What is GDP, really?

In 2025-26, India's GDP was ₹357.1 lakh crore (MoSPI). That is the total market value of everything produced within the country's borders in a year. It includes the shirt stitched in a Tiruppur factory, the haircut in a Madurai saloon, the flat you own and rent to yourself (yes, that counts), the wheat bought by the government, and the software written in Bengaluru and exported to Frankfurt.

Three words carry the whole definition. Final goods and services, not the intermediate ones that go into making them. Domestic, meaning production inside India, regardless of who owns the factory. And a flow per year: GDP is not the stock of wealth piled up over decades. It is a snapshot of economic activity in one year, measured in rupees at current prices (that is nominal GDP) or adjusted for inflation (real GDP). In 2025-26 prices, the real GDP was ₹201.9 lakh crore, meaning that, compared to a 2011-12 base, output after stripping out price changes had grown by 7.4% over the previous year.

GDP is not an ancient institution. National-income accounting is a twentieth-century invention, built in the 1930s and 1940s to manage the Great Depression and finance the Second World War. The American economist Simon Kuznets developed the early US estimates and delivered them to Congress in 1934. Before that, governments did not know, in any systematic way, how much their economies produced.

Kuznets himself warned about putting too much weight on the number. "The welfare of a nation," he wrote, "can scarcely be inferred from a measure of national income." His warning still holds: GDP was designed to measure production for war-planning, not to score happiness.

The urge to measure India's economy did not start with the government. In the 1860s and 1870s, Dadabhai Naoroji, a nationalist leader, made pioneering private estimates of India's per-capita income. His purpose was political: he argued that British rule was draining wealth. Naoroji's "drain theory" used rough national-income arithmetic to show that wealth was being siphoned off.

The official machinery arrived after Independence. In 1949, the government set up the National Income Committee, chaired by P. C. Mahalanobis, with V. K. R. V. Rao and D. R. Gadgil as members. Their report in the early 1950s gave India its first official national-income estimates and created the framework for what is today the National Accounts Statistics, compiled by the National Statistical Office under MoSPI. The earliest figure in that series, for 1950-51, put India's nominal GDP at ₹10,221.6 crore. That is a sliver compared to today's number, a reminder of how much the production base has expanded.

Think of three different ways to measure the same economic activity. You can add up the value added by every producer (the output approach), sum up everyone's incomes (the income approach), or total all spending on final goods and services (the expenditure approach). In principle, all three should give the same number. India uses all three, but the production approach is the backbone.

The production approach starts with Gross Value Added (GVA). For each sector, officials take the output value and subtract the inputs used up in production. For example, when a steel plant sells ₹10 crore of steel but bought ₹3 crore of iron ore and coal, its GVA is ₹7 crore. Summing GVA across all sectors and adding net product taxes (like GST minus subsidies) gives GDP. In 2025-26, GVA at basic prices was ₹323.5 lakh crore, and net taxes on products was ₹33.7 lakh crore, yielding the GDP figure.

The hard part is that a large share of India's activity is informal and unregistered. The street vendor selling pakodas, the carpenter who works without a written contract, the farm where family labour is not paid a wage. For these, the national accounts cannot simply tally invoices. They estimate rather than measure directly. The methodology often uses formal-sector or corporate data as a proxy, for instance, the MCA21 company database, and labour-input indicators to extrapolate. This is not unique to India, but the size of the informal sector makes the exercise particularly difficult and, as we will see, contentious.

The headline number tells you the sheer scale of production and its growth. India's nominal GDP has climbed from ₹10,221.6 crore in 1950-51 to ₹357.1 lakh crore in 2025-26. In real terms, GDP rose from ₹5 lakh crore in 1950-51 to ₹201.9 lakh crore in 2025-26, and per-capita real output moved from about ₹17,000 to ₹1.4 lakh over the same period.

Who is buying all this output? In 2025-26, private consumption accounted for 61.5% of GDP, while investment (gross fixed capital formation) made up 30%, government consumption 9.9%, and net exports were a small negative (-2.3%). That heavy reliance on household spending also makes growth sensitive to what happens in kitchens and kiranas across the country.

The composition of output tells its own story. In 1950-51, agriculture accounted for 53.2% of GVA; by 2025-26, that share had shrunk to 16.8%, while services rose to 56.4%. In rupee terms, agriculture contributed ₹54.3 lakh crore, industry ₹86.8 lakh crore (of which manufacturing was ₹45.5 lakh crore, 14.1% of GVA), and services ₹182.4 lakh crore. That shift mirrors the story of an economy moving from fields to cities.

