Governance struggles and policy processes in disaster risk reduction: A case study from Nepal

In the neo-liberal climate of reduced responsibility for the state, alongside global platforms established to implement the Hyogo Framework for Action, a new arena opens for a multitude of stakeholders to engage in disaster risk reduction (DRR). The key role that the state can play in instituting effective DRR tends to receive little attention, yet in situations where the state apparatus is weak, such as in Nepal, it becomes evident that integrating DRR into development is a particularly challenging task. Due to the political sit- uation in Nepal, progress has been stalled in providing a legislative context conducive to effective DRR. This paper traces the evolution of key DRR initiatives that have been developed in spite of the challenging governance context, such as the National Strategy for Disaster Risk Management and the Nepal Risk Reduction Consortium. Informed by in-depth interviews with key informants, the argument is made that the dedicated efforts of national and international non-governmental organisations, multilateral agencies and donors in mainstreaming DRR demonstrate that considerable progress can be made even where gov- ernment departments are protective of their own interests and are slow to enact policies to support DRR. The paper suggests however, that without stronger engagement of key political actors the prospects for further progress in DRR may be limited. The ﬁndings have implications for other post-conﬂict countries or weak states engaging in DRR. (cid:2) 2014 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY license


Introduction
It is evident that some governments have successfully adopted and implemented disaster risk reduction (DRR) policies, while others lag behind (Williams, 2011). The shortcomings in DRR are increasingly being regarded as a consequence of weak governance and lack of political will (Williams, 2011). Despite this, in the academic literature, little attention has been given to the processes of governance of DRR, such as the formulation of policy and the roles of different stakeholders. In addition there is a lack of evidence on the effectiveness of different governance systems (UNISDR, 2011). Ojha et al. (2009: 365) suggest that effective governance is particularly hindered by the complex interplay of power and knowledge among diverse groups of actors with unequal command over resources. In DRR, this complex interplay of power and knowledge among diverse stakeholder groups gives rise to different governance approaches. As a post-conflict country in the early stages of democracy with a weak system of governance, Nepal offers an interesting case study on DRR implementation. This paper aims to explore in depth, the governance struggles and policy processes in this post-conflict state in which numerous stakeholders are emerging to fill the 'gap' left by weak government apparatus.
A brief theoretical discussion on governance is presented below, followed by an account of the global context of DRR. Nepal's disaster, political and social context is then introduced and the methodology for the research undertaken is outlined. Drawing on key informant interviews, the discussion provides a critical overview of the policy processes and the governance of DRR. The conclusion reflects on the implications of the research for other post-conflict states and states where government responsibility is reducing as part of a neo-liberal agenda.
Governance 'Governance' refers to the actors, structures and processes by which societies share power and make collectively binding decisions (van Asselt and Renn, 2011: 431;Lebel et al., 2006: 2). 1 According to Goodwin (1998) adopting a 'governance perspective' entails giving attention to the distribution of political power both internal and external to the state. The distribution of power between state and non-state actors has changed significantly over the last twenty years, as a consequence of neoliberal economic and political restructuring. This is often referred to as the shift from 'government to governance'. Before discussing the nature of this shift further a brief introduction to neoliberalism is presented.

Neoliberalism and the redistribution of state functions
Neoliberalism has a complex history and has been ascribed different meanings by rival groups of political economists throughout the 20th century (Gane, 2012). 2 In a broad sense, it involves a (re)negotiation of the boundaries between the market, the state, and civil society so that more areas of people's lives are governed by an economic logic, as the market is regarded to be the best mechanism for allocating goods and services to meet the diverse needs of actors across the globe (Castree, 2008). A popular notion of neoliberalism is that it is a laissez-faire political and economic culture which demands government and the state to be limited in their power to intervene in the market or in the entrepreneurial activities of individuals (Gane, 2012). 3 Castree (2008) refers to neoliberalism as ''the fetishising of markets as forms of social organization''. 4 One of the processes popularly associated with neoliberalisation is deregulation. This entails the scaling back of states and their capacity to regulate (Igoe and Brockington, 2007) and interfere in areas of social and environmental life (Castree, 2008), as well as the rolling back of redistributive reforms of the mid-twentieth century that were implemented in most advanced industrial economies (Castree, 2008citing Harvey, 1989). This has resulted in an 'off-loading' of responsibilities to the private sector and/or civil society groups (Castree, 2008).
For Bulkeley and Jordan (2012) the 'hollowing out' or 'rolling back' of the nation state has given rise to a situation whereby its functions are redistributed 'upwards' to international institutions, 'downwards' to regional and local tiers of authority [decentralisation], 5 and 'outwards' to a range of non-state actors. Thus, state boundaries have become more politically and economically permeable to decisions and flows emanating from diverse, overlapping and integrated networks of power which operate beyond effective control by formal structures of government (Bulkeley and Jordan, 2012;Goodwin, 1998). The messiness and scope of this complexity raises questions concerning legitimacy and power (Goodwin, 1998) which have implications for DRR governance.

NGOs and the 'outwards' redistribution of 'state' functions
Non-governmental organisations (NGOs) have found a new place within the neoliberal global order in terms of the 'outwards' redistribution of state functions (Frewer, 2013). 6 NGOs are an extremely diverse group of independent organisations that are neither run by government nor driven by profit (Lewis and Kanji, 2009). 7 Poorer countries receiving official development assistance have seen an increased proportion of aid being channelled to and through NGOs (Suleiman, 2013, citing Edwards andHulme, 1996;Wallace et al., 1997) as part of the shift from government to governance, meaning not only have they proliferated in number, but their relative position has increased in relation to the state (Bebbington, 2005). 8 Lewis and Kanji (2009: 5) attribute the rise of the NGO against the ascendancy of the neoliberal policy agendas to their high degree of flexibility in institutional form and the wide spectrum of values. International donors have tended to view national and international NGOs as being more efficient than governmental actors as: they tend to work at the grassroots level with poorer and more marginalised groups adopting participatory approaches; they have principled motivations, clear mission and objectives, progressive development agendas and lower overhead costs (Suleiman, 2013;Davis and Murdie, 2012;Benson et al., 2001;Hulme and Edwards, 1997;Lewis and Kanji, 2009). NGOs are often seen as synonymous with, or representative of, 'civil society', acting as a pluralist counterweight to state power (Frewer, 2013;Benson et al., 2001).
