Storytelling and Policy Transfer

This article argues that policy narratives underpin the mechanisms driving policy transfers. It applies the Narrative Policy Framework (NPF) to examine the policy narratives of the transfer agent and recipient, involved in the introduction of disaster risk reduction policy to West Africa. The article analyses the narrative setting, character, plot and moral depicted by the actors in their policy narratives and explains how these narrative elements informed the observation of the transfer mechanism of obligation.


Introduction
Storytelling is a universal means of communication (Jones, 2018). Stories are used to "organize, process, and convey information" (Jones & McBeth 2010, p.329) and policy actors can organise information through stories presented in policy narratives (Kaplan 1986;McBeth et al. 2014). These narratives are often relatable and easy to comprehend as events and locations are webbed around actors, who can be individuals or groups, and they in turn, based on their actions, morph and define the state of the events and locations that are captured in the story. This article examines the role of policy narratives in policy transfer, particularly, how narratives used by policy transfer actors can shape the transfer mechanisms that drive the occurrence of policy transfer. It considers the case of transfer of disaster risk reduction policy to West Africa by the United Nations International Strategy for Disaster Reduction (UNISDR). It identifies and analyses the policy narratives used by the actors through the application of the Narrative Policy Framework (NPF). The policy narratives are analysed by identifying narrative elements recognised in the NPF literature, including setting, character, plot, and moral, and aggregating their observation. These policy narratives were identified from selected communication materials, such as newsletters, research reports, meeting reports, flyers, and policy documents, which had been generated by the transfer actors prior to or during the process of policy transfer. Thirty-one documents were analysed and, in addition to this, two elite interviews were conducted with officials of UNISDR and the Economic Community of West Africa (ECOWAS) DRR Unit. The application of the NPF allows making arguments for how the policy transfer was facilitated through an obligatory pact and the narratives that led to the emergence of this mechanism. I argue that the depiction of the narrative character of allies and victims, the plot of overcoming the monster, morals of value reorientation and managing cost and benefit, as well as situating a bureaucratic setting, shaped the manifestation of the transfer mechanism of obligation in the transfer of Disaster Risk Reduction (DRR) policy to ECOWAS by the UNISDR.
The study makes conceptual contributions to the policy transfer and NPF literatures. It introduces a methodical dimension to the analysis of the interaction of transfer actors, through the identification and examination of NPF's narrative elements, i.e. setting, character, plot, and moral. It also brings a transnational dimension to the application of NPF's analysis of policy narratives through the empirical case of DRR policy transfer by UNISDR. The rest of the article is structured as follows. First, there is a discussion of the policy transfer framework, followed by an overview of the NPF, and then the case study is presented. Next, the article examines which narrative elements are depicted by the actors and analyses how these depictions shaped the manifestation of the transfer mechanism of obligation. Lastly, the article concludes with a discussion of the relevance of applying communication approaches in transfer studies and reflects on the use of obligation by inter-governmental organizations (IGOs) for facilitating the adoption of global policies in sub-Sahara African countries.

