Skip to content

← Economics

Adding It All Up

Chapter 13 · The Whole Economy

In this chapter 5 sections
  1. Three ways to count one number
  2. From what is made to what a family can spend
  3. Money GDP and real GDP
  4. Why counting it in Nepal is hard
  5. What it leaves out

On her best day Anshu sells 280 glasses of tea and takes in Rs. 7,000. That one day can be counted in three different ways, and all three come out the same.

The first way asks what her shop made. Rs. 7,000 of tea went out across the counter, but Rs. 2,800 of it was milk, tea leaf, sugar and gas that other businesses had made and she had bought. What her shop added, by turning those things into hot tea in a glass, was Rs. 4,200. Economists call that the value added, मूल्य अभिवृद्धि, the same words that are in the name of VAT.

The second way asks who was paid. The same Rs. 4,200 went out again, to the people whose land, money and work the shop used: Rs. 500 for the room, Rs. 100 for the wear on the stove and pans and the interest on her loan, Rs. 1,400 to her two helpers, Rs. 700 for her own day, and Rs. 1,500 left over as profit. Add them up and you get Rs. 4,200 exactly. It can’t come out any other way. Profit is whatever is left after everyone else is paid, so what the shop adds and what it pays out are one sum seen from two sides.

The third way asks what was spent. Her customers spent Rs. 7,000 on tea. That looks bigger than Rs. 4,200, but the Rs. 2,800 she paid her suppliers was also value added, by the dairy, the tea garden, the sugar mill and the gas company, and each of them bought things from others in turn. Follow the chain all the way back and add what every business along it added, and the total is the Rs. 7,000 the drinkers paid at the end. Adding the dairy’s sales of milk to Anshu’s sales of tea would count the milk twice, so the spending count takes only what the final buyers paid, and the production count only what each business added. Some of those businesses were abroad: the gas came a long way. Their part was added somewhere else.

Done for a whole country, these are the three ways of counting what it makes in a year, its gross domestic product or GDP. By what is made, by what is earned, and by what is spent. Everything spent is someone’s income, and every income comes from something made, so in principle the three give one number.

Three ways to count one number

Counting by what is made is the production method, उत्पादन विधि. Nepal’s statistics office goes through every industry, works out what each produced and what it bought from others, and adds up the difference. For 2025/26 its first estimate, which it marks preliminary, put about Rs. 24 of every Rs. 100 added in farming, forests and fishing, about Rs. 14 in mines, factories, power and building, and about Rs. 62 in services: shops and traders above all, then schools, real estate, government offices, transport, banks and the rest. Then it adds one more thing. The price a buyer pays includes VAT and customs duties, which no producer keeps, so the office adds those taxes on goods, less any subsidies, to bring the count up to what buyers actually paid. They came to about Rs. 805 billion, and the total to Rs. 6,600 billion.

Counting by what is earned is the income method, आय विधि. Here the office sorts the same total by who received it. About 38% went to employees as wages and salaries. About 12% went to the government as taxes, mostly the VAT and duties on goods, less subsidies. The other half went to the owners of farms, shops and firms, as what the office calls operating surplus and mixed income. The second name is there for people like Anshu. When a family runs its own shop or farm, the owner’s pay for her day and her profit arrive as one sum, Rs. 2,200 on her best day, and nobody can say which part is which. Many Nepalis work for themselves or their families, so a large share of the country’s income is mixed in this way.

Counting by what is spent is the expenditure method, खर्च विधि. Every final good or service made in a year is bought by someone, and the buyers fall into four groups. Households buy food, clothes and bus rides; the government pays teachers, police and doctors; firms and the government buy buildings, machines and roads; and people abroad buy Nepal’s exports. One adjustment makes the sum work. Some of what Nepalis bought was made abroad, like the gas under Anshu’s pan, so the imports are taken away. In the textbooks’ shorthand, GDP = C + I + G + (X − M): consumption, investment, government spending, and exports minus imports. Nepal’s textbooks join the first and third into consumption, as the Dashboard did.

In 2025/26 consumption came to Rs. 5,959 billion and investment to Rs. 2,108 billion, which includes the goods firms added to their stocks. Exports of goods and services came to Rs. 658 billion and imports to Rs. 2,278 billion. So Nepalis spent far more than Nepal made, and the imports take the total back down, to Rs. 6,447 billion.

What was made: 6,6001,393farming821industry3,582services805taxes on goodsWhat was earned: 6,6002,512employees3,281owners and families807taxesWhat was spent: 6,447investmentconsumption 5,9592,108−1,621imports beyond exports153 not traced
Figure 13.1 Nepal’s GDP for 2025/26 counted three ways, in billions of rupees, the statistics office’s first, preliminary estimates. What was made: what each kind of industry added, then taxes on goods less subsidies. What was earned: pay to employees, what owners kept (including the mixed income of family shops and farms), and taxes on production less subsidies. What was spent: consumption and investment come to more than Nepal made, and imports beyond exports, in red, take the total back. Taxes are purple in both counts that have them. The dashed line marks the 6,600 made; the office couldn’t trace the last Rs. 153 billion of spending, the hatched sliver before it.