However, the enthusiasm towards NGOs as a vehicle for development has for a long time been under scrutiny within academia (Frewer, 2013). NGOs tend to reflect and reproduce both the hierarchies of the societies in which they are embedded, such as neopatrimonial networks, clientelism, corruption and inequality (Frewer, 2013;Suleiman, 2013;O'Reilly, 2010). Thus where the legitimacy of poor country governments is questionable, their civil societies may be equally disenfranchised, deeply divided, and ill 1 'Governance' is used in both a descriptive and a normative sense . In a descriptive sense, it refers to the complex web of manifold interactions between heterogeneous actors pertaining to a particular policy domain . In a normative sense, 'good governance' is an umbrella term for any package of public sector reforms or actions in civil society designed to create lasting and positive changes in accordance with the principles of: participation, representation, deliberation, accountability, empowerment, predictability, coherence, social justice, and organisation features Ahrens and Rudolph, 2006;Lebel et al., 2006). Good governance in the context of disaster risk reduction includes the adoption and promotion of robust and sound policies, legislation, coordination mechanisms and regulatory frameworks, and the creation of an enabling environment that is characterized by appropriate decision making processes to allow effective participation of stakeholders, complemented by the appropriate allocation of resources (WMO, 2012).
2 It is beyond the scope of this paper to consider these in depth (see Peck, 2008, for such as discussion -cited by Gane, 2012). 3 According to Medina (2010), a neoliberal rationality of government departs from classical liberalism on two key points. First, rather than proposing that the state limit its interventions into markets, neoliberalism proposes that the market form itself should serve as ''the organizational principle for the state'' (citing Lemke, 2001:200).
In a more Foucauldain reading of neoliberal ideology, neoliberalism addresses the appropriate powers of the state and the role it should play in ensuring the freedom of the market (Gane, 2012) which may entail ''a constant push to define and regulate social life through principles that come from the market'' (Gane, 2012). Many conceive of neoliberalism then, as a reorganisation of the role and practices of the state in relation to the economy, not a decline in state power (MacNeil and Paterson, 2012). 4 Castree (2008) highlights 'ideal-type characterisations' of neoliberalism as including privatization, marketisation, degregulation, re-regulation and the creation of market proxies. Of most relevance to this paper is the characterisation he refers to as ''the construction of flanking mechanisms in civil society'' which relates to the encouragement of charities, NGOs, 'communities' etc to provide services that interventionist states have provided in the past. 5 This is not considered in depth as it is beyond the scope of this paper but is intended to form the basis of future work. 6 A considerable proportion of donor funding now by-passes the state. Multilateral aid (channelled for example to the UN) stands at around 30% of official development assistance (OECD, 2010). It is argued that multilateral aid may be more effective than bilateral aid due to economies of scale, political neutrality and legitimacy, large scale of capital and knowledge resources, lower unit costs, and the provision of public goods. On the other hand institutional complexity, procedures which can be cumbersome or time-consuming, lack of transparency, higher absolute costs and salaries, remoteness and lack of accountability characterise multilateral institutions (OECD, 2010). 7 NGOs play very different roles and take different shapes and forms within and across different country contexts. In terms of structure, NGOs may be large or small formal or informal, flexible or bureaucratic. In terms of values, NGOs may be secular or faith based, vehicles for progressive change or may be regarded as part of market solutions to policy problems. They may play roles as implementers (goods and services, delivery role, contracted to provide specific tasks); catalysts (inspire, facilitate or contribute to promote change -lobbying, advocacy, research, policy influence) and partners (with government, donors, private sector, building capacity) (Lewis and Kanji, 2009 equipped to participate effectively in politics (Grindle, 2004). Furthermore, NGOs have been criticised for: functioning in a similar way to businesses to compete for financial support from donors; facilitating neo-liberal policy change (or at least being co-opted to serve hegemonic development agendas); being more accountable to donors (upwards accountability) than 'civil society'; being self-interested or based on family associations; lacking coordination and generating fragmentation and duplication; and their economic dependence on external funding which has sustainability implications (Ebrahim, 2003;Lewis and Kanji, 2009;Suleiman, 2013;Townsend et al., 2004;Booth, 2011).
Transnational networks and the 'upwards' redistribution of 'state' functions At the same time as there has been a redistribution of state functions 'outwards' to non-government actors, there has also been an expanding influence from 'beyond the state' (or 'upwards') through transnational networks of actors and institutions (Betsill and Bulkeley, 2004). These networks (such as global platforms in the case of DRR) mobilize information, knowledge, and values with the objective of integrating new conceptions of (environmental) phenomena into everyday worldviews and practices of private and public actors (Lipschutz, 1997:443 cited by Betsill and Bulkeley, 2004). These 'networks' can influence the direction of donor funding and approaches taken by multi-lateral agencies (such as promoting the principles in the Hyogo Framework for Action).

Governing beyond the state in disaster risk reduction
Authors such as Bulkeley and Jordan (2012) note that despite the shift towards 'governing beyond the state' where the role of transnational networks becomes of greater importance, it is not a matter of 'governing without the state' as the state is 'still the centre of considerable political power' (citing Pierre and Peters, 2000: 12). Indeed, within disaster management, the role of the state is seen as critical. Christoplos et al. (2001: 187), for example, regard governments and local institutions to be the most important set of actors in disaster management. Yet DRR, as the following section illustrates, is very much a global agenda which is being driven in poorer countries by international organisations such as the UN and funded by donors. 9 While donor support may be very valuable in areas such as disaster management where state expertise may be lacking or capacity is weak (Rajan, 2002) it may also simultaneously present a challenge for government 'ownership', 10 as there may be a 'latent' power evident in the way donor partners prioritise global agendas (Hyden, 2008). Thus, paradoxically, according to Berg (1993, cited by Godfrey et al., 2002) the donor or supply driven nature of technical assistance can actually lead to: weak local ownership and hence limited commitment; disorderly decision making; deepening dependence on foreign experts and weakening rather than building local institutions and capacities, exacerbating the tendencies of neoliberalism towards weakening the state.
