Skip to content

← Economics

Nepal's Economy

Chapter 28 · Nepal's Economy

In this chapter 4 sections
  1. Making and working
  2. Seven economies
  3. The constitution’s economy
  4. What it means, and how far it has gone

Follow a sack of rice to a plate in Kathmandu. It starts as paddy in a field in Dhanusha, planted in the monsoon and cut in the autumn by a family that eats some of it and sells the rest. A trader buys the surplus and trucks it to a mill in town, where machines strip off the husk and polish the grain. Then it is weighed into sacks with a brand printed on them, carried up the highway to a wholesaler in the valley, sold on to a shop in Kalimati, and finally cooked by a family, or by a hotel that charges a tourist for it with dal and vegetables on the side.

Economists sort the work along that road into three kinds. Growing the paddy is primary work, taking something straight from nature, as farming, forestry, fishing and mining do. Milling it is secondary work, turning materials into goods, as factories, power stations and builders do. Everything after that, the trucking, the trading, the shop, the hotel, along with teaching, banking, nursing and governing, is tertiary work: services. Every economy has all three, joined like the links of that road, and the mix tells you a good deal about a country (Geography’s Three kinds of work tells how the mix has shifted as countries grew rich).

Nepal’s mix has a shape of its own, and earlier chapters have already shown most of its pieces. What it makes comes to about Rs. 218,000 a person a year, by the measure of Chapter 13: a low income, though rising. Most of them still work on farms, many of them on plots too small to need all their hands (Chapter 24). Factories make little of what the country uses, so it buys far more from abroad than it sells (Chapter 20), and pays for the difference largely with the money that Nepalis working abroad send home (Chapter 32). Most work is informal, in family farms and small shops with no contract. Roads, power and water have improved fast but still leave much of the country hard to reach, and its forests, rivers and minerals are used far below what they could give (the subject of the next few chapters). And the country has been, since 2015, a federation, with seven provinces and 753 local governments, each with an economy of its own, under a constitution that says what kind of economy it should become.

This chapter puts those pieces side by side: what Nepal makes and who makes it, how the seven provinces compare, and what the constitution’s “socialism-oriented” economy is meant to mean.

Making and working

Start with what the country produces. In the year that ended in mid-July 2026, farming, forests, fishing and mining together gave about a quarter of the value Nepal added to everything it made. Factories, power, water and building gave about a seventh. Services gave the rest, about three-fifths: trade the largest of them, with transport, teaching, government, property, banks and hotels among the others.

Now ask who does the work, and the picture turns over. The two official counts don’t even agree with each other, and the difference is worth understanding.

Farms, forests, fishing, minesFactories, power, buildingServicesWhat was produced, 2025/2625%14%62%Everyone who worked, census 202158%12%30%People in paid work, survey 2017/1822%30%48%
Figure 28.1 Nepal’s three kinds of work counted three ways: the share of what was produced in 2025/26 (gross domestic product at basic prices, first estimate), the share of everyone aged 10 and over who did any work in the year before the 2021 census, own farm included, and the share of people in paid work in the Labour Force Survey of 2017/18. Mining is counted with farming in the first part of each bar, as the statistics office does.

The census of 2021 asked everyone aged 10 and over whether they had done any work in the past year, including on the family’s own land, whether or not it brought in money. Nearly fifteen million people said yes, and nearly three in five of them had worked mainly in farming, forests or fishing. The Labour Force Survey of 2017/18 used the stricter international rule of Chapter 24: a person is employed only if the work was for pay or profit. Growing food for the family’s own pot doesn’t count. By that rule only about a fifth of the employed were farming, and services took nearly half. The survey didn’t forget the farm. It counted separately more than eleven million people who had grown food for their own households, more than half of everyone aged 15 and over, and didn’t count that work as employment, since none of it was sold.