Policy Transfer
According to Dolowitz (2017, p.39), policy transfer is essentially "a process where knowledge of how to make things work in one political system is used in the development of similar solutions in another political system". As a public policy framework, policy transfer is not of itself a novel concept. It is an offshoot of a preceding concept, policy diffusion, which seeks to understand the phenomenon of different political jurisdictions having similar policies. Policy transfer was introduced in a bid to address the limitations of policy diffusion, chiefly, the limited attention to the micro-mechanisms driving the process (Dolowitz, 2017;Hadjiisky et al., 2017;Vigour, 2017). Policy transfer investigates the attendant events, actors, and arenas observed in the process of a specific jurisdiction developing a policy existent in another jurisdiction (Baker et al., 2019). This focus has also informed the introduction of a multi-level and multi-disciplinary perspective (Legrand, 2016), which showcases the multiplicity of possible transfer agents and recipients, and multiple pathways that can materialise in transfer processes. By following this approach, several transfer studies have been able to examine frictions that impeded a transfer process (Dunlop, 2009); how what was transferred, i.e. the transfer object, is faring in the recipient jurisdiction (Delpeuch and Vassileva, 2017), the activities of a particular group of transfer agents (Carroll, 2014), and the source or origin of decisions taken by domestic policy actors (Yilmaz, 2017). Often, in all these investigations, the transfer mechanism that is involved in facilitating the occurrence of transfer is identified and mentioned, but not expounded.
Policy transfer mechanisms are factors that facilitate the eventual occurrence or non-occurrence of policy transfer, i.e. they can explain why transfer takes place or otherwise, primarily based on the acceptance or resistance of the intended recipient. They are broadly grouped into voluntary and coercive mechanisms. The voluntary category primarily represents when a jurisdiction internally engages in learning from the experience of other jurisdictions for development of policies, administrative arrangements, institutions and ideas (Dolowitz and Marsh, 2000;Ertugal, 2018). With regards to the second category, Benson & Jordan (2012) note that the coercive group comprises conditionality, obligation, imposition, and semi-coercive mechanisms. Except for semi-coercive mechanism, the identification of the other three sub-types of the coercive category in instances of policy transfer has been extensively reported in the literature (Radaelli 2000;Nutley et al. 2012;Blum 2014;Carroll 2014;Ertugal, 2018;Dolowitz et al., 2019). Not all of these mechanisms are identical in the preceding concept, where learning, competition, emulation, coercion, and social construction are identified as mechanisms of policy diffusion (Graham et al., 2013;Maggetti & Gilardi, 2016). This is to be expected, since policy transfer is concerned with the micro-dynamics and political dimensions of the phenomenon these concepts explain (Hadjiisky et al., 2017). As such, transfer mechanisms are primarily premised on the demonstrated relations between the recipients and other transfer actors. Also, transfer mechanisms offer an in-depth perspective to the experience of actors and transfer objects, while mechanisms of diffusion help to present a broader perspective to the patterns and drivers of the spread of similar policies to varying political jurisdictions. Nonetheless, these are all relevant mechanisms and they can sometimes be interlinked (Zhang & Zhu, 2019) or observed to operate in parallel formations (van de Kuilen, 2019).
As earlier stated, this study focuses on the mechanism of obligation. Transfer mechanisms, in and of themselves, cannot materialise in policy transfer instances outside of their expression by policy transfer actors, and this makes the examination of interaction among policy actors imperative. It also shows that narratives should not be ignored. Indeed, critical policy studies have demonstrated that policy realities can be contextual and are shaped by the perceptions of actors. This knowledge has dawned on the examination of discourse, the consideration of the social construction of policy issues, drawing meanings from the narratives of policy actors to explain policy action, and the introduction of communication approaches in public policy research (Mcbeth, 2014;Yanow, 2000). While the investigation of policy narratives is yet to emerge as a principal approach in policy transfer studies, the references to policy narratives in the literature are few and far between (Onursal-Beşgül, 2016;Soremi, 2019;Taylor, 2018). Despite this, recent developments in the literature attest to a growing interest in the consideration of communication approaches for investigating policy transfer cases. A recent review of the literature also presents evidence that supports the importance of communication in the transfer process and its relevance for mutual understanding and preventing miscommunication (Minkman et al., 2018). Some studies have employed communication-based approaches, leading to the incorporation of different concepts with policy transfer, such as discursive institutionalism, instrument constituency, and public relations theory. Table I   How are transfer objects re-interpreted and translated?
Communicative understanding of policy transfer, with respect to how borrowed policies are framed and how they perform. (Wahlström & Sundberg, 2018) The recognition of persuasive argumentation for legitimisation. It adds to the understanding of policy transfer as not simply driven by cognitive justification (Nordin and Sundberg, 2016).

Public Relations Theory
How do communication types between policy lenders and recipients result in different modes of translation in the transfer process?
Identification of translation modes in policy transfer process based on communication types i.e. one way-communication and two-way communication. The resultant translation modes can be distorted, rational, discrete and democratic (Park et al., 2014(Park et al., , 2017.

Instrument Constituencies
How do instrument constituencies shape policy transfer?
What are the ideas and discourse that inform the role of actors and their agency in promoting international policy transfers?
Understanding the social construction of transfer objects and its influence on the movement of policies across jurisdictions (Foli et al., 2017).

Source: The Author
Notably, the emphasis on the examination of the interactions of actors in policy transfer research can be credited to the emergence of the concepts of policy mobilities (McCann and Ward, 2012) and policy translation (Clarke et al., 2015;Prince 2010;Stone 2017). Policy mobilities engages with the travels of policy-relevant knowledge, and policy translation examines the morphing and embedding of transfer objects in the recipient jurisdiction. Going with this tangent, these concepts are communication inclined and they emphasize "discourse, language, discursive frames, … and cultural framing" (Hadjiisky et al., 2017, p.15). Without a doubt, these concepts have etched their mark in the field of policy transfer through the demonstration of the viability of understanding the role of interactions amongst actors and its influence on the transfer process. They have also enabled venturing into aspects of the transfer process that otherwise might be discounted, including international conferences (Porto de Oliveira, 2017; Montero, 2019), study tours (Cook et al., 2015), official visits (Baker and McGuirk, 2019), and policy tourism (Hudson and Kim, 2014).
Notwithstanding the influence of the mobilities and translation approaches, many of the studies that address the interaction among actors do not often adopt a methodical approach to the analysis of discourse. The mainstay in the transfer literature is the multiplicity of research directions, which limits the commonality of research language and practice in transfer studies. This article addresses this gap by applying the NPF. The NPF is a public policy theory for investigating the role of policy narratives in the policy process (Jones, 2018). The framework has the potential capability and efficacy for a consistent approach to examining communication among transfer actors. Importantly, a focus on policy narratives offers a lens to navigate the course of actions in a transfer process. It allows delving into underlying statements and reactions that inform and exemplify the intents, interpretations, and undertakings that are identified in transfer cases. Through the analysis of narratives, we can interrogate the power and dependency relations (Crossley, 2018) and investigate the "often overlooked role of motivation" (Dolowitz, 2017, p. 40). To examine motivations, narratives matter in delimiting the space where the actors aim to influence, identifying how and where actors are situated, and portraying the envisioned targets and movements of the transfer object. Narratives can also highlight preferences and aversions, as well as affinity and disinterest, towards varying transfer pathways, sources, agents and recipients. This positions the analysis of policy narratives as a veritable tool for examining the policy transfer process and, based on this claim, this article draws on the NPF to explain how narratives shape transfer mechanisms.
The next section introduces the NPF, presents the key narrative elements, setting, character, plot, and moral, and discusses how these can help us to understand how policy transfer mechanisms evolve and can be identified.