Neoliberalism and the role of donors in development in weak states 11 An underlying rationale for the imposition of the neoliberal project on poor countries may be based on the assumption that corrupt and inefficient states restrict economic growth, so that if states were less intrusive, people's lives would naturally improve (Igoe and Brockington, 2007). Yet as Mosse (2013: 236) notes, in the age of neo-liberal reform, the terms weak state, fragile 12 state, crisis state and collapsed state are especially prominent in the development policy lexicon. While not explicitly highlighting the mechanisms underlying such a connection, he argues that the preferred solutions to the question of how market-led development was to be governed in poor countries with weak states, was through the donor coordination (of regulatory institutions, decentralisation, democratic process). The capacity of weak states to make effective use of aid has been questioned however, and is a contentious issue. Grindle (2011) suggests that strengthening non-governmental forms of governance makes sense for failed or collapsed states as there is very little upon which to build, but weakly institutionalized states have more upon which to build and strengthening government structures may best serve as sites for intervention. 13 This paper, then, examines in detail the tensions that arise between the need for strong centralised state support for, and expertise in, DRR and disaster management and the neo-liberal agenda which has 'hollowed out' the state. This research aims to explore the extent to which a shift from 'government' to 'governance' is evident in the DRR process and the degree to which such a shift may constitute power-sharing between the state and the multitude of other stakeholders emerging on the DRR stage. The role of international organisations and donors in advancing the DRR agenda is examined along with the implications that this may have for state 'ownership' over the DRR process. The challenges and constraints to progressing the DRR agenda in the context of a post-conflict state with 'weak' governance are also explored. The following section discusses the broader international agenda and transnational networks driving DRR processes.

Disaster risk reduction, governance and the international agenda
Disaster management frameworks such as the Hyogo Framework for Action (HFA) 2005-2015 have played an important role 9 National policies in turn are often influenced by such international policy concerns and funding (Ensor et al., 2009) particularly in less 'traditional' sectors such as environmental governance (Persson, 2009) and disaster management. This is particularly the case in areas such as international environmental governance (Persson, 2009) and disaster management but perhaps less so in the more traditional welfare sectors (for example Ensor et al.'s (2009) work on policy implementation in the health sector in Nepal suggests it was more domestically led than donor led). Persson (2009 citing Laugen andLunde, 1994) argues that the proliferation of overseas development assistance policy goals in the 1980s and 1990s (poverty, reduction, gender, environment and to this list can be added disaster risk reduction) result in a cognitive burden and 'mainstreaming fatigue'. With respect to environmental policy, mainstreaming (for which there are many parallels with DRR), Persson (2009) notes that the process is hampered by lack of recipient country institutional capacity, weak environmental ministries and a lack of commitment in the face of other development priorities.
10 Recipient governments can be said to 'own' an aid activity when they believe it empowers them and serves their interests (Godfrey et al., 2002:357, citing van de Walle and Johnston, 1996). 11 It is relevant to note that the concept of a 'weak state' has been challenged. Gainsborough (2010) for example notes that states can be 'strong and weak depending on where and how we look'. They may exhibit gaps in security, performance and legitimacy, but still be capable of repression and authoritarianism (GSDRC, 2013). Furthermore in the absence of a formal state, informal forms of security, governance and order, may emerge (GSDRC, 2013). 12 OECD's definition of fragile state is a state with weak capacity to carry out the basic state functions of governing a population and its territory and that lacks the ability or political will to develop mutually constructive and reinforcing relations with society (OECD, 2013). According to the World Bank, ''Fragile states'' is the term used for countries facing particularly severe development challenges: weak institutional capacity, poor governance, and political instability (World Bank, 2013a). The World Bank's definition of fragile states covers low-income countries scoring 3.2 and below on the Country Policy and Institutional Assessment (CPIA). Nepal falls just outside this category with a score of 3.28 (World Bank, 2013b). 13 The state of knowledge and practice regarding the establishment of effective governance mechanisms in post-conflict countries is still in its infancy (Brinkerhoff, 2005), but the relative importance of the state vis-à-vis NGOs in rebuilding legitimacy, security and effectiveness and the influence of external donors over such processes are key areas of interest (Rondinelli and Montgomery, 2005).
in advancing the agenda for DRR (Djalante et al., 2012). 14 The HFA was adopted by 168 countries (including Nepal) and supports the efforts of nations and communities to become more resilient to, and cope better with, the hazards that threaten their development gains. The HFA recognises that sustainable development, poverty reduction, good governance and DRR are mutually supportive objectives (UNISDR, 2007a) and sets out a number of goals and priorities, summarized in Table 1. DRR governance, upon which this paper focuses, relates to HFA Priority 1, which is concerned with, in particular, institutional and legal frameworks for DRR, and resource availability. Central to this is the establishment at global, regional and national levels of what are commonly known as 'platforms'. These platforms bring together the stakeholders (both government and non-government) engaged in DRR to share knowledge, experience and expertise. National Platforms are nationally owned and led, and ''aim to contribute to the establishment and development of a comprehensive national DRR system as appropriate to each country'' (UNISDR, 2007b: 4). The HFA promotes decentralization to the sub-national level to the provincial, district and ward level; while the National Platform feeds ''upwards'' into the Regional and Global Platforms.
For Gaillard and Mercer (2012: 94), the HFA is ''a non-binding treaty with no concrete targets. . ..too vague to entail concrete outcomes at the national level'', a sentiment shared by others (see, for example, Lavell and Maskrey, 2013). The UNISDR's bi-annual Global Assessment Reports (2009, 2011 and 2013) share the findings of individual countries' self-reporting against a series of core indicators. While countries have reported significant progress against Priority 1 through the establishment of national policies, legal frameworks and coordination mechanisms (UNISDR, 2013), it was recognised by a number of reporting countries that having these frameworks in place does not necessarily translate into effective DRR on the ground. Barriers reported by participating countries to the implementation of Priority 1 include a lack of political will, with DRR not seen as a top priority; the struggle to mobilise sufficient resources to support DRR at different levels; and uneven local capacities impacting on community participation and decentralisation (UNISDR, 2013). Similar challenges were highlighted by Manyena et al. (2013), who reviewed disaster legislation in five countries (India, South Africa, the Philippines, United Kingdom and Zimbabwe) post-2002. For Djalante et al. (2012), whose review focused on Indonesia, a lack of commitment from government to mainstream DRR into broader development agendas was a notable impediment to implementation.

