP-type programs to strengthen the resilience capacities of the beneficiaries.While the paper demonstrated the conceptual link between ASP and resilience, it would remain important to improve our understanding of the actual impact pathway(s) through which this resilience strengthening effect is achieved effectively.At the present time our understanding of which activity, or combination of activities, in an ASP program is more effective in building the resilience capacities of households and communities is still low.Whether for instance it is more important to empower women at the household level than to invest in community level governance as a way to strengthen people adaptive capacities and resilience-or whether we need a combination of both-is still not clear.Furthermore, it is likely that the appropriate 'package' of activities is context-and possibly shock-specific.Yet at the present time our knowledge on what works and what does not is patchy to say the least.There is therefore a need for more empirical research on these questions.More systematic rigorous evaluations are also needed to strengthen the case.
At the policy level the principal lesson that has emerged so far from the evaluation of the World Bank ASP pilot program is the confirmation of the importance of dialogues and consultations between the main relevant actors and policy makers at national level.There are two distinct reasons for this.One is related to the ASP key principle 1 "Promote institutional coordination" and the fact that by definition ASP is about dialogue and collaboration between ministries and communities of practice involved in social protection, but also climate change adaptation, humanitarian responses and disaster risk reduction.The second reason why policy dialogues and consultations are important is the fact that without those there is a serious risk that some of the key actors will reject or at least resist the necessary transformational changes.In the case of the World Bank ASP pilot program for instance the initial underestimation of the importance of these consultations is thought to have resulted in lost opportunities for cross-learning and has been identified as the reason why some consensus around important elements of ASP are still not fully achieved in some of the countries where the program is operating.Acknowledging and embracing the political economy of transformational changes is therefore key.

Figure 1 .
Figure 1.The place of Shock Responsive Social Protection (SRSP) and Adaptive Social Protection (ASP) with respect to resilience interventions, short-term humanitarian assistance, climate change adaption projects, and long-term development programs, and their link to the humanitariandevelopment continuum.The 3D resilience framework (absorptive, adaptive, transformative capacities) is derived from [58] looking at the conceptual relationship between resilience and food security interventions.

Figure 1 .
Figure 1.The place of Shock Responsive Social Protection (SRSP) and Adaptive Social Protection (ASP) with respect to resilience interventions, short-term humanitarian assistance, climate change adaption projects, and long-term development programs, and their link to the humanitarian-development continuum.The 3D resilience framework (absorptive, adaptive, transformative capacities) is derived from [58] looking at the conceptual relationship between resilience and food security interventions.

Table 1 .
The 12 recommendations that emerged from the Shock-Responsive Social Protection Systems program.In bold the recommendations that focus specifically on the ways safety net programs can make disaster or humanitarian interventions more effective.