We are in the midst of the federal restructuring of the state and the basic parameters of our federal model are under intense discussion. Although the Political Dialogue and Consensus Committee chaired by Baburam Bhattarai falied to settle the contentious issues of the constitution and missed its September 6 deadline, the political parties have agreed to provide more time to forge consensus on disputes. The three major political parties have also agreed to hold a national political conference aimed at finding consensus in relation to the demands of the CPN-Maoist, including other forces outside the Constituent Assembly (CA). Hence, it is unlikely that the new constitution will be drafted by January 2015, the self-imposed deadline of the political parties.

In the first CA, identity came under intense discussion and in this respect, five criteria were established—ethnic, dialectal, cultural, geographical, and historical identities. However, consultations were focussed mostly on ethnicity, with widespread misunderstandings that identity was equivalent to ethnicity. This caused great confusion among stakeholders and eventually led to the dissolution of the first CA.

Furthermore, economic interrelationships and capability, development of infrastructure and viability, availability of natural resources, and administration accessibility were the four criteria established in relation to capability. It is needless to mention that there is a lack of district-level data related to identity and capability, and the main basis for state operation is revenue, which is still constrained and limited.

Towards fewer provinces

During the first CA, a majority in the CA committee and the state restructuring commission proposed 14 and 12 provinces respectively. On the contrary, merely a few days before the dissolution of the first CA, the political parties had agreed to an 11 provinces federal model. However, the major political parties are still seriously discussing the issue of provinces, and they are more or less close to consensus on a federal model with fewer provinces.

In this context, the report on Natural Resources, Economic Rights and Revenue Sharing has proposed customs duty, value added tax, corporate income tax, and personal income tax, among others, for the central government. Similarly, local-level revenue items, such as entertainment tax, fees and fines, vehicle tax, service fees, sales tax, and land tax have been recommended for both provincial and local governments. The revenue headings presently received by the local bodies, namely land registration, electricity, mountaineering, and forests have been proposed for all three levels of government—federal, provincial and local. Most revenue headings presently exploited by local bodies have been proposed for the provincial government. The report has actually proposed very few revenue assignments for local governments.

On the basis of the division of revenues proposed for the federal, provincial and local governments, it can be analysed whether the districts will be able to afford administrative expenses. This article provides an analysis on finding economically-viable districts based on the revenues of Village Development Committees (VDCs), Municipalities (Muns) and District Development Committees (DDCs) for the fiscal years 2011/12 and 2012/13. Moreover, depending on the average internal revenues of these local bodies, provinces have been proposed. The district revenue data used in this article was derived from the internal revenues of the three sets of local bodies, ie, VDCs, Muns and DDCs.

According to this data, the Kathmandu and Bajura districts score the highest and the lowest internal revenues respectively. In overall district revenue, Kathmandu has a share of 15.67 percent. In a central treasury/revenue system, this district covers 28.83 percent. Kathmandu district has the largest population (1.7 million) with Jorpati VDC, which has the highest population and collects the largest revenue, falling in this district. Needless to say, Kathmandu district dominates most sectors.

Low viability

The share of internal revenue in relation to total district expenses is only 13.90 percent. This clearly reveals that most districts in Nepal are dependent on central government grants. Districts which the largest proportion of internal revenue compared to the total expenses are Kathmandu (48.72 percent), Bhaktpur (43.72 percent), Rupandehi (43.40 percent), Dhading (41.66 percent), Lalitpur (36.69 percent), and Makwanpur (31.61 percent). On the other end of the scale, Bajura, Achham, Kalikot and Mugu score below 2 percent.

Among the 75 districts, the average proportion of internal revenue compared to total district expenses is less than 10 percent. However, district expenses between 20-30 percent for administrative purposes are a rule of thumb. The Financial Report of the Local Bodies Fiscal Commission has derived this figure to be 23.18 percent. But even if we assume the lower limit of 20 percent as a benchmark for administrative expenses, only 18 districts are economically sustainable. If we reduce the benchmark by a further five percent to 15 percent administrative expenses, the number of economically sustainable districts increases to 29, which comprises Kathmandu, Rupandehi, Bhaktpur, Lalitpur, Dhading, Morang, Makwanpur, Kaski, Kapilbastu, Jhapa, Chitwan, Sarlahi, Bhara, Banke, Sunsari, Navalparasi, Dang, Kanchanpur, Parsa, Kailali, Kavrepalanchock, Nuwakot, Solukhumbu, Sindupalchock, Surkhet, Dhanusa, Mustang, Tanahu, and Bardiya. Out of these 29 districts, four are from the Eastern Development Region; 13 from the Central Development Region; six from the Western Development Region; four from the Mid-Western Development Region and two from the Far Western Development Region. The remaining 46 districts are not economically sustainable under a 15 percent benchmark. Based on that benchmark, a maximally viable map of provinces has been developed and put up for discussion.

A seven province model

In the course of the discussion on the division of provinces, the major concerns are the number of provinces, their geographical boundaries, and their names. Here, seven provinces with identity-based names and their geographical boundaries have been designed in such a way that they are optimally balanced in terms of population, revenue and expenses, and will be able to sustain the administrative expenses by their own revenue as far as possible. Out of the seven provinces illustrated in the map, only four are economically viable, which are Narayani, Bagmati, Lumbini-Gandaki and Koshi. The other provinces proposed in the map need to be supported with grants from the central government to cover their administrative expenses as they are not economically viable. The most strongly dependent province is Bheri-Karnali followed by Seti-Mahakali and Sagarmatha-Janakpur. Based on the existing structure of districts, the status of internal revenue and other indicators related to economic viability, it can be argued that the country should not be divided into more than four provinces.

Devkota holds a PhD in fiscal decentralisation and is a former member of the Local Bodies Fiscal Commission