The Narrative Policy Framework (NPF)
The NPF was borne from an understanding of the social construction of policy realities (Jones, 2018). It is positioned as a public policy framework for exploring the role of narratives in the policy process (Shanahan et al., 2017). Policy narratives are used by diverse policy actors for expressing preferences, engaging in debates, disseminating information to the public, and mobilising support. The NPF posits that there are identifiable features that constitute a narrative structure and are common to most stories. These features can be compiled, compared, and contrasted to examine how policy actors make use of narratives in communicating policy issues and assessing the role of narratives in the policy process. The identifiable features are grouped into narrative elements, namely, setting, character, plot, and moral, and they are the building blocks of conducting narrative analysis using the NPF (Shanahan et al., 2018). A description of these elements is presented in Table II.

Setting
The problem environment or context

• Bureaucratic
Situating the policy transfer as a focus on institutional and administrative frameworks • Community Situating the policy transfer as a focus on delivering service to a specific group in the population

• Sectoral
Situating the policy transfer as a focus on a sector and addressing a challenge within the sector

• Collective
Situating the policy transfer as a focus on providing service to all groups of the population

Character
The representation of the identity and/or role of individuals, groups or organisations referred to in a policy narrative • Hero An actor that has made a notable achievement, attained success, or has a remarkable capability

• Ally
A supporter of another actor, in relation to its identity, position, policy position, vision, or project; or actors that are partnering together towards achieving an objective

• Villain
Who or what is causing or contributing to harm

• Antagonist
Who is antagonising/opposing another actor, or a policy stance or solution, or a project

• Victim
Who or what is negatively affected by a situation or challenge

Plot
The journey or overview of the sequence of events that the story will take its audience through (Mayer, 2014). It also emphasizes the temporal element of policy narratives as stories with a beginning, middle and end (Jones and McBeth, 2010). The emergence of investigating the influence of interactions among transfer actors places the NPF as a timely tool that can extend the import of transfer studies. As a policy communication approach, it allows a comprehensive focus on how interactions can inform what and/ or how actors conceive of their engagement in the transfer process, what they deduce of the components of what is being transferred, the composition of transfer agents, their reflection of challenges or triumphs during the transfer process, and even propensity for extended engagement in transfers. The narrative analysis entails examining a communication form such as newsletters, speeches, or tweets that consist of at least one character and are related to a policy area. Following the descriptions in Table II, the specific narrative elements are identified in the communication materials, and the extent of their depictions can be compared for different narrators. In comparison to the approaches identified in Table I, the NPF can be incorporated not only as a theoretical approach, but also as a methodical tool embedded in transfer studies. While these other approaches bring new insights to transfer studies, the NPF can also enhance their application through employing the model for narrative analysis advanced by the NPF. In this manner, what we already know of the transfer process can be further expounded through the lens of the narrative elements.
The focus on narratives is more so important in studying policy transfer, based on the analysis of the influence of international organisations, which, having limited or no jurisdictional authority, rely on coercive and persuasive measures to introduce policies in recipient jurisdictions (Minkman et al., 2018;Montero, 2019). To understand coercive and persuasive devices employed by this group of actors, policy narratives enable the identification of strategies that mask unpleasant information that may obstruct the coercive transfer or mar the perception of the transfer agent or object by the recipients. An example of this is the framing of a local policy problem, such as natural disasters, to align with policy preferences of international actors (Lawlor and Crow, 2018). Another strategy is the likely use of an ally-seeking narrative, whereby the narrative characters of villain and victim are eschewed in a bid to ensure the recipient is not portrayed negatively, so as to secure support or consent for the transfer object (Soremi, 2018).
With respect to transfer mechanisms, policy narratives can portray the perceptions and intents of transfer actors, which reflect the manner of relationship that typify specific mechanisms (Minkman 2018, Stone 2012. A narrative could, for instance, depict a story of a hero helping a victim, or a hero conquering a villain. These depictions may be linked to likely instances of internally-initiated policy transfer and possibly be driven by the mechanism of learning. On the other hand, narratives that depict a hero helping victims to conquer a villain, or a hero working with an ally to help victims or conquer a villain, may be linked to externally-initiated transfers and, possibly, mechanisms of conditionality, obligation, imposition or persuasion. In addition to depicting perceptions, the analysis of narratives can also represent an aspect of policy transfer study that can be applied across diverse geographical locations, since stories are regarded as a universal communication means (Jones et al., 2014), and transfer actors are often engaged in looped interactions (Walker, 2018). Importantly, because the concept of policy transfer highlights the movement of ideas and intermediation of policy actors, narratives can show from where and to where have policy ideas moved, and who mobilised them. This makes it tenable to apply the NPF, not only for examining transfer mechanisms, but also for addressing the other key questions that give an overview of policy transfer research. The narrative elements of setting, character, plot and moral can all be linked to the standard policy transfer questions of who, what, when, where and why (Dolowitz, 2017). As most communication content on policy transfers would already meet the minimum conditions for examining policy narratives, i.e. one character and a distinct policy area, the NPF therefore offers transfer scholars a ready-made tool for investigating cases using consistent and comparable language, which will go a long way to consolidating the literature.
The next section presents the case study and discusses the UNISDR's role in the transfer of DRR policy to ECOWAS and its reliance on the transfer mechanism of obligation. Following that, the findings from the analysis of the actors' use of narrative elements and how these shaped the manifestation of the mechanism of obligation will be considered.