That is Rs. 153 billion short of the production count. In theory the gap should be nothing. In practice each method is built from different records, surveys of farms and firms for one, tax and customs records and household surveys for another, and they never quite meet. Nepal’s statistics office prints the gap honestly as a line of its own, the statistical discrepancy, and treats the production count as the GDP. Many countries’ accounts carry such a line.

From what is made to what a family can spend

GDP counts what is made inside the borders, whoever makes it. A second number counts by whose it is. Gross national product, कुल राष्ट्रिय उत्पादन, or GNP, starts from GDP, adds what the people and firms who live in the country earn abroad from their work and their money, and takes away what those who live elsewhere earn in it. The difference is called the net factor income from abroad, साधनको खुद वैदेशिक आय. Most statistics offices now call the result gross national income, GNI, which is the same thing measured as income. Nepal’s came to Rs. 6,697 billion in 2025/26, only about 1.5% more than its GDP.

That small gap is a surprise in a country where workers abroad send home about a third as much as the whole economy makes. Where did Aman’s money go?

The answer is a rule about who belongs where. National accounts everywhere follow the same rule book, kept by the United Nations, and it says a person belongs to the economy where they have lived, or mean to live, for a year or more. Aman, Anshu’s brother, signed a two-year contract in Qatar, so while he works there the accounts treat him as part of Qatar’s economy. His wage is Qatar’s income, and the money he sends home arrives in Nepal as a transfer, like a gift from one household to another. A worker who goes to India for a few months of the harvest and comes back is still part of Nepal’s economy, and his pay counts in Nepal’s GNI.

Transfers from abroad come in one step further down. Add them to GNI and you get gross national disposable income, what the whole country has to spend or save, and Nepal’s came to Rs. 8,914 billion, a third more than its GDP. That is also where the high national saving on the Dashboard comes from. So a remittance is in neither GDP nor GNP, though it is always quoted as a share of GDP. The share is only a ruler, a way to say how big the money is beside something everyone knows.

The rest of the way down is a series of subtractions. Machines, trucks and buildings wear out as they are used, and some of each year’s investment only replaces them. Take that wear, called depreciation, ह्रासकट्टी, out of GNP and you get net national product, खुद राष्ट्रिय उत्पादन, or NNP: what a country could use up without becoming poorer. Take away the taxes on goods, which are in the prices but are nobody’s pay, and add back the subsidies, and you get national income, राष्ट्रिय आय. That is the sum of the four payments: all the rent, wages, interest and profit earned by the country’s land, labor, capital and enterprise.

Not all of that reaches people’s hands. A company keeps part of its profit to grow, and pays tax on it, and part of a salary goes into social security. Meanwhile some people receive money they didn’t earn that year, such as an old-age allowance from the government. Take away the first and add the second, and you get personal income, व्यक्तिगत आय. Take away the income tax people pay out of it, and what is left is disposable income, व्यय योग्य आय, the money families can actually choose to spend or save.

Step How you get it from the step above Nepal, 2025/26
GDP everything made inside the country Rs. 6,600 billion
GNP or GNI add income earned abroad, take away income paid to foreigners Rs. 6,697 billion
NNP take away the wear on machines and buildings —
National income take away taxes on goods, add subsidies —
Personal income take away profits firms keep, company tax and social security; add transfers such as allowances —
Disposable income take away income tax —

The tables the statistics office released with these estimates stop at GNI and the country’s disposable income. They have no estimate of the wear on Nepal’s machines and buildings, so the four steps below have no Nepali figures.

Divide any of these by the number of people and you get a figure per head, प्रतिव्यक्ति आय. Nepal’s GDP per head in 2025/26 was about Rs. 218,000, or 1,513 US dollars, and its GNI per head about Rs. 221,000. Its disposable income per head, with the remittances in, was about Rs. 294,000. None of these is what any real person earns. It is an average, and it says nothing about who gets how much, which a later chapter takes up.

Money GDP and real GDP

Suppose that next year Anshu sells the same 280 glasses on her best day, but at Rs. 27.50, because milk and sugar have gone up. She takes in Rs. 7,700, 10% more. Nobody drank any more tea. Her takings grew only because the price did.

A country’s GDP has the same problem. Counted at each year’s own prices, it is money GDP, or nominal GDP, मौद्रिक कुल गार्हस्थ उत्पादन, and it grows whenever prices rise. To see whether more was really made, the statistics office counts every year’s goods and services again at the prices of one fixed year, called the base year. That gives real GDP, वास्तविक कुल गार्हस्थ उत्पादन.

Nepal’s base year is 2010/11. In that year its GDP came to Rs. 1,559 billion. By 2025/26 it had reached Rs. 6,600 billion, more than four times as much. But at 2010/11 prices, the same year’s goods and services came to Rs. 2,901 billion, less than twice as much. In fifteen years Nepal made about 86% more, and most of the rest of the rise was prices.

