Nepal’s Quest for Federalism: A Driver of New Violence

EU and German policymakers often promote federalism as a means to strengthen peace in societies emerging from civil war. In Nepal, however, the 2015 constitution which restructured the country to make it a federal republic has become a driver of new violence. This contribution argues that it is not only the substance of the federalist law that has inspired unrest, but also the manner in which it was introduced.

• Constitutional provisions stipulating geography-based federalism in Nepal sparked protests among Madhesis and Tharus that killed 57 people and stalled imports of petroleum and earthquake relief material from India from September 2015 to February 2016.

• The Madhesis and Tharus feel that the federalist law further discriminates against them and are demanding the delineation of federal units according to identity. While the constitution was initially perceived as a major breakthrough in the frequently stalled peace process with former Maoist rebels, the Madhesi and Tharu protests represent a new escalation in a long-polarised debate.

• Three factors have contributed to the violent escalation of the protests: (1) the dominance in the public debate of hill-upper-caste groups at the expense of marginalised groups; (2) the concurrence of the federalist reform with other institutional reforms that, as a whole, have sparked fears of discrimination among Madhesis and Tharus; and (3) increasing involvement on the part of China and India, which is further polarising the government and the marginalised groups.

• The example of Nepal reflects and accentuates the recent debate on post-war institutional reform – that is, that reforms have joint effects on peace rather than working in isolation.

Policy Implications

Policymakers must ensure that a future solution to minority demands in Nepal is identified based on dialogue that includes minorities to a larger extent; any elite proposition could lead to new unrest. Nepal also demonstrates that subnational and national institutional reforms as a whole can ease or exacerbate minority fears. Donors should pay attention to such interaction rather than limiting projects to one policy field.

Violent Unrest in Post-Earthquake Nepal

On 20 September 2015, 89 per cent of Nepal’s Constituent Assembly representatives approved a new constitution for the Himalayan country. The urgent need to begin reconstruction after the devastating 7.8- and 7.3-magnitude earthquakes on 25 April and 12 May 2015, which killed over 8,700 people, had accelerated a long-stalled constitutional debate. The final document represented a breakthrough in the peace negotiations with former Maoist rebels, which had begun nine years earlier with the signing of the Comprehensive Peace Agreement (CPA). A central provision of the new constitution is the restructuring of Nepal into seven federal provinces that are delineated according to physical geographical characteristics, as well as on the basis of existing administrative divisions.

Federalist reform has long been one of the most contentious issues in Nepal’s constitutional process, but since August 2015, Nepalese society has become ever more polarised on the question of whether federal states should be delineated according to identity rather than geography. Identity-based demands are an integral element of Nepalese politics, and cleavages as based on various identity markers – such as ethnicity, religion, gender, region, or caste – making it difficult to give exact figures on how strongly all groups are represented in the population. The 2011 census, for instance, records 125 caste and ethnic groups, 123 languages spoken as a mother tongue, and 10 religious groups. Table 1 outlines the most important identity groups for the purpose of this contribution.

In August 2015, the Tharu and Madhesi communities from Nepal’s southern Tarai plains – identity groups who have long held substantial resentments against the state for its failure to address their political and economic exclusion and systematic discrimination – began protesting against the constitutional provisions for federalism.

These groups argued that the federal structures stipulated in the new constitution would oppress their interests because they would, for instance, result in unequal resource distribution and their political underrepresentation at the centre and would thereby further marginalise the already marginalised. Their accompanying demands focused on four central issues: (1) identity-based federalism instead of geography-based federalism, (2) increased proportional representation within state institutions and the electoral system, (3) equal citizenship rights for men and women, and (3) the delineation of electoral constituencies based on population density.

In order to make their demands heard, the United Democratic Madhesi Front (UDMF) – an umbrella organisation for Madhesi parties – called for the blockading of customs checkpoints at the Nepal–India border on 23 September 2015, effectively stalling imports of petroleum, medicine, food, and earthquake relief material from India. The poor and those affected by the earthquake were hit the hardest. Nepal’s government immediately accused the Indian government of supporting the blockade, not least because the long-standing cultural, family, and socio-economic links of Nepal’s Madhesi population to the north-Indian state of Bihar mean that Madhesis are often not seen as “true” Nepalis. Some hill-based elites feared that granting Madhesis more autonomy would lead to the disintegration of Nepal.