The governance and disaster context in Nepal
Located along the Himalayan Arc, Nepal is highly susceptible to floods, landslides, glacial lake outburst floods and earthquakes (Petley et al., 2007;Nepal Disaster Report, 2011). In addition Nepal is classified as a low income country and a low human development nation, ranked 157 out of 187 countries in the composite Human Development Index (UNDP, 2013). High levels of poverty and social inequality based on ethnic and caste-based discrimination prevail (Murshed and Gates, 2005;Von Einsiedel et al., 2012). This combination of geophysical and social vulnerability renders Nepal highly susceptible to a high magnitude disaster, with Kathmandu classified as one of the most high risk cities to earthquakes worldwide (GeoHazards International, 2001).
Understanding the socio-cultural and political context in Nepal provides much needed insight into the ongoing governance and development challenges faced. Over the past six decades, Nepal has been caught in an internal struggle for peace and development, culminating in a decade-long Maoist insurgency between 1996(Von Einsiedel et al., 2012. The insurgency is argued to have arisen from a democratic deficit after 1990 in which new leaders failed to institutionalise democracy by promoting inclusion, representation and responsiveness (Thapa and Sharma, 2011) and was reinforced by ethnic and socio-economic cleavages (Bohara et al., 2006). Despite replacing its centuries-old monarchy with a federal republic, drafting an interim constitution and holding successful elections in 2008 -all relatively peacefully (Transparency International, 2009), the extent to which the democratic deficit has been addressed is questionable. According to Transparency International (2009: 12) ''[P]olitical instability, lawlessness, nepotism, clientelism, and lack of accountability prevail in the society and corruption is perceived to be a major concern''. In 2012, the Failed State Index, placed Nepal 27th overall and in the bottom 20 countries in terms of state delegitimisation (Fund for Peace, 2012).
Appointed following elections in 2008, the interim Maoist-led government (or Constituent Assembly -CA) was dissolved in May 2012 having failed to draft a new constitution during its term of office. A caretaker government was subsequently appointed, headed by the Chief Justice of the Supreme Court, and charged with holding new elections in 2013. In the absence of a parliament for more than a year, the government has been only partially functional. As noted by the IFRC (2011: 7), in a review of legislation related to DRR in Nepal, ''legislative reform processes have slowed and become less clear, as has the process for implementation of new policy''. With a new CA elected, and in place, it is hoped that progress will be made.
Like other conflict-affected countries, development assistance is increasing in Nepal and now makes up about 26% of Nepal's national budget (MoF, 2013). In 2011-12 Nepal received development assistance totalling $1.04 billion from over 40 donors. The top five multi-lateral development partners include the World Bank Group, ADB, UN Country Team, EU and the Global Fund. The top five bilateral partners include the UK, India, Japan, Norway and Germany (MoF, 2013). The flow of bilateral aid from DfID's expenditure delivered through multilateral agencies and NGOs increased from 19% in 2007/8 to 40% in 201040% in /11 (UKAid, 2012; while Pandey (2012) suggests that 30% of all donor money is channelled to NGOs. According to Yogi (2012), foreign aid has supported a rapid proliferation of local NGOs in Nepal. The number of local, national and international NGOs in Nepal stood at 221 in Table 1 HFA priorities and actions. Source: UNISDR (2007: 6) Priority/theme Action 1. Governance Ensure that DRR is a national and a local priority with a strong institutional basis for implementation 2. Risk assessment Identify, assess and monitor disaster risks and enhance early warning 3. Knowledge and education Use knowledge, innovation and education to build a culture of safety and resilience at all levels 4. Risk management and vulnerability reduction Reduce the underlying risk factors 5. Disaster preparedness and response Strengthen disaster preparedness for effective response at all levels 14 The shift from a response-led, 'expert-outsider' driven, top-down, technologycentred interventionist and structural approach in disaster management to a disaster risk reduction agenda that emphasises reducing vulnerability and building resilience by focusing on prevention and preparedness is not considered in depth here as it is discussed elsewhere in the literature (Maskrey, 1989 With such a strong reliance on overseas development assistance, and in the absence of a fully functioning government, the growing influence of donors in Nepal and the NGOs that they support, has given rise to considerable debate within Nepal -particularly regarding ethnic tensions that may arise from donor support for 'inclusive democracy', and neo-liberal development policies 15 (Manandhar, 2013;Bhatta, 2013;Newar, 2013;IRIN, 2013). 16 At the same time there has been no official mechanism to monitor the transparency of NGOs (of which there are 36,000 registered with the Social Welfare Council) as the anticorruption agency (the Commission for the Investigation of Abuse of Authority -CIAA) currently only monitors the activities of the government (Manandhar, 2013). In addition, the position of chief of the CIAA and national auditor-general are still vacant after seven years, and there is no Public Accounts Committee -a parliamentary body that tracks spending of donor monies (IRIN, 2013). Tighter controls have been proposed by government authorities with the aim of ''regulating the inflow of many unaccounted billions of rupees into the country every year'' (The Himalayan Times, 2014). This will involve the SWC enhancing its monitoring capacity, introducing a centralised database system, collecting registration fees and amending the Social Welfare Act (The Himalayan Times, 2014).
The nature of aid flows through Nepal have been largely development rather than humanitarian oriented (Kellett and Sparks, 2012), reflecting the fact that, unlike other countries in the region, Nepal has not experienced a high magnitude disaster for many decades. By comparison, in Bangladesh, which is often cited as an exemplar for DRR governance in the developing world (Ahrens and Rudolph, 2006), the high frequency of disaster events and subsequent humanitarian assistance has led to a transformational approach to DRR in country (UNDP, 2011).