Transfer of Disaster Risk Reduction Policy to ECOWAS
The world's attention was turned to DRR concerns after the occurrence of multiple natural disasters across Asia and the Americas in the 1960s (UNISDR, 2012). In response to this, the United Nations General Assembly, in 1965, mandated the UN secretariat to assist countries experiencing natural disasters by providing emergency services (Pelling and Uitto, 2001;UNIS-DR, 2012). However, after the occurrence of another earthquake in Iran, in 1968, which killed about 12,000 people, in 1970 the UN General Assembly in 1970 identified the need for disaster prevention and reducing damage and deaths (Pelling and Uitto, 2001;UNISDR, 2012). In the 1980s, the UN christened the forthcoming decade, the 1990s, the International Decade for Natural Disaster Reduction (UNISDR, 2012). During this decade, however, the disasters challenging West Africa and its intergovernmental regional body, the Economic Community of West Africa (ECOWAS), were not natural but man-made. These problems were also not environmental, but political. With the break-out of civil wars and attempted toppling of sitting governments in Liberia, Sierra Leone, and Guinea Bissau, managing the political crises ensuing from the war and ending the war altogether topped ECOWAS' agenda in the 1990s (Alter et al., 2013;Kufuor, 2006 (UNISDR, 2012). In response to this call, UNISDR, the United Nations Development Programme (UNDP) and the Africa Development Bank (ADB), partnered with the AU and the New Partnership for Africa's Development (NEPAD), to introduce activities targeted at developing continent-wide food security and disaster management policies (AU and NEPAD, 2004). ECOWAS contributed to the development of the African Regional Strategy for Disaster Risk Reduction (AU and NEPAD, 2004;ECOWAS, 2006) and then inaugurated a Technical Committee on Disaster Management.
ECOWAS' interest in DRR can be linked to the establishment of an African office of the UNIS-DR Vordzorgbe, 2006). UNISDR Africa identified and worked with relevant stakeholders like the AU, NEPAD, and ECOWAS. In particular, after the 2005 World Conference on DRR and the development of the Hyogo Framework of Action (HFA), UNISDR Africa set out to coordinate the domestication of the recommendations from the conference in African countries. It setup the Africa Advisory group "to advise and provide technical support for mobilizing political commitment and resources for DRR from national, sub-regional and regional authorities (e.g. ECOWAS, AU/NEPAD)" (UNISDR Africa, 2005a, p. 44). It also broadened the membership of an existing Africa Working Group on DRR to include regional organisations. ECOWAS then held a meeting with international DRR experts to deliberate on the development of DRR policy for West Africa (UNISDR, 2005b). In 2007, the policy was approved by Heads of States of ECOWAS nations (Bhavnani et al., 2008;UNECA and UNISDR, 2013 As shown in Figure I, UNISDR's intervention in the development of ECOWAS DRR policy centres on the domestication of the HFA and it employed the obligatory apparatus of the AU. According to Bulmer & Padgett (2005), a hierarchical relational approach to policy transfer can be associated with the transfer mechanisms of obligation and conditionality, but if obligation is present, it is often directed at the domestication of policies. ECOWAS also notes that the DRR policy was developed because African regional economic communities "are required to design disaster reduction policies and programmes under the Africa Strategy for Disaster Risk Reduction" (ECOWAS, 2006, p. 8).
Having discussed the case of DRR policy transfer to ECOWAS, the next section turns to the narrative elements used by the transfer agent, UNISDR, and the recipient, ECOWAS. The use of each of the NPF narrative elements, i.e. setting, character, plot, and moral, is analysed. Also, I examine how these narrative elements shaped the manifestation of the transfer mechanism of obligation.