So both counts are right about different questions. The census says how most Nepalis spend their working days: on land. The survey says where the paid jobs are: in shops, on building sites, in schools and workshops. Neither is up to date. The survey is nearly nine years old, and a new one has been in the field since March 2026; the census counts Nepal once every ten years.

Put the bars together and the most important fact about Nepal’s economy appears, even allowing for the counts being from different years and the census including people who farmed only part of the year. By the census’s count, more than half of those who work produce about a quarter of the output. So each person’s work on a farm adds far less, on average, than work elsewhere. That is not because farmers work less. A day spent weeding a small plot with hand tools, on land watered only by the rain, simply yields less than a day at a loom or a bank counter. It is why poverty in Nepal is still mostly rural, and why moving work off the farm has been part of how the countries that grew rich did it (Chapter 21). Chapter 30 takes up why farming earns so little.

The same split runs through villages and towns, though less neatly than the words suggest. In the rural municipalities, गाउँपालिका, nearly three in four of the people who worked did so mainly on farms or in forests. The economy there is paddy and maize, buffalo and goats, firewood and fodder, with cottage industries, a little trade and, in places, homestays for visitors. In the municipalities, नगरपालिका, the share was lower, but still almost half. That surprises people until they remember that “municipality” in Nepal is a legal title, not a description: two in three Nepalis live in one, but by density alone only about a quarter live in anything like a city (Geography’s What counts as a city). In the real towns the work is what you would expect: trade and transport, factories and workshops, building and selling land and houses, hospitals, colleges, offices and banks. The two economies feed each other. Towns buy the countryside’s grain, milk and vegetables and sell it cement, phones, fertilizer and loans, and the young people who leave a village for a town, or for the Gulf, send money back to it.

Seven economies

Since 2018 Nepal’s national accounts have been split among the seven provinces, so each now has an estimate of its own output, its provincial GDP. Its first estimate for the year that ended in mid-July 2026 put the whole country’s output at about Rs. 6,600 billion, and divided it very unevenly.

Bagmati alone produced well over a third of it. Koshi produced about a sixth, Lumbini about a seventh, Madhesh about an eighth, Gandaki about a tenth, Sudurpashchim less than a tenth and Karnali about a twenty-fifth. Bagmati made more than Madhesh, Lumbini and Karnali together. Part of that is simply people, since provinces differ in size. Dividing by them gives a fairer comparison.

Farms, forests, fishing, minesFactories, power, buildingServices0100200300400Bagmati381Gandaki238Koshi203Lumbini174Sudurpashchim169Karnali160Madhesh135Nepal218GDP per person, thousand rupees
Figure 28.2 Gross domestic product per person in each province in 2025/26, first estimates, in thousands of rupees, each bar split by the share of the province’s output that came from each kind of work. The split is approximate: it applies the shares of output at basic prices to output per person at market prices.

Per person, Bagmati made about Rs. 381,000 in the year, nearly three times as much as Madhesh, at about Rs. 135,000. Only Bagmati and Gandaki were above the national average of about Rs. 218,000. In dollars, Bagmati’s figure was about $2,640 a person and Madhesh’s about $930.

The split inside each bar shows where the gap comes from. The part from farming, forests and mining is roughly the same size everywhere, between about Rs. 45,000 and Rs. 67,000 a person. What differs is services. Bagmati holds the capital, with its ministries, head offices, banks, hospitals, colleges and airport, and services made more than three-quarters of its output: the statistics office found trade its largest single activity, and property second. Everywhere else farming came first. Madhesh depended on it the most, for over a third of its output, and in 2025/26, when its farm output fell, it was the slowest-growing province, at about 1.3%, while Bagmati grew fastest, at about 5.4%.

The order is the familiar one from the rest of this book. Karnali and Sudurpashchim have the highest shares of their people below the poverty line, while Madhesh and Lumbini hold the most poor people (Chapter 22), and Madhesh had the lowest human development when the provinces were last compared (Chapter 21). Gandaki, a small province by people, comes second per person, with more of its output from factories, power and building than any other province.

