Consumption feeds today. Investment feeds tomorrow.
Some spending is gone by morning; other spending keeps producing for decades. The national accounts call the second kind GFCF, and its slumps foretell the slow years.
The charts on this page draw the national accounts at 2011-12 prices, the series that ran to July to September 2025. The NSO now publishes GDP at 2022-23 prices. The idea on this page works the same on either ruler; the latest figures are in the GDP guide.
The word, defined
First principlesSome spending is gone by morning: the meal, the movie ticket, the monthly groceries. Other spending stays: the machine, the factory shed, the highway, the house. The national accounts keep the two kinds apart, and the second has a name of its own, gross fixed capital formation. In plain terms, investment in things that last.
The reason to watch it is simple. Machines installed in one year are the output of the next several; a country that eats its whole harvest has no seed for the following one. When investment slumps, the slow years arrive later. That makes GFCF a forecast written in the present tense, and the series below shows it working.
The widest swing in the accounts
y/y %, quarterlyNo major series in the accounts swings as wide as this one. Investment is the easiest spending to postpone: a firm can run an old machine another year, but a household cannot skip dinner. That asymmetry is the whole chart.
Investment fell two-fifths in one quarter: then warned, and recovered
Above the zero rule, the economy built more than a year earlier; below it, less. The plunge and the spike are one event read twice: building nearly stopped in the lockdown, so the next year, measured against that wreckage, printed –. Treat both as arithmetic before treating either as news.
What deserves attention is the quiet part of the line: the stretches below zero with no pandemic anywhere near them. Those are the slumps that foretell slow years.
Investment warned before the pandemic did. Through the second half of FY20, GFCF printed three straight quarters below zero: –. By the last quarter before COVID-19, GDP growth had slid to –, the weakest of the pre-pandemic years.
The slump in building came first; the slow economy followed. That sequence is what this series exists to reveal.
What the recovery looks like
After the base effectsThe recovery since the pandemic is genuine rather than arithmetical. The last four quarters read –: positive on bases that no longer flatter, and close to the pre-pandemic average of –. The economy is adding capacity at a steady pace, roughly a third of everything it produces.
Two cautions belong with any reading of this series. The accounts count the building, not the borrowing behind it, so a credit-fuelled boom looks identical to an honest one until the loans sour. And GFCF measures how much is built, never whether it was worth building. The number is a quantity, not a judgement.
How to misread this concept
- i.
“Investment means the stock market.”
No. In the accounts, investment means new physical capacity: machines, buildings, roads. Buying shares transfers ownership of assets already built and adds nothing to what the economy can produce. The two move on different logic, and confusing them confuses everything downstream.
- ii.
“All spending is equal for growth.”
A rupee on a machine and a rupee on a meal both count in GDP today. Only the machine keeps producing tomorrow. Two economies with identical growth can face very different futures: one built, the other consumed. Read the composition, not only the total.
- iii.
“A construction boom is always healthy.”
Not when it is built on debt nobody can repay. India's great investment surge of the late 2000s ended in a decade of bad loans and stalled projects. The accounts record the cranes going up; they say nothing about who pays for them. Quantity first, judgement after.