Narrative Setting
The UNISDR set the scene for the transfer of DRR policy with a call for "strong legal, institutional frameworks" (Kamara, 2004, p. 61) and a request for African governments "to demonstrate their commitments through providing national policies and strategy on disaster reduction to its population and involve all stakeholders in all processes" (UN/ISDR Africa, 2005b, p. 49). With this use of a bureaucratic setting in its narratives, the transfer agent sets the scene for an obligatory policy transfer that will entail the creation of a new agency, specifically the establishment of national platforms on DRR in West African countries (AU and NEPAD, 2004;ECOWAS, 2006). With reference to other cases of obligated transfer in the literature (Blum, 2014;Eccleston and Woodward, 2013;Mamudu et al., 2015), the setting that often features is a sectoral or collective setting, whereby the transfer is focused on one sector in the society or the people within the jurisdiction. This makes the situation of a bureaucratic setting an unexpected observation.
This incongruity can also be seen in how the UNISDR relates differently to transfer recipients in developed countries such as Germany, Australia, Denmark, and the USA. In these countries, the transfer object is mostly in the form of a policy idea that will be incorporated into existing government structures (DKKV, 2017;UNISDR, 2017). Hence, the obligated policy transfer hardly results in the creation of new administrative arrangements or the development of new policy documents. UNISDR's depiction of a bureaucratic setting as against a sectoral or collective setting in the transfer of DRR policy to ECOWAS may be explained by its "significant involvement" (Marshall, 2005, p , 2003). This manner of intervention for ensuring the occurrence of the policy transfer therefore gave the UNISDR an edge in relating with ECOWAS, and with this advantage it was able to put forward the proposition for establishing and coordinating new institutional frameworks for managing DRR issues within West Africa.
Following in the footsteps of UNISDR, ECOWAS member states also situated a bureaucratic setting in their policy narratives and narratives and acquiesced to the recommendation for "the establishment of a formal institutional structure" (Olokesusi, 2005, p. 18). We can assume that the depiction of a similar narrative setting implies that the transfer recipient acknowledges the existence of a common problem and agrees to the recommendation for addressing the challenge. However, this is not always the case in instances of obligated transfer, since the recipient jurisdiction is often at liberty to decide how to meet its obligation and incorporate the transfer object within the governmental functions in its domain. For example, in the introduction of the Freedom of Information to Ireland and Portugal, which was informed by an EU directive on Access to Information on the Environment (AIE), the two recipient jurisdictions were at liberty to devise the appropriate mode of incorporating the directive so that Portugal went for enacting the Access to Documents Act, an act which also had a mix of French imports, while in Ireland the directive was "implemented through regulations" (Bugdahn, 2007, p. 134).
In addition, the transfer agent situated a collective narrative setting, as the policy was contextualised as an effort to give all citizens access to information on government activities and decisions (EUR-LEX, 2017; European Commission, 2016).
The seeming absence of contestation of the setting situated by the transfer agent is quite noteworthy, considering that the broader policy issue to which DRR is linked, i.e. climate change, is often a contentious topic between African policymakers and international actors. More so, the view that African nations need to make commitments to address climate change has often been countered with the concept of climate justice to maintain that the region did not cause the problem, but, because it would still be impacted negatively by the effects of climate change, it needs to be compensated (Okereke and Coventry, 2016). A possible explanation therefore, for the recipient's use of bureaucratic setting in an obligated policy transfer could be that it was unaware of, or was not familiar with, the possibility of having the liberty to meet its obligations and decide the setting of the policy transfer as it thinks best.
On the other hand, ECOWAS' use of bureaucratic setting could be credited as successful advocacy efforts of the transfer agent. UNISDR sought to engage with the topmost echelons of authority in ECOWAS and the relevant officials in a bid to secure alignment with the Hyogo Framework of Action. It participated in local summits, assigned resident consultants to regional organisations, and promoted DRR in different West African countries. Primarily, UNISDR was able to ensure that the transfer recipient acquiesced to not only the idea of obligated policy transfer on a contentious issue by securing their commitment to the HFA, but also to align with its preferred context for domesticating the HFA in West Africa.
Nonetheless, UNISDR's depiction of a bureaucratic setting, buoyed by its significant involvement in the policy transfer, led some of the ECOWAS member states to maintain a disposition that made it seem as though they were at the receiving end of conditional policy transfers. Some of the countries forthrightly expressed expectations of financial support from the UNIS-DR and other international development partners. For instance, the Chad government asserted its dependence on the international community for dealing with "hazards with disastrous impact on ecosystems" (Ibrahim, 2004, p. 10), while the Nigerian government emphasised its reliance on the international community, particularly the UNISDR, in order to fulfil some of the provisions of the HFA (Ojo, 2004, p. 20). Also, the AU's NEPAD, as though forewarning the UNISDR on what should be their additional contribution after playing a "critical role … in shaping the agenda in disaster risk reduction" (Rukato, 2004, p. 4), advised the transfer agent that it should not absolve itself from bearing some of the responsibilities for carrying out the agenda. Such expressions, which can also be detected in ECOWAS' consistent depiction of the character of victim in the two instances of policy transfer (as will be shown in the next section), made it seem that the recipient jurisdictions had conditioned themselves to expect conditional policy transfers.

Heroes and Allies
The introductory message in the newsletter published by the UNISDR and tagged Disaster Reduction in Africa: ISDR Informs (Valency 2003, p.1), gives an insight into how the UNISDR views its involvement in facilitating the policy transfer and its relationship with the intended transfer recipients. The message presented a space for co-ownership of the publication with its target audience. With emphasis on joint efforts and shared responsibilities, it highlighted the necessity for the direct contribution of other parties and portrayed an envisioned alliance with those vulnerable to disasters. UNISDR's narrative may have been informed by the need to ensure that the intended transfer recipients assume joint ownership of the problem and acknowledge the need for joining resources together. Knowing that what prompts obligated policy transfers is a drive for joint action to address a common challenge (Bulmer & Padgett 2005;Sobaci 2009), the emphasis on joint effort and shared responsibilities portrays the envisioned role of the recipients as allies. This narrative character portrayed the expected relations between UNISDR and ECOWAS. An example of this depiction can be seen in the quote from UNISDR below. Specifically, UNISDR employed a narrative character that positions both the transfer agents and recipients on a seemingly similar pedestal, despite the clear hierarchy that distinguishes them. As Radaelli (2003, p.33) notes, "[h]armonisation reduces … diversity, typically by providing a level playing field", and the depiction of the transfer agent and recipient as allies tends to be a means for presenting a level playing field in obligated policy transfers.
To face up this challenge, UNISDR Africa has forged a very sound partnership with African national governments and regional organisations to promote understanding of disaster risk reduction (UNISDR Africa, 2006, p.12).
When we consider that the transfer recipient in this case of obligated transfer is a regional body consisting of countries that are classified as 'developing' and 'least developed' countries, the recognition of a possible capability of these countries to contribute (even in the least measure) to international policy development efforts further highlights the acknowledgment of a shared problem and shared responsibility. Also, the casting of the recipients as allies by the UNISDR reiterates the non-reliance on the use of force in obligated transfers. Consequently, while there may be a hierarchical difference in the position of the transfer agent and recipient, i.e., international and regional, this hierarchy is not reflected in the transfer agent's portrayal of the recipient. In addition, the depiction of allies was also used to clarify the form of partnership that the transfer agent desired and send a message to the recipient that "partnership is about combining assets (strengths) not liabilities (weaknesses), Africa should contribute its assets -not its liabilities" (Valency 2004, p.1).
Similar to the observations in the actors' use of narrative setting, ECOWAS also rarely deployed grand statements or depiction of heroism in referring to its own identity or activities; instead, it was portrayed as an ally. The notion of what would be achieved via the transfer process could have informed this depiction. This is because with the expectation of joining in to address a global challenge, ECOWAS likely considered its engagement in the policy transfer as a piece required to solve a big puzzle.