These numbers are estimates in a deeper sense than most. The statistics office doesn’t add up each province’s output from its own records. It takes each industry’s national total and shares it out among the provinces by indicators, such as where the crops are grown, the factories registered and the teachers paid. The office itself says what this can’t do: there are few reliable indicators by province, and no suitable price indexes for them, so its figures can’t show how much prices differ between Jumla and Kathmandu. So the gap in output is not quite a gap in living standards, and nobody can yet say how far the two differ.

The provinces’ governments are new too. Most of their money comes from the federal government’s transfers (Chapter 19), and how much difference a province’s own policies can make to its economy is one of the questions federalism is still answering.

The constitution’s economy

Nepal’s constitution of 2015 does more than set up governments. It says what kind of economy the country should have. Its preamble commits the country to “socialism based on democratic norms and values”, लोकतान्त्रिक मूल्य र मान्यतामा आधारित समाजवाद, in order to build a prosperous nation. Article 4 lists what kind of state Nepal is: independent, indivisible, sovereign, secular, inclusive, democratic, “socialism-oriented”, समाजवाद उन्मुख, and a federal democratic republic. And Article 50, among the state’s guiding principles, gives its economic objective in one long sentence, here in the Law Commission’s English:

The economic objective of the State shall be to attain a sustainable economic development, while achieving rapid economic growth, by way of maximum mobilization of the available means and resources through participation and development of public, private and cooperatives; and to develop a socialism-oriented independent and prosperous economy while making the national economy independent, self-reliant and progressive in order to build an exploitation-free society by abolishing economic inequality through equitable distribution of the achievements made.

In plainer words: grow fast and for good, with the state, private business and cooperatives all taking part, share the gains fairly, and end up with an economy that stands on its own feet and leans toward socialism. That sentence is the source of what Nepal’s plans and school books call the three-pillar economic policy, तीन खम्बे अर्थनीति. The constitution itself doesn’t use the word “pillar”. It names three sectors, public, private and cooperative, and asks all three to take part and to grow. The policies that follow, in Article 51, spell it out: strengthen the national economy through the participation and free development of all three; value the private sector’s role; promote cooperatives and use them for national development; regulate for fairness and competition; and end black-marketing, monopoly and artificial shortages (Chapter 11).

Each pillar has a size, and they are very different.

The public sector is the government and the companies it owns. Besides its ministries, schools and hospitals, the state owned 45 public enterprises in mid-July 2025, by the Ministry of Finance’s latest count, among them the electricity authority, the oil corporation, the airline and the telecom company, with Rs. 799 billion of the government’s money in them. Sixteen of them made a loss, and the profitable ones paid the government about Rs. 8 billion in dividends. Of all the money Nepal spent on new buildings, roads, machines and other fixed investment in 2024/25, the government and its companies did about two-fifths. Their share swings from year to year with how much of the development budget gets spent (Chapter 19).

The private sector is everything else that is owned privately, and it is most of the economy. It is the farm family, the tea shop, the bus owner, the cement works, the private school and the commercial bank. It made most of the fixed investment, and nearly all the work Nepalis do, on family farms and in family businesses, is private. Most of it is small and informal.

The cooperative sector is the pillar with the most members. Nepal had about 32,000 cooperatives in mid-March 2026, by the Ministry of Finance’s count, for savings and credit, milk, farming, tea, coffee and much else, with about ten million memberships among them and about 82,000 people working for them. A dairy cooperative that collects milk from a hundred households and sells it to a chilling center does what no single farmer could. But the savings and credit cooperatives, run outside the central bank’s supervision, also produced one of Nepal’s worst financial failures: from 2022 many could not return their members’ savings, as Chapter 15 tells in Outside the walls. By mid-March 2026 the government had declared 23 of them troubled; about 80,000 people had claimed Rs. 48 billion from them, and fewer than 4,000 had been paid in full. A law of 2025 set up a separate regulator to supervise them.