Methodology
The findings of this paper are based primarily upon 26 qualitative interviews conducted by the first and second authors in June and September 2011 and March 2013, and a participatory workshop in January 2013. All stakeholders that were deemed to be important in DRR arising from the participatory workshop were interviewed. These included: government ministries, multi-lateral organisations such as UNDP, international and national NGOs, and bi-lateral and multi-lateral donors. The paper is also informed by informal conversations, Nepali grey literature and policy documents. Questions explored the nature of the role of different stakeholders in advancing the DRR agenda and how these have changed over time to examine the extent to which new actors are emerging in the DRR arena (a shift from government to governance). How stakeholders feel about these changes (e.g. whether government bodies felt a lack of a sense of ownership in a donor driven agenda); who has power in governing DRR policies and practices and the challenges of working in a post-conflict context were key areas explored in the interviews.

Discussion
The section below presents a historical overview of disaster management in Nepal and draws on the reflections of key informants to highlight the convoluted nature of policy processes in the advancement of the disaster risk reduction agenda. The subsequent section reflects further on the dynamics of stakeholder relationships and the power struggles they entail.

Policy processes: Changes in the legislative and policy context of DRR in Nepal
Disasters in Nepal were traditionally managed on an ad-hoc basis and attended to as and when they occurred. In 1982 the Natural Disaster Relief Act (also known as the Natural Calamities Act) was ratified, leading to the establishment of Central, Regional, District and Local Level Natural Disaster Relief Committees, as well a system for allocating relief funds (MoHA, 1982). The Act also set out clear organisational responsibilities, giving the Ministry of Home Affairs (hereafter the Home Ministry) responsibility for the formulation of national policies and their implementation; preparedness and mitigation activities; immediate rescue and relief work; data collection and dissemination; and the distribution of disaster relief funds (Pradhan, 2007). This Act remains in place with only minor amendments having been made since it was passed (Aryal and Dobson, 2011).
In 1996, the Government of Nepal (GoN) produced the National Action Plan for Disaster Risk Management (MoHA, 1996) in accordance with the International Decade of Natural Disaster Reduction. This plan deals with different stages of a disaster (pre, during and post) and was, in theory, supported by the 1999 Local Self Governance Act which advocated devolution of responsibility to lower levels of the politico-administrative hierarchy. However, without any guidance or specific funds provided to local government, this had little impact (Pradhan, 2007).
In 2005, the HFA provided a much needed stimulus for change. In 2006 two separate initiatives were instigated: the Disaster Management Policy and Act; and the National Strategy for Disaster Risk Management (hereafter the National Strategy). Both initiatives reflect to varying degrees the wider paradigm shift from relief and response towards DRR and propose new institutional structures to facilitate this shift, recognising that while it is appropriate for the Home Ministry to continue to manage response and recovery efforts in the country (as they command the police and the 15 Nepal also introduced Structural Adjustment Policies in the mid 1980s. An 'independent' evaluation by the World Bank suggested that a dispute with India in early 1989 disrupted trade and created losses in output (Yogi, 2012). 16 It is beyond the scope of this paper to explore the extent to which neoliberalism may have weakened or contributed to the fragility of the Nepalese state. Suffice it to say that some authors consider there to be a direct relationship between neoliberalism and fragility in Nepal; some consider the relationship to be more indirect; while others would attribute the cause of fragility to more internal factors such as the ethnic tensions discussed above. Suggesting a direct relationship between neoliberalism and fragility, Dahal (2012: 9), for instance, argues that Nepal's mix of feudal and neoliberal features into its political culture oiled by patronage ''further exposed the state into fragility''. He points to the neoliberal regime having ''privatised over three dozen national industries for rent seeking, dichotomised health, education and communication to discriminate poor citizens from their very birth and adopted financial capitalism in tune with external advice than national need'' (2012: 8). Presenting a more indirect link between neoliberalism and fragility in Nepal, Bhatta (2013) writing for the Telegraph Nepal, argues that ''Reality is such that donor supported developmental policies after 1990s (the neo-liberal policies) and political agendas (inclusive democracy) after 2006 have pushed Nepal to political chaos. Neoliberal policies inundated Nepal with NGOs who promoted 'rights based' social society. These very (I)NGOs later floated political agendas such as federalism, and secularism, which have now become major bones of contention. It is widely believed that these agendas would become source of latent conflicts in Nepal''. Bhatta (2013) blames the international community for having caused ''regime instability, weakened the state institutions and divided people along different social lines''. Some politicians, analysts and journalists have also painted Nepal's international donors as instigators of ethnic tension (IRIN, 2013). Thus, rather than being a direct consequence of neoliberalism's economic logic, this relates more one of Castree's (2008) 'ideal-type' characteristics of neoliberalism -the 'construction of flanking mechanisms in civil society' leading to an increasingly important role for NGOs -and the specific type of development engagement that donors have been promoting. army and are the most powerful ministry), their capacity for dealing with DRR is limited.
The first of the two initiatives emerged when Oxfam funded a national NGO (The National Centre for Disaster Management) to draft a new Policy and Act. A consultative exercise was undertaken involving key stakeholders and a series of workshops within and outside Kathmandu in the five regional centres. The National Policy was written in 2007 and presented to the National Planning Commission and to various ministries for consideration. However, there were many changes in government and as noted by a representative from a national NGO ''nobody knows what happened in the past. . .all the time [we] remind them [the government] and give new set of documents''. As a consequence, the Policy seems almost lost from the memory of all but a few people who were involved in it. The simultaneous process of drafting the new Act was highly contested among interviewees with some suggesting that several reputable lawyers were involved, while others attributed the drafting of the document to a single lawyer who had just completed his law degree. This proposed new legislation was promoted through a symposium and a series of workshops with members of the CA. However, due to the volatility of the CA, the passing of legislation was stalled. Respondents noted that politicians had other priorities at the time, principally drafting a new constitution.
