What does India's lower-middle-income label really measure?

What number puts India in the lower-middle-income box?

Start with the box itself. For its fiscal year 2027 classification, the World Bank sorted economies using their estimated 2025 Atlas GNIGNI (gross national income)GNI starts with everything produced inside the country, then adjusts for net primary income across the border. That includes wages and investment income residents receive from abroad, minus comparable income paid to non-residents. Personal remittances are generally transfers, so they are not the adjustment that turns GDP into GNI.The World Bank divides Atlas-method GNI by population to set the analytical income classification. It is a national-accounts average, not a household pay packet. per person. Low income ended at $1,175. Lower middle income ran from $1,176 to $4,635. Upper middle income began at $4,636, and high income began above $14,375. India's estimate was $2,760, so it landed squarely in the second band. That is the label's formal claim: India's average national income per resident, converted by the Bank's method, falls inside that dollar interval.

India's $2,760 was 59.5% of the $4,636 upper-middle entry line. That does not mean India is 59.5% developed, or that the journey is 59.5% complete. It is simply one number divided by the boundary above it. If a future estimate rises above that year's boundary, the Bank will move India into its upper-middle-income tables. It will not reclassify each Indian household.

Both the income figure and the boundary need a date attached. The 2025 income is an estimate published in July 2026 and may be revised. The boundary is updated every year for international inflation. So India is chasing a moving line, not filling a fixed $4,636 bucket.

How does the Bank get India's number? It starts with GNI, not GDPGDP (gross domestic product)GDP is the market value of production inside a country's borders. A foreign-owned factory in India adds to Indian GDP because the production happens here; profits later paid abroad are part of the cross-border income adjustment between GDP and GNI.India can rank among the world's largest economies by total GDP and remain lower middle income after national income is divided across its very large population.. GDP counts production inside India; GNI adds net wages and investment income received from abroad and subtracts comparable income paid abroad. The national total is then divided by about 1.46 billion residents. This is an accounting average, not the median salary, household income or what a typical person can spend. For the wider distinction, see How big is India's economy?.

The final step converts rupees into dollars with the Atlas method. The Bank uses the current exchange rate and the previous two years, adjusted for the gap between Indian and international inflation. That reduces the effect of one bad currency year, but it does not remove currency effects. It is also not PPPPPP (purchasing power parity)PPP compares price levels by asking how much a common basket of goods and services costs in each economy. An international dollar is a statistical unit with the purchasing power of a US dollar in the United States, not cash that can be spent abroad.India's PPP GNI per person is much higher than its Atlas figure because many local goods and services are cheaper. PPP answers a purchasing-power question, not the classification question.: the calculation does not ask what $2,760 can buy in a kirana, hospital or rented home in India.

India has moved closer to the upper-middle line, but not every year. It was classified as low income from 1987 through 2006. When it entered lower middle income in 2007, its Atlas income was only 24.6% of the next line. The share reached 51.2% in 2019, fell to 46.4% during the pandemic and recovered to 59.5% by 2025. India's income can rise while its relative position falls if population growth, currency conversion or the threshold moves against it.

Why does a giant economy rank so low per person?

India looks enormous before the total is divided by its population. The same release puts GDP at about $3.96 trillion using market exchange rates. After adjusting for local prices through PPP, it is about $17.2 trillion. Atlas GNI is about $4.04 trillion. These totals measure the whole economy. The per-person figures spread national income across about 1.46 billion residents.

That division changes India's rank dramatically. India is 3rd among 197 published PPP GDP estimates and 6th among 204 market-GDP estimates. But its PPP GNI per person is $11,600, ranked 132nd among 196 estimates. Its Atlas GNI per person is $2,760, ranked 156th among 200. The number of economies changes because the four workbooks do not all have estimates for the same places.

PPP and Atlas also answer different questions. PPP adjusts for local prices: how much would the same basket of goods and services cost in each country? Atlas converts income with a three-year exchange-rate average for the classification. India's PPP figure is about 4.2 times its Atlas figure. The multiple is about 4.3 in Pakistan, 3.8 in Bangladesh, 2.0 in China and 1.0 in the United States. India has a much larger PPP figure because many locally produced services cost less than they do in richer countries.

The larger PPP figure is useful, but it is not a pile of dollars India can spend abroad. It cannot pay for imported oil or repay debt denominated in dollars. Bangladesh makes the difference clear. Its Atlas GNI per person is $2,840 against India's $2,760, while India leads at PPP, $11,600 to $10,710. The ranking flips because one measure converts through exchange rates and the other adjusts for local prices. For the gap between a national average and people's actual incomes, see How rich is the average Indian?.

Who belongs to the lower-middle-income world?

The biggest global change since 1987 is that far fewer people now live in countries labelled low income. Among economies with both classification and population data, the share fell from 61.2% in 1987 to 9.4% in 2025. The lower-middle share rose from 13.0% to 36.0%, and the upper-middle share from 8.6% to 37.6%. The high-income share barely changed, staying near 17%. Most of the movement was into the two middle groups.