When you see a dollar figure, like the World Bank's estimate of India's GDP at $3.91 trillion in 2024 (market exchange rates), it is a convenient rank card. By this measure, India's output was just behind Germany ($4.69 trillion) and Japan ($4.03 trillion), and ahead of the United Kingdom ($3.69 trillion). But a dollar figure at market exchange rates can be misleading because it does not reflect how much a rupee actually buys in India. A haircut in Ballia costs a fraction of one in New York. That is why economists also look at purchasing-power parity (PPP). By PPP, India's GDP was about $16.2 trillion in 2024, making it one of the largest economies and far larger than the market-rate figure suggests.

India's Gross National Income (GNI), which adds income earned abroad by Indians and subtracts income sent home by foreigners here, was ₹351.6 lakh crore in 2025-26, about ₹5.5 lakh crore less than GDP. That gap, net primary income paid abroad, is largely profits and interest flowing out. It is a reminder that GDP measures production inside the border, not the income accruing to residents.

GDP is a production measure, and what it excludes is as important as what it includes. It does not tell you who got the money. A rising GDP can coexist with rising inequality. It does not count unpaid work. Where the statistician draws the "production boundary" is a convention, not a law of nature. A meal cooked and sold in a restaurant counts; the same meal cooked at home does not. Feminist economist Marilyn Waring showed in her 1988 book If Women Counted that this boundary leaves most of women's unpaid care and domestic work invisible. In India, where women spend an outsized share of their day on unrecognised domestic labour, the GDP silence is loud.

GDP also ignores the depreciation of natural capital. When a mine is exhausted or a river polluted, the costs do not subtract from GDP. They are externalities, not outputs. Even the physical capital that gets used up is accounted for only when you move to net domestic product (NDP). GDP minus depreciation gave an NDP of ₹317.5 lakh crore in 2025-26. The gap, roughly ₹39.7 lakh crore, is the value of machinery, buildings, and roads that wore out during the year.

The treatment of the informal sector compounds the blind spot. Because the accounts estimate informality via formal-sector proxies, shifts that hit the informal economy hard can get smeared. If small businesses close after a shock like demonetisation or GST implementation but large firms keep growing, the numbers may overstate overall growth. That takes us to the live argument over India's measurement.

A heated, public debate questions whether India really grew as much as the official figures claim. The argument pivots on two problems: formal-sector proxy and deflators.

In a March 2026 working paper from the Peterson Institute for International Economics, economists Abhishek Anand, Josh Felman, and Arvind Subramanian argued that India's growth was likely underestimated by about 1 to 1.5 percentage points a year during the 2005-2011 boom and overestimated by about 1.5 to 2 percentage points a year from 2012 to 2023. The reason: after 2015, events like demonetisation, the introduction of GST, and the COVID-19 pandemic hit the informal sector far harder than the formal sector. National accounts, which use formal-sector data (like MCA21 filings) to infer informal activity, missed the differential collapse. In effect, the methodology assumed that the informal sector moved in lockstep with the formal, when in reality it was bleeding.

A second weakness, they argue, lies in sectoral deflators. To compute real growth, statisticians strip out price changes using price indices. In manufacturing, many of these deflators rely on commodity-price indexes, which fell sharply in the 2014-2016 period. A falling deflator can mathematically inflate real output, producing a faster growth print than what happened on the ground. An earlier 2019 paper by Subramanian alone had put the overestimation at roughly 2.5 percentage points per year for 2011-2017.

The government has pushed back. After the 2019 paper, the Prime Minister's Economic Advisory Council issued a point-by-point rebuttal. It called the cross-country-regression method "unusual" and its conclusions unsupported, noting that the analysis relied on private data and underplayed agriculture and services. Officials have also pointed out that India revised its GDP methodology again in February 2026, partly to address measurement concerns of exactly this kind. The debate is unsettled. Both sides agree the number could be wrong; they disagree on how much and in which direction.

GDP is not useless. It is the best guess we have, built on painstaking work by the National Statistical Office. But it is an estimate, not a census. The assumptions that make it possible also make it fragile, especially in an economy where most work is informal. Treat the growth rate as a pointer, not a thermometer: it tells you the direction of travel, but the speedometer may be sticky.