Victims
The use of the narrative character of victim is aimed at emphasising the existence of a problem. It is also directed at mobilising the necessary support from allies in order to address a problem. In its narratives, the UNISDR emphasised catastrophes that were being experienced by com-munities within Africa, even though some of these were not directly related to environmental hazards. We can argue that the transfer agent drew from existing non-disaster related circumstances to project the victim status in a bid to establish a policy problem and communicate this effectively. This is in line with the notion that commitments and contributions for addressing a challenge can be more guaranteed when there is an identifiable victim (Clemons et al. 2012). As such, despite the issue of DRR being predicated on climate-related events (Bojang 2003), the UNISDR presents African countries as being harmed by diseases, poverty, armed conflicts, epidemics, and "large-scale social, political and economic problems" (UNISDR Africa 2003, p.4) in order to justify the existence of a problem worthy of attention. Thus, through its emphasis on the harm being done it shaped the materialisation of the mechanism of obligation, since "if a community does not believe it is at risk then it is unlikely to take part in disaster risk reduction programmes" (UN/ISDR Africa 2004b, p.5).
ECOWAS' member states also did not shy away from being situated as the victims in their narratives. They openly presented information on the occurrence of calamitous events within their jurisdiction. Other events that are non-environmental related were also linked to environmental hazards (Directorate of Civil Defence, 2006;Nying and Nyangado, 2005). In a way, these depictions signify the transfer recipient's acknowledgment of the existence of DRR challenges in West Africa. They also connote the validation of the transfer agent's engagement in policy transfer by the recipient, as an alternative depiction may imply an unwanted or unwarranted intrusion on the part of the transfer agent. The use of the narrative element by the ECOWAS member states signifies its association with the global problem of environmental disaster and this helps to readily suggest its acceptance of an obligatory responsibility to domesticate the HFA.

Villains and Antagonists
In the NPF literature, the villain is depicted as the person, group or thing that is responsible for causing or worsening a particular problem (Lybecker et al., 2016). The identification of the villain also signifies the portrayal of who or what should be blamed for a problem, such as "profit-driven corporations that are polluting the environment, radical environmental activists harming economic development and threatening individual liberty" (Pierce et al., 2014, p. 31).
The villain was also found to be the most frequently identified character in policy narratives analysed by NPF scholars (Pierce et al., 2014). However, the finding from this study revealed a different level of depiction of villains, as most of the documents analysed, particularly those authored by the transfer agent, did not identify a villain at all. Also, where the character of villain was depicted, it was associated with a non-human identity, and when humans are identified, there is no specific person, group, organisation named, but instead a generalised reference to human beings is given. The difference between the extent to which the character of villain is depicted in NPF studies and what can be observed from this study can be explained by the absence of "policy battles" (McBeth et al., 2014, p. 52), whereby opponents with differing policy ideas spar using competitive stories. One might say that in policy transfer, especially with regards to the relations between the transfer agent and recipient, there will likely be no use of competitive stories as the two parties are not opponents. Instead, the transfer agent uses stories to convince the transfer recipient to adopt a transfer object. The transfer recipient also uses stories to indicate whether it accedes to the proposition by the transfer agent or not through a positive or negative portrayal of the transfer agent and the proposed transfer object. Saddled with the task of convincing the recipient, transfer agents tend to be careful not to ruffle feathers by not pinpointing who might be causing harm or aggravating it.
In addition to not casting a villain that can be held liable for harm, the UNISDR also did not cast anyone as an antagonist. By using stories to woo another party that is resident in a different jurisdiction, the transfer agent avoided the depiction of a character with a distinctive negative connotation. Unlike the heroes and allies, who are depicted as contributing to addressing a challenge, or the victims, who are depicted as people who are being harmed and in need of support, the character of villain has a distinctive negative connotation. Based on this, and as though following an approach of 'see no evil, hear no evil, speak no evil', the transfer agent turned away its story lens from capturing past or recent untoward intentions or deeds. The reason transfer agents will likely shy away from casting villains and antagonists may be because they do not want anything to distract the prospective recipients from the issue at hand, i.e. the transfer object, as the casting of villain tends to lean towards emphasising differences in opinions and winning people over to one's side of the argument. So, rather than go along with what obtains in many policy narratives, where groups of actors use competitive stories, policy transfer agents are more likely to opt for uniting actors on both ends of the transfer process, regardless of differing opinions, in a bid to achieve a specific goal, i.e. adoption of the transfer object.