The number that links the two is the GDP deflator, कुल गार्हस्थ उत्पादन डिफ्लेटर: money GDP divided by real GDP, times 100. Worked out that way for 2025/26, it comes to about 228. The office prints its own, 225, worked out before the taxes on goods are added. Either way, on average, what Nepal makes cost about 2.3 times what it did in the base year. Turned round, real GDP is money GDP divided by the deflator, times 100. The same sums work from one year to the next. In 2025/26 money GDP grew by about 6.5% and real GDP by 3.85%, so prices rose about 2.5% over the year. The 3.85% is the growth everyone quotes. The deflator follows the prices of everything made in Nepal, machines and buildings included. That is why it differs from the price index of what families buy, which takes in the imports they buy and leaves out the machines; that index is the subject of a later chapter.

The difference between the two kinds of GDP was a problem from the very first count. In January 1934 the United States Department of Commerce sent the Senate a report called National Income, 1929–32, asked for during the Great Depression, and an economist named Simon Kuznets had been put in full charge of it. It found that the income produced in the country had fallen by more than half in three years. But prices had fallen too, so the real fall was smaller. Kuznets had no good price index for everything, and could only guess that, at steady prices, it was 30 to 40%.

Why counting it in Nepal is hard

Every country follows the same rule book, but the rules ask more of some countries than others.

Take food grown and eaten at home. Under the rules it counts. A family’s rice from its own field is as much a product as rice sold at Kalimati, and the family could have sold it, so it is in GDP at what it would have fetched. The same goes for a house built by its owners. The difficulty is that nobody writes it down. No receipt is made when a family eats its own rice, so the office must estimate how much was grown from farm surveys and value it at market prices. In a country where a quarter of what is made comes from farms, much of it eaten where it was grown, that is a large estimate.

The same goes for houses people live in themselves. A family renting a house pays for a service, and that is counted. A family living in a house it owns gets the same service and pays nobody, so the office estimates the rent the family would have paid. That imagined rent goes into real estate, which Nepal’s accounts put at eight rupees in every hundred.

Then there is work no one records. In the labor force survey of 2017/18, about 85% of the people with jobs were in informal work: working for themselves or their families in businesses that aren’t registered, or employed with no paid leave and nothing paid towards their social security. A carpenter paid in cash, a woman selling vegetables by the road, a porter, a tailor working from a room at home. All of it is in GDP in principle, and all of it has to be estimated from surveys of a sample of people and businesses, then scaled up.

So Nepal’s GDP is revised, sometimes a lot. The first figure for a year is made before the year has ended, as the Dashboard’s was, and is corrected as better records come in. Every few years the office also starts again from a new base year, with fresh surveys of what the economy now looks like. When Nepal moved its base from 2000/01 to 2010/11, in March 2021, its GDP was revised up by 14.1%. Nepal hadn’t grown overnight. Shops, banks and landlords were simply counted better. Bigger revisions happen too. When Nigeria moved its base year from 1990 to 2010, in April 2014, its GDP rose by 89%, mostly from services the old count had largely missed, phone companies among them, and it became Africa’s largest economy in a day.

What it leaves out

One kind of work is left out of GDP on purpose. The rule book counts goods a household makes for itself, but not services: the cooking, cleaning and washing done at home, the care of children and the old. It leaves them out, it says, because counting them would swamp the money flows the accounts are meant to follow. So when a family pays someone to cook, the cook’s wage is in GDP. When someone in the family does the same cooking for nothing, it isn’t.

Kuznets knew this. His report of 1934 said the work of housewives and others at home was large enough to change any estimate, and left it out because nobody could value it reliably. Fourteen years before him the English economist Arthur Pigou had made the same point with a joke: “if a man marries his housekeeper or his cook, the national dividend is diminished.” She does the same work, now unpaid, and the count goes down. Pigou called it a paradox. In a country like Nepal, where women do most of the unpaid work in homes and on family farms, it is a large one.

GDP misses other things too. A year of longer hours and a year of more rest can show the same output. A factory that pours smoke over a town adds to GDP, and the cleaning up afterwards adds more. Rebuilding after a flood or an earthquake shows up as output, though the country is only getting back what it had. And GDP says nothing about who gets the income. A country where a few people take most of it can have the same GDP per head as one where it is shared out evenly.

Kuznets warned about all this in his first report. Measuring income, he said, couldn’t show the unpleasantness of the work that went into earning it, or who received it, and so “the welfare of a nation can, therefore, scarcely be inferred from a measurement of national income”. People did it anyway, and still do.

That doesn’t make GDP useless. It does one job well: it adds up, in one consistent way, what a country makes and what that earns. Everything spent is someone’s income, so the total can be counted by what is made, by what is earned or by what is spent, and the three should meet. GDP counts what is made inside the borders; GNP counts what the people and firms who live in the country earn, at home or abroad; and the steps down from it, through the wear on machines, taxes and transfers, end at the money families can actually spend. Money GDP rises with prices, so real GDP, counted at a base year’s prices, shows what was really made, and the deflator is the gap between them. In Nepal much of it has to be estimated, from home-grown food to informal work, and the work done at home for nothing isn’t counted at all. The remittances that keep so many Nepali families going sit outside both GDP and GNP, as transfers. Next comes the thing all these sums are counted in, and why anyone accepts it: money.

Found a mistake? Please tell us.