The bars classify countries, not families. Everyone living in a country is placed in the same bar. When China changes category, hundreds of millions of people move between bars on the update date. That does not mean every Chinese household crossed an income line that year.

India dominates its present group. Of roughly 2.95 billion people living in lower-middle-income economies in 2025, 49.7% live in India. Pakistan accounts for 8.7%, Nigeria 8.1%, Bangladesh 6.0% and Egypt 4.0%. The other 42 economies together make up 23.6%. There are 47 countries in the group, but count people rather than countries and the group is half India.

That makes an eventual Indian upgrade unusually large on paper. Nearly 1.5 billion people would move from the lower-middle bar to the upper-middle bar in one annual update. The global chart would change dramatically. The lives of those people would not change on the update date.

What does history say from India's position?

A shared label does not produce a shared journey. China crossed the upper-middle line in 2010 and reached more than three times the entry threshold by 2025. Indonesia crossed in 2019, fell back in 2020 and 2021, then returned in 2022. Sri Lanka first crossed in 2018, dropped back the next year and returned only in 2025.

Vietnam reached upper middle income for the first time in 2025. Bangladesh remained lower middle income, but at 61.3% of the next line it sat slightly ahead of India's 59.5%. The Philippines also crossed in 2025, although it is not one of the six lines plotted here.

The full record is slower than the familiar Asian success stories suggest. We found 67 economies that entered the historical data after 1987 as lower middle income. Each was then between 55% and 65% of the next line, the band around India's current position. About 13.3% crossed within five years. The estimated share reached 29.2% within ten years, 48.0% within fifteen and 64.3% within twenty.

Twenty-six of those economies had still not crossed by 2025. We did not delete them or count them as permanent failures. Each contributes all the years for which it can be observed, and its clock simply stops in 2025. Statisticians call this a Kaplan-Meier estimate with right-censoring. We also exclude economies already in the starting band in 1987 because the workbook cannot tell us when their journeys began.

It is not India's probability. The 67 economies include commodity exporters, factory economies, islands and countries hit by wars or financial crises. The calculation makes one narrower point: being three-fifths of the way to the line does not make a crossing within ten years normal or automatic.

Crossing often stuck, but not always. We found 76 economies that moved from low or lower middle income into upper middle or high income during the workbook's 1987 to 2025 window. Of those, 54 stayed at least upper middle for the rest of the observed period. Sixteen fell back and later recovered. Six were still below the line in 2025.

Put differently, 22 of 76 observed first crossers, nearly 29%, fell back at least once. Indonesia and Sri Lanka are part of a wider pattern. Currency moves, crises, commodity cycles and statistical revisions can all contribute. Because the workbook begins in 1987, the count covers only reversals visible from that year onward.

What moved India's own number?

The bridge below replays India's 2015 to 2025 change one piece at a time. India began at 39.1% of the upper-middle line. First, update only GNI measured in rupees and leave population, currency conversion and the threshold at their 2015 settings. The running figure rises to 98.1%.

Now update the other pieces. Population growth spreads the income across more people, lowering the running figure to 89.0%. Updating the Atlas currency-and-inflation conversion lowers it to 68.4%. Finally, updating the upper-middle threshold brings the result to India's observed 59.5%.

The middle bars are bookkeeping steps, not alternative histories. Only the first and last bars are observed. Changing the order of the steps would change the middle values but not the final 59.5%. The safe conclusion is that rupee GNI growth pushed India upward, while population growth, Atlas conversion and a rising threshold absorbed much of that gain.

The six economies promoted in 2025 show why this accounting matters. Vietnam's Atlas GNI per person rose 10.7%, from $4,490 to $4,970, alongside strong growth and exports. The Philippines rose 8.5%, from $4,470 to $4,850, with broad-based growth. Sri Lanka's 20.7% jump, from $3,870 to $4,670, reflected recovery from crisis.

The less familiar cases show how better measurement can change a category. Jordan updated the base year and coverage of its national accounts. The new accounts captured activity the older system had missed and made the measured economy nearly 10% larger. Its Atlas GNI per person rose 7.8%. Togo's new population estimate was about 11.7% lower. Dividing national income among fewer estimated people helped lift GNI per person by 10.7%, enough to change category. Micronesia moved after a more modest post-Covid recovery.

Jordan did not build those newly measured businesses overnight, and Togo's households did not suddenly become richer when its population estimate changed. The revisions corrected older statistics. This is why a classification can move because the economy changed, because its measurement improved, or because both happened together. The bars show one-year changes; they do not measure how much each possible cause contributed.

What would crossing change, and when might it happen?

The World Bank uses one ladder to describe income and another to make lending decisions. They overlap, but crossing one line does not automatically cross the other. In the 2025 data, the ceiling used for IDA allocations was $1,365. IDA provides grants and very low-cost loans to poorer countries, and India is listed as having graduated from it in fiscal year 2014. India was at $2,760, while upper middle income began at $4,636. A discussion about graduating from regular IBRD lending began much higher, at $8,105. High income began above $14,375.