Plot and Moral
The main plot portrayed in this case of DRR policy transfer by the UNISDR and ECOWAS is the plot of overcoming the monster (Booker, 2004), and the monster in the story was stipulated to be disasters. These disasters, however, took varying forms. At the onset of the operations of UNISDR, the monster being tackled took the form of environmental hazards ranging from hurricanes to earthquakes. However, upon its engagement with African countries, the form of disaster being contemplated extended from environmental hazard to conflicts. In particular, having conflict as the monster to be tackled under the issue of DRR was part of the debate that emerged between the transfer agent and the recipient, as the African regional bodies, including ECOWAS, expressed that "some space had to be provided for digressing views and interests if the synergies and energies of different stakeholders were to be captured" (UN/ISDR Africa 2006, p.5). So, "[t]he question of whether disaster risk reduction could include conflict prevention was raised" in line with "the African context where it can be said that conflict is a disaster" (UN/ISDR Africa 2005a, p.42). Without refuting the point raised by the regional economic communities, the UNISDR noted that "[c]onflict can sometimes be triggered by a disaster; disasters can sometimes represent a means of resolving conflicts -therefore reducing risk and vulnerability can also be applied to conflicts" (UN/ISDR Africa 2005a, p.42). As a result, conflict was accommodated as part of what will be discussed among the actors. Following this, the ECOWAS' Executive Secretary reiterated "the determination of ECOWAS to reduce not only (natural) disasters but also armed conflicts" (UN/ISDR Africa 2005b, p.56).
Using a plot of overcoming the monsters, stories of calamities that could happen within the African continent were told by the UNISDR to make the recipient recognise the danger that lurked ahead and motivate decision makers within the recipient jurisdiction to act. The portrayal of this plot, needless to say, necessitated presenting vivid pictures and relatable numbers such as "millions of people" (UN/ISDR & NEPAD 2004, p.1;Annan 2004, p.33) and "decades" (Valency 2003, p.1;UN/ISDR Africa 2004, p.7;UNISDR 2005, p.1) of development that will be lost, to depict what might befall the African continent should they not pay attention to DRR. In addition, future climatic events that could wreak havoc or gory stories  were also referred to, all in a bid to communicate the possible debilitating effect of the monster, the urgency of dealing with it, and the notion that it will be insufficient for individual countries to go at it alone. The plot of overcoming the monster shows the transfer agent's reliance on making a call to action to ensure the adoption of the transfer object. Examples of the depiction of the plot can be seen in the two quotes below.
Recent news reports … indicate that, within months, 30 million Africans could be facing starvation, the immediate cause of which is drought (Hartnady, 2003, p. 26).
Mali, like several countries of the subregion, experienced important damage as areas deprived of water for several years received an excessive quantity of water. In addition, crops are threatened every year by locust invasion and various epidemics (Traore, 2005, p. 33).
A distinctive feature that characterises the depiction of the plot of overcoming the monster in policy transfer stories is the emphasis on the inability of the transfer recipient to individually tackle the policy problem. In obligated policy transfer, the plot of overcoming the monster also signals to the intended recipient the urgent need to come on board an international effort to address the problem. This realisation may have made the UNISDR to go extra lengths to acknowledge the theme of conflict, even when it did not resonate with its agenda. It may also have made it possible for ECOWAS to look past its heroic efforts to establish the ECOWAS Mechanism for Conflict Prevention, and to recognise that "while each state has primary responsibility for its own economic and social development, an enabling international environment is vital to stimulate and contribute to developing the knowledge, capacities, and motivation needed to build disaster resilient nations and communities" (UNISDR 2005, p.13). The portrayal of a plot overcoming the monster in obligated transfer also points to the recognition of the recipient's capability to have an input and shape what is manifested in the mechanism of obligation, whereby the transfer agent brings together and motivates willing and/or capable parties to commit to addressing a challenge by signing an accord or agreement. The transfer agent then invokes the appending of the accord or agreement to ensure the domestication of what was agreed. This form of relationship and basis for policy transfer, especially to developing countries, accords the recipient a decent degree of dignity and respect, whereby the recipient is not viewed as a liability, but as a party that can contribute towards solving global challenges. ECOWAS realigns its singular focus on conflicts and admits to the issue of natural disasters. It first set up a Technical Committee and then adopted the ECOWAS disaster risk reduction policy.