In November 2015, it was reported that the border blockade had resulted in a worse economic crisis than the earthquake and UNICEF (2015) warned that a shortage of food and medicine meant that three million children were at risk of death or disease. By January 2016, 57 people – including eight police officers – had died in clashes between officers and protesters in the Tarai. The government and the protesting parties had already met for several rounds of talks, but when the largest political parties unilaterally amended the constitution on 23 January 2016 to increase the Madhesi presence in government bodies through proportional representation, the UDMF boycotted the vote and argued that this amendment was incomplete, as it did not address its key demand: the revision of federal boundaries. On 8 February 2016, four days after the end of the 135-day blockade, however, the UDMF formally announced an end to protest activities, partly because many of its protesters were so frustrated and disillusioned by the process that they were less and less willing to block the border checkpoints.

While the first constitutional amendment played a role in ending the blockade, it did not solve the underlying conflict, and the UDMF has stated that it will con­tinue to protest until Tharus and Madhesis are granted separate states. The elites from the main political parties in Kathmandu strongly oppose this demand and argue that such identity-based federalism would create “ethnic ghettoes” and increase social divisions in Nepal (ICG 2016). The risk of escalating violence in the Tarai thus remains, and given that the Tarai has a history of armed groups (there were more than 100 armed groups active in 2009), continued disillusionment on the part of the Madhesis might drive an angry population towards radical groups. Thus, while the federalist provisions in the new constitution represent the symbolic end of the peace process with Maoist rebels, they have at the same time driven new violence, thereby raising the following questions: What makes federalism a problem rather than a solution in Nepal? How can the crisis be solved? And what lessons can we learn from Nepal for the future promotion of federalism in post-war states?

Federalism as an Instrument for Post-War Peacebuilding

In Nepal, it was the Maoist rebels belonging to the Communist Party of Nepal (Maoist) who, after waging a decade-long civil war against the government from 1996 to 2006, demanded a new constitution that would restructure the country into a federalist republic. The Maoist request thereby reflected a global trend, as federalism has increasingly emerged since the end of the Cold War as a tool for peacebuilding promoted by international policymakers and a demand made by warring parties engaged in violence. It has been discussed or implemented in places as diverse as Bosnia and Herzegovina (in the 1995 Dayton Agreement); Sudan (where it was a key issue in various peace agreements from 2004 to 2007); and more recently Syria, where Kurdish parties in particular have regularly called for federalism as a solution to the ongoing civil war.

Several analyses have promoted federalism as an effective tool for post-war peacebuilding and regard it as a vital ingredient for the stabilisation of peace after war. These proponents stress that federalist arrangements that devolve political authority from the central government to federal provinces mitigate fears of exploitation among marginalised identity groups by limiting discriminatory practices within state institutions and providing for a system of checks and balances (Lake and Rothchild 2005). These authors also argue that the federal restructuring of post-war states addresses the economic insecurities of marginalised groups, increases efficiency within the war-torn civic administration system, and generally satisfies local demands for cultural and political autonomy.

In contrast, opponents of the implementation of federalist structures after war argue that such arrangements can create parallel structures of political authority that further weaken war-torn societies if tasks are not clearly defined and separated between political levels (Wolff 2009). These authors further believe that federalism impairs prospects for post-war peace because it reinforces divisive group identities (for instance, between ethnic groups in Bosnia), and that it creates new instability and secessionist demands – for instance, because it provides regional identity groups with the ability to finance and mobilise for new violence to a greater extent than in centralised states. These authors have also stressed that federalism on its own is often insufficient to generate long-term solutions to civil war and that federalist reforms, if they are to stand a chance of furthering the peace process, must be embedded in broader institutional reform processes at the central government level. Empirical evidence shows that in post-war societies, federalism is most helpful when combined with other conflict resolution instruments, particularly power-sharing arrangements within the central government (Wolff 2009).

The Evolution of the Federalist Debate in Nepal

The debate in Nepal reflects these opposing positions voiced within the academic community. While the debate has been the single most divisive issue in Nepal’s political discourse since the CPA, federalism is by no means a new topic; it has been passionately discussed since it was first proposed in the 1940s. At that time, Nepal was ruled through an authoritarian system of hereditary prime ministers within the Rana family that systematically benefited Chhetris and Bahuns – the hill-upper-caste Hindus who continue to make up the political and economic elite today. The Madhesis from the southern Tarai, on the other hand, had to obtain written permission to enter the capital Kathmandu (Hachhethu 2007). After a short period of democracy in the 1950s, authoritarian rule was reinstated and remained in place until 1990, when the first People’s Movement started a transition to democracy.