The second initiative, the National Strategy, was coordinated by UNDP and funded by the European Commission's Humanitarian Aid Department, with a national NGO, the National Society for Earthquake Technology (NSET), serving as the technical consultant. It is not clear whether UNDP was aware of the already written National Policy which could have provided a starting point for developing the new legislative framework for disaster risk management, but it is clear that a separate process occurred in the formulation of the National Strategy.
As part of the formulation of the National Strategy, a meeting was held on the 26 February 2007 bringing together 136 representatives from government, non-government and international organisations. The National Strategy clearly arose from a thoroughly consultative and participatory process. It was intended to ''facilitate the fulfilment of the commitments made by Nepal through various international conventions and forums towards DRR'' (NSDRM, 2008: iv), particularly the HFA by putting forth ''suggestions regarding institutional reorganisation and development and strategic improvement in the existing policy and legal environment'' (NSDRM, 2008: iv). The National Strategy appears to be widely accepted and supported at the national level, with the stakeholders interviewed expressing an alignment with the National Strategy and no criticism of either the process or the content. Some interviewees argued that the National Strategy was necessary to guide the implementation of DRR activities in the absence of a supportive legislative context. For some this amounted to a need to ''bypass the government'' to enable DRR implementation.
The status and content of the current draft Act remains unclear. A representative from the Home Ministry explained that the current draft Act is based on the 2009 approved National Strategy but does not include all its recommendations as this would have severe legal repercussions. It was noted in an IFRC report in 2011 that the revised draft Act will soon go to Cabinet and be made publically available, however this still appears to be the situation. Without the Act, there is no budgetary allocation to DRR (Aryal and Dobson, 2011) which hinders the progress that the government can make in effectively implementing DRR.
It is against this backdrop of a well-regarded National Strategy and a stagnant legislative context that the Nepal Risk Reduction Consortium (hereafter, the Consortium) was instigated in 2009 by UNDP, and officially launched by the GoN in 2011 (NRRC, 2013). Based on the Hyogo Framework and the National Strategy, the Consortium aims to support the GoN to develop a long-term disaster risk management action plan (NRRC, 2013). The Consortium consists of multiple stakeholders working within the humanitarian and development sectors, including the government, UN agencies, NGOs, international donors and communities. It has a steering committee comprising 13 partners and more than 100 organisations and government entities contributing to its programmes.
The Consortium has five priority or flagship areas (Table 3). Some interviewees complained that the choice of flagship programmes had not been made in a transparent or participatory way and was very much led by the UN. While there have been notable achievements under the five flagship areas, some of the Consortium's more ambitious targets e.g., large-scale retrofitting of hospitals and schools, will not be met fully by 2015 (Taylor et al., 2013). 17 This has been attributed to a range of factors including the scale and breadth of the planned activities; bureaucratic challenges; and tensions between stakeholders who have not traditionally worked together including the development and humanitarian sectors (Taylor et al., 2013). However, the majority of stakeholders interviewed commended the work of the Consortium, suggesting that it has been particularly successful in raising the profile of DRR within government and among other stakeholder groups, and is potentially attracting further funding for DRR.
Following a review of the Consortium in 2013 (a few months after the interviews for this paper were undertaken), the Steering Committee, which includes senior government officials, took the decision that the Consortium will continue beyond 2015 to 2020. However, the impact on the Consortium of the recent elections and the Disaster Management Act, if passed, remains unclear.

Governance struggles: Stakeholder relationships and power dynamics
In a recent article for the Guardian newspaper, Robert Piper wrote: ''After five years of working on this in Nepal, I have come to recognise that addressing Nepal's vulnerability to natural hazards is first a governance problem'' (Piper, 2013). This section discusses further some of the dynamics of governance of DRR in post-conflict Nepal and the dynamics of power and influence of the various stakeholder groups underlying these policy outcomes.

Challenges brought about by the proliferation of NGOs and non-state actors
The emergence of a multi-stakeholder governance landscape for DRR is very much evident in Nepal and is unsurprising given the rise of NGOs. Although some organisations have been working in Nepal in the field of disaster management for considerable time, 18 over the last decade or so, there has clearly been a proliferation of NGOs engaging in DRR. Many of these were already engaged in development work in Nepal and increasingly turned their focus to DRR. 19 In some areas this has given rise to duplication and inefficiency and the development of the National Policy and the National 17 The Consortium has an ambitious funding target. The estimated total budget of the five-year programme is US$ 195.8 million (NRRC, 2013) and by August 2012 funding for the consortium had reached US$ 65.2 million (Harris et al., 2013). Only a third of funds have been raised so far. 18 UNDP's first projects, for example, were some twenty years ago and the Nepal Red Cross Society have engaged in humanitarian work since 1963, and began disaster preparedness activities since the mid 1970s. Nepal's national NGO the National Society for Earthquake Technology was founded in 1993. 19 To give an indication of the scale of NGO activity in DRR in Nepal, the Association of International NGOs (AIN) in Nepal has 110 members and 12 claim on AIN's website to engage in aspects of disaster management (response, relief, risk reduction, preparedness). In addition, 57 local partners, 33 implementing partners and 29 donors have registered with Flagship 4 (Flagship 4, 2013). DPNet-Nepal has 82 members including members from government, faith-based organisations, NGOs, municipalities, civil society organisations, UN departments and military departments. Strategy concurrently by different organisations is a case in point. As a representative from a UN organisation remarked ''when I came I noticed that there was one group of people that were more involved in the Strategy and another group of people that were more involved in the Act. . ..certain people were more champions of one and other people were more champions of the other''. It is evident that some animosity arose through this process. One interviewee, for example, suggested that UNDP denied that the draft Act had ever been presented to the government by Oxfam, in what another respondent described as ''a battle of egos''.
In other respects, NGOs are presented as filling a gap, particularly in terms of implementation, that the government is unable to address. As one international NGO representative noted: ''currently the DRR works that's being done in Nepal implemented at the field level it's more done by the NGO sector and still we don't have in government''. The focus on NGOs by donors was justified in terms of a low level of government capacity: ''We are trying to work through the government, build government's disaster risk management capacity. .. but systems simply don't exist currently'' (donor representative). This view is shared by the NGO community: ''NGOs are there because the government is not doing so -if the government's perfect there is no need of NGO'' (INGO representative). The government does not necessarily agree with this situation however, as a representative of a bi-lateral donor organisation noted: ''the government doesn't want the money to be going to international NGOs -they understandably want it to be going through them, so there is always this tightrope really and there is a lot of tension''.