What it means, and how far it has gone

What, then, is a “socialism-oriented” economy? The constitution doesn’t define it. A Nepali Congress politician who sat in the assembly that wrote it has said the phrase was part of a consensus with the two communist parties, and that his own party, which also counts socialism among its principles, understands it differently.

Nepal’s school books give the official reading. A socialism-oriented economy, they say, is not a fully socialist one in the sense of Chapter 4, where the state owns the means of production and a plan sets prices. It is a market and mixed economy, gradually oriented toward socialist values and goals. The state keeps the legal power to make rules, the private sector supplies most of the economic power to produce, and cooperatives add the social power of people pooling what they have. Its two bases are social justice, सामाजिक न्याय, and prosperity, समृद्धि. Social justice in this reading has four sides: distributing opportunities fairly; redistributing income, through taxes that take more from those who have more; remedying old disadvantages, through quotas and reserved places for groups left behind; and protecting the vulnerable, through allowances and cheaper health care for the old. Prosperity is a rising, fairly shared standard of living, which the plans summed up in the slogan “Prosperous Nepal, Happy Nepali”. Chapter 23 tells how far the measures of Bringing people in have narrowed Nepal’s gaps.

How far has it gone? Some of it is written into law. The constitution made employment, food and social security fundamental rights, with laws to carry them out within three years, and in 2018 Parliament passed a cluster of them: a Right to Employment Act, a Right to Food Act, a Social Security Act, an act for free and compulsory schooling, one for public health. The federal budget for 2026/27 set aside about Rs. 120 billion for social security programs, the old-age allowance of Rs. 4,000 a month the best known of them.

But much of the constitution’s economic chapter is a promise, not a rule. Article 55 says that no court may be asked whether anything in the part on the state’s guiding principles has been carried out. A committee of Parliament watches over them instead. The fundamental rights can be taken to court, but what they give depends on the laws and on the money. Meanwhile most of the economy is private and small, the state’s companies are mixed in their results, and many cooperatives’ members are still waiting for their savings.

People in Nepal disagree about what to make of this, and the argument grew louder before the elections of March 2026. The Federation of Nepalese Chambers of Commerce and Industry, the main body of the country’s businesses, urged the parties to take “socialism-oriented” and the three-pillar policy out of the constitution. Its president said the wording “no longer reflects Nepal’s economic realities or aspirations”, and that the private sector has always been wary of it; a former finance minister said it puts off investors from democratic countries. Even the Congress politician who remembered the consensus wanted the phrase revised. Those who would keep it reply that a goal isn’t wrong because it hasn’t been reached: dropping the phrase because it hasn’t delivered, one Marxist scholar said, would be like abolishing the anti-corruption commission because corruption is widespread. “The constitution does not belong to any single group,” he said, “and calling for changes on such grounds is misguided.” As of the latest reports we could open, the words stood as they were written in 2015. Changing them needs two-thirds of all the members of both houses of Parliament.

An economist can’t settle that argument, because it is partly about values: how much weight to give fairness against growth, and how far the state should go in shaping the market. But economics can say what each side would need to show. Those who want the words out would need evidence that they, rather than slow courts, unclear rules or poor roads, are what keeps investors away. Those who want them kept would need to show that they change what governments do: whether the allowance arrives, the school has teachers and the clinic has medicine. Both are questions about what happens, not what is written, and both can be checked.

Back on the road from Dhanusha, the sack of rice passes through all three pillars and all three kinds of work. It is grown on a private family farm, perhaps with seed and credit from a cooperative, carried on a road the state built, milled in a private factory, and sold in a market whose rules the state writes. Nepal’s economy is still mostly farms by the number of people and mostly services by the value made, mostly private, unevenly spread across seven provinces, and bound by a constitution that asks for growth and fairness together. How well those fit is the question the rest of this part of the book turns over, one sector at a time.

Found a mistake? Please tell us.