An upgrade would first move India into a different column in World Bank tables. It could also matter for policies or programmes that refer to income groups. It would not automatically end IBRD borrowing, change Indian taxes or raise salaries. Lending decisions also depend on whether a country can repay and on wider institutional conditions. Crossing $4,636 would record higher average national income. It would not flip every policy switch.

The possible crossing date depends heavily on which part of India's past we extend. From 2007 to 2025, India's position relative to the line improved by about 5.0% a year. Continue that pace and the arithmetic reaches the line around 2036. Starting in 2015 gives about 2037, starting in 2021 gives about 2038, and using the slower period since 2019 gives about 2046.

These calculations compound growth in the ratio; they do not add the same number of percentage points each year. India's 59.5% position must grow to 100%, while the dollar threshold continues to move.

The 2036 to 2046 range is not a forecast. It shows how much the answer changes when we choose a different starting year. Future rupee income, population, inflation and exchange rates will all change India's Atlas figure, and the threshold will be updated each year. A weaker rupee can offset part of real growth. A statistical revision can also move the date without changing anyone's life that year.

Would an upper-middle label fit the India people live in?

India's latest modelled estimate is about 27 deaths before age five for every 1,000 live births in 2024. The comparison is deliberately tied to the income milestone rather than one calendar year. For each peer, we use the year it first entered upper middle income. The rates were 15.7 in China, 21.8 in Indonesia, 17.3 in Vietnam and 26.5 in the Philippines. They were 14.1 in Thailand, 15.1 in Malaysia and 7.6 in Sri Lanka.

India is close to the Philippines at its crossing. But child mortality is much higher than it was in Sri Lanka, China or Thailand at theirs. The same income milestone can therefore arrive with very different levels of child survival. The comparison cannot tell us how much nutrition, sanitation, vaccination, maternal care or health-system reach produced those differences.

The production structure differs too. Manufacturing accounts for about 13.5% of India's GDP. At their first upper-middle classification, the share was 31.1% in China and 30.9% in Thailand. It was 25.8% in Malaysia, 24.5% in Vietnam and 19.7% in Indonesia. The Philippines, at 15.3%, was the closest selected peer, followed by Sri Lanka at 15.0%.

Most selected Asian crossers reached the line with a larger factory base than India has today. Manufacturing can employ large numbers of workers and support exports, but the comparison does not prove that factories caused the crossing. Services, productivity, education, infrastructure, trade and demographics also matter. The longer development argument is in Why India stayed poor while Asia got rich.

Household life is the final test. MoSPI groups surveyed households from the poorest 5% to the richest 5% by monthly consumption per person. The poorest rural group averaged ₹1,677 a month, while the poorest urban group averaged ₹2,376. In the group just above the middle, the figures were ₹3,866 in rural India and ₹6,334 in urban India. For the richest 5%, they rose to ₹10,137 and ₹20,310.

These figures are not an alternative estimate of GNI. The survey records household consumption and divides it by the number of people in the household. GNI comes from the national accounts and includes income that never appears as cash in a household budget. The measures cover different things and different periods, so converting one into the other would create a false comparison.

Those gaps are why the country label cannot describe a typical Indian. Urban consumption is higher at every point shown, but urban prices are also different, so the gap is not a direct measure of living standards. HCES does not reveal inequality within a household, and consumption is usually distributed more evenly than income or wealth. "Lower middle income" places India's national average on an international scale. It does not tell us where a typical household sits. See How unequal is India? and Has India ended poverty? for the distribution and threshold questions.

How should you read these numbers?

Sources. The headline values come from the World Bank's July 2026 workbooks for 2025 GNI, GNI per person, GDP, PPP GDP and population. The Bank uses those estimates for its fiscal year 2027 groups. Historical classifications and thresholds come from its history workbook. Historical income, population, child mortality and manufacturing data come from World Development Indicators. The explanations for the six newly promoted economies come from the Bank's release. Household consumption comes separately from MoSPI's HCES 2023-24.

Calculations. The progress line divides each year's Atlas GNI per person by that year's upper-middle threshold. The 67-country comparison starts when an economy is first observed after 1987 as lower middle income and between 55% and 65% of the line. Economies that have not crossed remain in the calculation for every year we can observe them. The reversal count covers 76 first crossings visible in the workbook. The India bridge updates rupee GNI, population, Atlas conversion and the threshold one after another. Its middle bars depend on that order; its first and last bars do not. The scenarios continue historical growth in India's ratio until it reaches 100%.

Limits. Every 2025 World Bank value is an estimate and may be revised. The four rank workbooks cover different numbers of economies because some estimates are missing. GNI differs from GDP through wages and investment income crossing borders; ordinary personal remittances are transfers, not part of that adjustment. Classifications are annual, while social indicators can arrive later. PPP and Atlas answer different questions, and the historical comparison mixes very different countries and periods. HCES consumption is neither income nor GNI, and rural and urban prices differ. The label is useful for comparing countries. It is not a welfare score or a development finish line.