But even with the rise of democracy in the early 1990s, the traditional ruling elite of male, high-caste hill-origin Hindus monopolised the political system, and Nepal remained a unitary monarchy with Nepali as its sole official language. Demands for federalism and increased autonomy continued to be made in the Tarai, where Madhesis and Tharus still experienced the state apparatus as distant and largely unresponsive to their grievances. Well-paid jobs in the administration were held by high-caste groups, and the security forces in the Tarai – particularly the police – were largely made up of hill-origin officers. Peaceful movements in the early 1990s called for increased representation and autonomy rights for the various identity groups, and these movements began to influence the political discourse on how to restructure the state.

From 1996 onwards, it was the Maoist rebels who capitalised on the aspirations of marginalised communities. Their insurgency boosted the federalist agenda in the Tarai, as the rebels induced Madhesis and Tharus to join their campaign by promising to fight for these identity groups’ federal autonomy. In November 2006, the Maoists and the Seven Party Alliance – a coalition of the biggest political parties in Kathmandu – signed the CPA, and Nepal was declared a republic following the Maoists’ electoral victory in April 2008. The Maoists then politically promoted the idea of delineating 13 federal provinces, the majority on the basis of identity groups.

While the transition to federalism was a predominantly Maoist aim after the CPA, today most parties – some more enthusiastically than others – have accepted the idea that Nepal should become a federalist republic, and federalism was thus mentioned in almost all party manifestos in the 2008 and 2013 Constitutional Assembly (CA) elections. Nevertheless, the parties differ considerably in their ideas of what type of federalism should be implemented. The Maoist 13-provinces identity-based model has been heavily criticised by the country’s two largest political parties, the Nepali Congress (NC) and the Communist Party of Nepal – Unified Marxist Leninist (UML). Both parties cite the critical perspective on federalism as brought forward by the academic literature and argue that an identity-based system would ruin Nepal’s unity and create divisions in a society that has not seen major ethnic clashes in the past (ICG 2011): “Nepali Congress will never accept an ethnic-based federalism. This federalism will ruin Nepal’s unity and stability,” said late NC president Sushil Koirala (Spotlight Nepal 2012). NC and UML instead contend that geographical boundaries with a mix of identity groups are the best way to ensure the viability of federal units.

In 2012, differences between the parties meant that the State Restructuring High Level Recommendation Commission – formed by the political parties represented in the CA to suggest an appropriate model of federalism for Nepal – failed to agree on one model and submitted two reports. One report promoted the division of Nepal into 11 provinces based on identity, and the second report proposed restructuring Nepal into six provinces based on economic and geographical characteristics. These differences are highlighted in Figure 1, which plots the level of support for federalism in general (left) and for delineating units according to identity (right). The graphs show that while there is strong support for federalist reform throughout society, the views of upper-caste, hill-based Hindus on identity-based federalism differ from those of Madhesi respondents.

These divisions between the groups were so strong that throughout the rule of the interim government (2007–2008) and the first CA (2008–2012), the Maoists were unable to bring forward the policies they had promised the Madhesis and Tharus during the war. The polarisation of positions on the topic even led to the dissolution of the first CA in 2012 without a constitution.

The failure of the Maoists to push for identity-based federalism also triggered an episode of violence in 2007: as it became apparent that the interim constitution did not fully address Madhesi demands, protests erupted in the Tarai, resulting in the deaths of over 30 people. The result was an eight-point agreement between the interim government and the UDMF accepting the Madhesis’ call for an autonomous “Madhesh” province and thus their desire for identity-based federalism. The Maoists only moved closer to the NC and UML positions as a result of the 16-point compromise deal of 8 June 2015 on constitutional provisions in the aftermath of the 2015 earthquakes.

Factors behind the 2015 Escalation of Violence

The main political parties thus arrived at a peaceful answer to the federalist question by promulgating the September 2015 constitution and deferring some of the contentious issues to be decided later. But this much awaited promulgation did not deepen social cohesion in Nepal. Instead, identity groups radicalised, and the chasms between the state and its citizens in the Tarai deepened. Several elements of the constitution-making process explain why the Madhesis and Tharus have felt increasingly sidelined, thus leading to the violence in 2015 and 2016. First and foremost, the Tarai communities have perceived the federalist debate as being severely dominated by upper-caste hill-origin Hindu politicians in Kathmandu, at the expense of marginalised groups’ voices and even as a result of the suppression of their voices through violent police action against protesters from August 2015 onwards. The Madhesi and Tharu communities – which fought in the Maoists’ People’s War in the hope that identity-based federalism would be the outcome of a Maoist victory – also perceive the Maoists’ acceptance of the geographical division of federal units under the new constitution as a major betrayal. They feel abandoned by their representatives in the capital.