Comments made by interviewees about the problems that arise from channelling money down to local NGOs reflect those in the literature. For example, one independent consultant noted that ''there are zillions of NGOs [in Nepal] -it's kind of an entrepreneurial thing''. A representative from a bi-lateral donor described the situation as ''a bit of a festival of NGOs because...they know that there is the resilience thing and donors are interested in this and everybody wants to try and get involved''. Coordination of the many local implementing NGOs working on the ground is challenging not only between organisations, but within organisations. It was claimed that in some areas, different departments in donor organisations ''don't even know quite what they're supporting as an organisation -the left hand not knowing what the right hand is doing'' and their projects in particular areas may have no connection with each other, creating overlap and wasted resources. A donor representative confirmed that ''donor coordination at the best of times is very, very difficult.'' The uneven distribution of NGOs is also problematic.
A well functioning state would at least enable a consistent approach to DRR planning at the local level and facilitate coverage across the whole country. Guidelines have been produced by the Development Ministry, supported by Oxfam, to standardise disaster risk management planning at the lower levels of government which offers opportunities for a more systematic approach. Furthermore Flagship 4 is tracking CBDRR implementation across Nepal with a view to enhancing coordination and consistency. It is advocating 9 characteristics of a disaster resilient community as a minimum standard for community based DRR projects. It is hoped that such initiatives will improve donor coordination and strengthen NGO-led DRR projects.
While donors tend to fund international NGOs, which in turn work with local NGOs on the ground for project implementation, some of the national NGOs argue that they are being by-passed by donors. As one director of a national NGO put it ''. . .the West has trained us. . .there has been a significant, you know, capacity built up here. Compare [national NGO] with any other organisation in the West. We are strong. Why should we be barred from competing?'' With regards to the Consortium specifically, a key criticism was the way the agenda was being taken over by international organisations and foreign experts, and for some, undermining existing national capacity. There is a question though, of whether existing national level institutions might have been equally able to perform the Consortium's current functions. For example, the Disaster Preparedness Network Nepal (DPNet-Nepal) a multi-stakeholder network engaged in disaster preparedness was identified as a possible mechanism for this. However, its capacity was not deemed by some stakeholders interviewed to be particularly high. A representative of a national NGO noted that ''the flagship programme is really under the, you can say, patronage, umbrella, whatever you call it, of the UNDP. That has a very kind of, you know, influence on the government and on the international scene. . . [we] need that kind of leverage. [With DPNet-Nepal] that kind of weight is not there. It [DPNet-Nepal] can do a lot of work, it can really mobilise a lot of people it has the connections and membership of so many of the major stakeholders. . ..but then at the same time it's after all an NGO''.
Similarly it was suggested by some interviewees that UNDP might have invested their efforts in developing an effective National Platform to align to UNISDR's recommendations, rather than creating a new institutional structure in the form of the Consortium. This was also posed as a question by a representative of the Consortium: ''So I asked about the National Platform. . .and I was told that it hadn't been fully functional in the last few years and that there was also some confusion around its status, given that it had been created as a NGO''. One interviewee noted that the documentation for the Global Platform had been produced by an outside consultant in the absence of a functioning National Platform. It could be argued then that the Consortium was created as a parallel structure because the capacity and authority did not exist at the national level to fulfil the functions that the Consortium currently serves and this would have taken too long to build. Robert Piper, the then UN Resident and Humanitarian Coordinator for Nepal, has mentioned elsewhere that the urgency of the situation warranted high level, effective and rapid intervention. For example in a UNISDR meeting in Paris, he noted that ''it's a race against time to save as many lives as possible before Nepal faces its next large scale disaster. That is why we have put together the Nepal Risk Reduction Consortium'' (McClean, 2012).
Government ownership, power struggles, capacity and commitment As more than one interviewee explained, the Home Ministry is not a development ministry and 'you can't manage disasters by batons and guns'. A representative from a NGO explained that ''it is imperative that other development ministries such as the health, education, public works, and local development should be made capable enough to lead disaster risk management initiatives as development agenda. Managing disaster risk is a development challenge''. It is thoroughly logical therefore for the Ministry of Federal Affairs and Local Development (hereafter the Development Ministry) to District guidelines for local level mainstreaming underway be granted the responsibility for rolling out DRR to the local level due to their role in local development and leading the shift from disaster response to DRR. The Home Ministry, however, seems reluctant to relinquish any power.
The Home Ministry has demonstrated its power (and perhaps its resistance to be dictated to) in a number of other ways. For example, changes were made to the draft National Strategy by government before approval. The National Strategy proposed that the new national authority should sit directly under the Prime Minister's Office where it would have authority over other ministries to mainstream DRR. If the new authority sits within the Home Ministry it will not have the constitutional or budgetary powers to effectively enable mainstreaming throughout other ministries. The approved National Strategy is somewhat ambiguous stating that 'the Government of Nepal will build upon existing structures and gradually transform them into new structures as necessary' (MoHA, 2009: 75). Table 2 documents other changes (e.g. at the District level) illustrating the Home Ministry's efforts to retain power.
While there are still few signs within government of progress towards a national disaster management authority headed by the Prime Minister, there is movement within the Home Ministry towards creating a larger division. This would be an opportunity to build a higher capacity within government and provided some ground for optimism among some of the interviewees. It was suggested that the Consortium could be redundant in the future as it could be replaced by a National Platform that the government would be strong enough to coordinate. Others argued however, that the Consortium is reaching the limits of what it can achieve without the government's full commitment. As a representative of the Consortium explained: ''[We said to the government] we are losing interest. You are talking to me about the fact that there needs to be more resources and I am telling you very frankly there aren't going to be more resources without a sign that the government is serious and that the government is putting in more absorptive capacity.'' A key challenge currently faced is the extremely low staffing and limited DRR expertise in the Home Ministry. Currently the Home Ministry's Disaster Management Section consists of only three staff members who were not disaster management experts when they come into post. They also have other responsibilities besides disaster management. Despite the training that international organisations offer (for example, the UNDP as part of Flagship 5) the rapid through flow of staff creates a challenge for building expertise within the government. A larger division demonstrates some effort by the government to increase its absorptive capacity but is a long way from a national authority under the Prime Minister's office.