Source: The author
With respect to the narrative moral, the transfer agent presented the morals of value reorientation and managing costs and benefits. In depicting the moral of value reorientation, the UNISDR called for a new approach to addressing disasters, whereby there is more focus on disaster prevention rather than disaster relief. The moral to sum up its message is that a change in the disaster management approach in West Africa is crucial. The narrative moral particularly emphasises the need to move away from an approach limited to post-disaster activities and pursuing provisions for relief after disaster events, and to a more proactive approach in which disasters are prevented and vulnerability to disasters are reduced. With the depiction of managing costs and benefits, UNISDR tabled possibilities that could be considered in deciding whether to embrace the concept of DRR or not. It noted the possibility of the increased cost of humanitarian relief if casualties are not averted in the event of environmental hazards. It also highlighted the development gains that can be achieved if disasters are prevented, especially by saving lives and property (UN/ISDR & NEPAD 2004;UNISDR Africa & UNDP 2004;Vordzorgbe 2006). By attempting to introduce financial undertones, such as "cost-effectiveness" (UNISDR Africa 2004a, p.52;Vordzorgbe 2006, p.3) and "investments" (UNISDR Africa & UNDP 2004, p.15;UNISDR Africa 2005c, p.26), the UNISDR seemed to show awareness of its audience and their policymaking innings. It gave due regard to considerations that will be necessary for individual governments to make, particularly budgeting. It also presented content that could be useful for internal debates within the recipient jurisdictions to facilitate domestication of the HFA.
The narratives of some ECOWAS countries, on the other hand, were summed up with emphasis on the limited resources at the disposal of the governments (Ibrahim 2004;Ndiaye 2005;Odei 2005). It appears that ECOWAS joined in with telling the story of DRR based on an expecta-tion that its statement of commitment might be compensated through the receipt of support from the UN and, possibly, other aid donors. Indeed, this is not obligation as is known in policy transfer studies. Neither is it conditionality, as no conditions were stipulated to the recipient. It may be referred to as obligation with significant involvement of the transfer agent. Table III summarises the messaging that is implied with the depiction of the narrative elements and shows how it builds up to the shaping of an obligated transfer process with significant involvement of the transfer agent. In this mode of obligated transfer, the recipient is reliant on external support for fulfilling its obligation. While this dependency is reflected in the bureaucratic setting staged for the policy transfer, the liberty and nobility of the recipient's participation is nonetheless reflected in the ally relationship hewed for the transfer process, and a story plot bearing a message of collaboration.

Conclusion
This article has argued that the narratives used by actors in the transfer of DRR policy to ECO-WAS contributed to shaping the transfer mechanism. It shows how the policy transfer in this case was hinged on ECOWAS' obligation to the African Strategy on DRR and the HFA. It also presents the narrative elements that were used by the UNISDR and ECOWAS. The transfer agent and recipient depicted a bureaucratic setting, indicating that the transfer process was focused on altering administrative arrangements within the recipient jurisdiction. While this may indicate that the two sides are aligned, this setting differs from what is obtained in the UNISDR's intervention in developed countries. It was therefore argued that what likely informed this portrayal of narrative setting was the UNISDR's significant involvement in the transfer process through the provision of financial and/or in-kind support. The emphasis on allies by both actors, on the other hand, was an indication of the joint acknowledgment of a policy problem and a joint commitment to address the challenge. Also, the portrayal of victims and the plot of overcoming the monster presented a call to action, while the moral of value reorientation and that of managing costs and benefits summed up the essence of adopting the transfer object. In this case, the narrative moral focused on a change from disaster relief to disaster risk reduction, as well as considering the cost to be avoided if the safety of lives and properties are prioritised. This moral gives relevance not only to what is to be achieved collectively, but also the concerns that need to be contemplated domestically.
The identification of transfer mechanisms is often a response to the question of why transfer, since the engagement of actors may not always be linked to the importance of, or demand for a transfer object, but mainly to the association between the parties involved in the policy transfer. Regardless of this assertion, there has not been an explanation applicable widely across policy transfer case studies to demonstrate how the relations among actors contribute to the occurrence of policy transfer. This study has shown the relevance of the relations between the transfer recipient and agent or source, not by describing what transpired among the actors, but by analytically examining what constitutes their policy narratives and the ability thereof to inform the transfer mechanisms that facilitate the eventual occurrence of policy transfer.
With respect to the NPF, this study reiterates the significance of the framework and the potential of its utility for investigating the role of narratives in the policy process. By using the NPF's explanation of the depictions of narrative elements to illustrate how the actors' policy narratives shaped the transfer mechanisms, the research showcases the applicability of the framework for investigating circumstances that are not focused on 'policy battles' by interest groups. By applying the NPF to policy transfer research, this study has launched the NPF into the terrain of international communication among policy actors, as against the communication on domestic issues which currently dominates the NPF literature. With this venture into an international context, there is a broadened overview of how policy problems are perceived in different jurisdictions. As shown in this case of DRR policy transfer, whereas the international community defined the problem of disasters to be one of environmental concern, the West African community held a perception of disasters that pertained to conflicts. However, ECOWAS was able to come around to the internationally-held view of disasters through its relations with the UNISDR and the drive for the domestication of HFA.
Empirically, this study points out the possibility of transfer agents and recipients to decide on the mechanism to be employed in a transfer process by strategically communicating this in their narratives. For example, this case of DRR policy transfer could have been like other instances of obligated transfer but for the use of a narrative setting which connotes a broad external influence in the domestic affairs of the recipient jurisdiction. Also, should ECOWAS have been proactive in deciding how to meet its obligation to the African Strategy on DRR and HFA, and own its story, this instance of policy transfer might have emerged as an egalitarian transaction, as was likely envisioned by the transfer agent. In agreement with Dolowitz et al. (2019, p. 6), domestic actors need to tell stories that enable them to "alter the power relations with international agents and subsequently the course of the transfer".