The perception of elite domination of the discourse and the fear of being patronised by upper-caste hill-origin Hindus goes beyond the political division of Kathmandu from the periphery: the Tharus and Madhesis also fear that the federalist boundaries have been delineated to suit local political and economic ruling elites in the designated federal states (Bhatt and Murshed 2009). This is partly because the lack of proportional representation in national state institutions has in the past given increased political leverage to upper-caste Chhetris and Bahuns from Nepal’s regions, instead of to the historically marginalised communities. The latter expect that identity-based federalism would expand their political representation while limiting that of the traditional ruling classes, and that upper-caste elites within identity-based federal units would inevitably have to cede some power to minorities and give Madhesis, Tharus, and Janajatis majority rule over some provinces.

As a result, Madhesis and Tharus perceive the current geographical delineation as further undermining their political representation, because the units divide these groups so that they do not make up a majority in any of the provinces, which favours the ruling elites. Particularly the Tharus have voiced their discontent at the prospect of being split in two, and the Madhesis say the current model draws borders directly through their ancestral homeland. They also point out that resource distribution is highly unequal across the proposed units, with a potential long-term impact on their development. For instance, two of the provinces do not share a border with China, which is expected to put them at a distinct trade disadvantage in the long run.

The delineation of federal provinces has also coincided with a number of institutional reforms and constitutional provisions that have accelerated fears among the Madhesis and Tharus that they will be pushed further to the margins, because these reforms reverse many of their past achievements. These fears relate to the debate on how federalism works in concordance with other reforms, underlining recent findings that post-war institutional reforms have joint effects on peace rather than working in isolation.

In particular, the federalist reforms have been accompanied by the introduction of draconian citizenship laws that hurt the rights of both women and Madhesis, and of Madhesi women in particular. Prior to the 2006 Nepal Citizenship Act, only a father with Nepalese citizenship could transfer this citizenship to a child. Article 11.2.b of the 2015 constitution states that a child whose father or mother is Nepalese can become a Nepalese citizen – it had been father and mother in previous drafts. However, legal experts argue that other clauses in the constitution override this statement. For instance, for women with Nepalese citizenship – but not men – the constitution requires children born in Nepal to obtain citizenship by descent, but while the children of Nepalese men married to a foreigner are entitled to citizenship by descent, those of women married to a foreigner can only obtain citizenship through naturalisation. Madhesis – who often intermarry with families on the other side of the border in north India’s Bihar – perceive this law to be discriminatory.

In addition to citizenship requirements, Nepal’s constitution originally also introduced an electoral law that dropped previous regulations on proportional representation and that would have left Tarai groups underrepresented in the parliament. Specifically, under the 2015 constitution, a smaller percentage of parliamentary seats were to be elected through proportional representation than under the 2007 interim constitution (45 per cent as compared to 58 per cent of seats).

Finally, the role of regional actors also complicates the situation. Nepal’s neighbours India and China, which are competing for influence within Nepal, both perceive themselves as being affected by federal reforms in Nepal. Both countries want fewer provinces on their border so that it is easier to deal with security issues. The Chinese government is said to strongly oppose identity-based federalism because it fears a spillover of such ideas across Nepal’s border to the Tibet Autonomous Region. It has thus suggested to Nepal’s government that identity-based federalism should be avoided. During the blockade, Nepal turned to the Chinese government to seek new import opportunities, something which is logistically difficult due to the challenges of trading across the Himalayas. China has doubled its emergency aid to Nepal since the 2015 earthquake and will likely use Nepal’s dependency on Chinese aid to support parties opposed to identity-based federalism. The Indian government, on the other hand, is generally in favour of federalism in Nepal, but – because it historically shares deep cultural ties with the Madhesis – is said to support the Madhesi struggle for identity-based federalism. It hosted rare high-profile visits of Madhesi leaders to India in 2015, and it has conveyed its interests through local Madhesi parties. India has more influence than China in Nepal’s internal affairs, though it has denied direct involvement in the border blockade. The influence of both regional powers can be regarded as affecting the opportunity structure of the state and of the Tarai protesters by further deepening already existing divisions and contributing to the polarisation of the debate within Nepal.