In some interviews, a more harmonious picture was presented and there appeared to be little evidence of a conflicting agenda between government and donors and multilateral agencies. A representative from the Disaster Management Section in the Home Ministry was quite clear that they lead on legal and policy matters around DRR and did not regard donors as influencing their practice. One Home Ministry representative noted ''We don't like to say like 'influences' because 'influences' has something like a negative connotation. . .What 'inspire' us to change our policy?''. He explained proudly that they had fed into the Global Platform evidence of good practice in Nepal -thus both shaping and being informed by practices elsewhere that may enable them to improve practice. He did however acknowledge the significance of donor funding: ''the national government doesn't have lot of funds for the prevention of disaster activity and if we follow their activity or their recommendations then it is easy for us to get the fund from outside''. As a UN representative explained with respect to DRR ''The Government Ministries have a choice whether or not they want to do this.. .There are sticks and carrots in the whole process and there's no doubt that the fact that there are resources there helps.'' While 'sticks and carrots' are clearly quite significant in the DRR process, it is difficult to locate centres of power or identify who is leading the process, as a degree of circularity seems to be in operation. Many donor and multilateral agencies laboured the point that DRR and the Consortium were 'government led' to convey a sense of ownership. Donors claim that they ''fund the gaps'' or ''respond to government requests'' and that the ''government leads'' but at the same time they shape government priorities. For example, a representative of the Consortium noted that ''if people are speaking about it at the same time, and people are hopefully using the same words, and the same rationale then hopefully they [the government] get a sense that this is something that is very important to the international community. Which, incidentally, we're only doing on the behalf of our partners in government -MoHA [the Home Ministry] has requested it and we're supporting them in that request''. Similarly a representative from a bi-lateral donor noted that ''we need to try and help the government to prioritise more so that we can properly support that''. A representative from a multi-lateral donor explained that ''investments on reducing risks has to be led by government. Development partners cannot be of support, unless it is the government's priority.'' Thus donors and multi-lateral agencies avoid using a language that makes it appear that they are leading DRR (while at the same time threatening to withdraw funding if the government do not show more commitment). Without these interventions, Nepal would be much further behind in addressing HFA priorities. The government is not being by-passed or ignored in the process of supporting DRR. While being classed as a 'fragile' state, it was pointed out by many interviewees that the state is powerful when it wants to achieve something (such as the road widening scheme in Kathmandu), and will not engage in anything it does not want to do. But donor funding is a clear 'carrot' in the DRR process.

Conclusions
''It's all a bit messy, I think, to an outsider looking in. Depending on your point of view you might say it's. . .been a fantastic achievement in terms of bringing those groups together. Pessimistically you could say that we're all falling all over each other and there's no real coordination. I mean that's one of the challenges of development actually, for the most part.'' [Bi-lateral donor representative] Drawing a conclusion to the governance struggles and policy processes evident in DRR in Nepal presents quite a challenge. Opinions were found to be manifold and conflicting, with each interviewee presenting a different perspective on the challenges and achievements that have been made. As the quotation above suggests, it is very easy to view the same situation from different positions. A single, consensual 'truth' does not exist. It is hoped, at least, that this research has provided insights into the complexity of processes around policy formation in DRR in a post-conflict country.
To some degree, in line with the neoliberal approach, a shift from government to governance is evident in the context of DRR in Nepal. This includes the influence of 'upwards' transnational networks (implementing the HFA), along with the growing 'outwards' role of non-government actors in DRR. As a country relying heavily on donor funding, the UN and donor organisations have a great deal of influence on the development agenda in Nepal. It is apparent that the GoN would not be engaging in DRR without the pressure and resources from the international community. 20 In the absence of state level expertise and capacity, international and national NGOs also play an important role in the implementation of the DRR programmes, while the Consortium plays a vital coordination role in what has been quite a haphazard DRR landscape due to the proliferation of NGOs. Despite this, the state is still the centre of political power in Nepal.
Tensions were noted arising from the complex interplay of power, knowledge and differential access to resources of the many different stakeholders engaged in the governance of DRR. The significant role that non-state actors play creates challenges for government ownership. There is some concern that the progress made through the Consortium ''reduces the incentive to centralise the issue as a core priority for the GoN, leaving it generally underserved in national budgeting processes'' (Taylor et al., 2013: 38). In addition, the sometimes hegemonic approach of the international community to DRR is the source of tension between the international community and national NGOs who in some cases felt as though their expertise and experience was not valued. Representatives from the UN and donor organisations emphasised repeatedly that they were working at the request of, and in partnership with, the GoN and that they are working through existing governance structures, and drawing on national expertise, where and when they could. However, it was also acknowledged that these mechanisms and expertise do not always exist. This could be a priority area of capacity-building for donors, via the Consortium, as part of its exit strategy.
What, then, are the implications of this messiness and complexity? There are aspects of DRR that require strong centralised governance, for example establishing and overseeing national emergency operating centres and maintaining critical lifeline services. This role does not necessarily need to be played by the state, but it does require a long-term vision, resources and authority. Currently the Consortium is playing a strong coordinating role in this area. However, this raises questions regarding the sustainability of disaster risk governance in Nepal. If the Consortium exits too soon, it may leave very little legacy; however if it stays too long, there is a danger that national capacity and ownership will not be developed. Findings suggest that the GoN needs to accord higher priority to developing institutional structures such as a National Disaster Management Authority and a National Platform so they can own the process and lead future resilience building in Nepal, but they are still some way from genuine ownership and institutionalisation. Brass's (2012: 229) conclusions from Kenya are relevant here: ''One hopes to avoid a mass exodus of international funds until the capacity of local NGOs, government and civil society has grown to fill the space occupied by them''. It is likely that other post-conflict states embarking on donor-driven DRR will face similar challenges.