Requirements for a Peaceful Solution

At first glance, the federalist debate in Nepal appears to be deadlocked particularly due to substantive differences about whether federal units should be delineated according to geography or identity; however, to a large extent it is also the manner in which the federalist laws have been introduced that has driven the recent violence. Both sides in the debate have presented reasonable arguments for why they prefer one federalist model over the other, but due to elitism, the combination of laws that have accelerated and increased perceptions of marginalisation among ­Madhesis and Tharus, and the further polarising role of regional powers, the sides have thus far failed to engage in a constructive dialogue. The federalist restructuring of Nepal has therefore deepened rather than mitigated the cleavages between the state and its citizens in the Tarai. Highly exaggerated and emotional claims on the benefits and pitfalls of federalist restructuring have further polarised the debate. These findings are of interest for both scholars and practitioners, who often look at the consequences of the actual design of post-war institutions but not at the process of producing these institutional reforms.

Based on these findings, the most central and basic recommendation for policy­makers and the international development community in Nepal is to ensure that any future solutions to the remaining demands of Madhesis and Tharus in Nepal are identified through a political dialogue that includes the marginalised groups in the Tarai to a significantly larger extent than has been the case thus far. As long as these groups feel marginalised in the process of designing future federal structures, as they do now, any proposition from the political elites in Kathmandu could lead to new violent unrest in the Tarai plains. The final model of federalist restructuring in Nepal – identity-based, geography-based, or a mix of both – should feel like a win-win situation to all parties involved, which is something that only an inclusive political dialogue can achieve. Substantively, policymakers should also ensure the wide-ranging resolution of minority demands instead of undertaking minor amendments to single laws. In particular, they should pursue a citizenship law that does not result in double discrimination against women and Madhesis.

More generally, the case of Nepal and the combined effect that constitutional provisions regarding (1) federalism, (2) citizenship, (3) a delineation of electoral constituencies, and (4) proportional representation have on marginalised groups’ fears of being sidelined provide a broader lesson for international policymakers: institutional reforms at the subnational and national level do not work in isolation but ­rather interact closely, and together they mitigate or accelerate minority fears. Donors should thus pay greater attention to the interaction of post-war institutional reforms on the ground rather than funding projects limited to one policy field or sector.

• Bhatt, Shiv Raj, and Syed Mansoob Murshed (2009), Nepal: Federalism for Lasting Peace, in: Journal of Law and Conflict Resolution, 1, 6, 121–140.

• Karki, Budhi, and Rohan Edrisinha (eds) (2014), The Federalism Debate in Nepal: Post Peace Agreement Constitution Making in Nepal, volume II, Kathmandu: United Nations Development Program.

• Gautam, Kul Chandra (2015), Lost in Translation: Rebuilding Nepal from the Maoist Mayhem and Mega Earthquake:, Kathmandu: Nepalaya.

• Hangen, Susan, and Lawoti Mahendra (2013), Nationalism and Ethnic Conflict in Nepal: Identities and Mobilization after 1990, New York: Routledge.

• Hachhethu, Krishna (2007), Madhesi Nationalism and Restructuring the Nepali State, Kirtipur: Tribhuvan University, Centre for Nepal and Asian Studies.

• International Crisis Group (ICG) (2011), Nepal: Identity Politics and Federalism, International Crisis Group Asia Report, 199, Brussels and Kathmandu: ICG.

• International Crisis Group (ICG) (2016), Nepal’s Divisive New Constitution: An Existential Crisis, International Crisis Group Asia Report, 276: Brussels, Delhi, and Kathmandu: ICG.

• International IDEA (2013), Citizen Survey 2013: Nepal in Transition, Stockholm: International Institute for Democracy and Electoral Assistance.

• Lake, David, and Donald Rothchild (2005), Territorial Decentralization and Civil War Settlements, in: Philip G. Roeder and Donald S. Rothchild (eds), Sustainable Peace: Power and Democracy after Civil Wars, Ithaca: Cornell University Press, 109–132.

• Spotlight Nepal (2012), Dispute on Federalism: Capability or Identity, 5, 21, 18 May, www.spotlightnepal.com/News/Article/Dispute-On-Federalism-Capability-Or-Identity (15 March 2016).

• UNICEF (2015), Nepal: Serious Shortage of Essential Supplies Threatens Millions of Children This Winter, press release, 30 November (15 March 2016).

• Wolff, Stefan (2009), Complex Power-Sharing and the Centrality of Territorial Self-Governance in Contemporary Conflict Settlements, in: Ethnopolitics, 8, 1, 27–45.