How an RBI repo rate cut moves through banks, NBFCs, autos and real estate: who feels it first, why bank margins get squeezed before they heal, and when the demand actually arrives.
Starting in February 2025, the RBI began cutting the repo rate, the rate at which it lends to banks. It came in three steps: a quarter-point in February, to 6.25% (Business Standard), another quarter-point in April, to 6.0% (Business Standard), and a bigger half-point in June, to 5.5% (Business Standard). By June the repo had come down a full percentage point, from 6.5% to 5.5%, and the RBI also freed up bank money by cutting the share of deposits banks must park with it.
The repo rate is the price of money for the whole economy. When it falls, every loan in the country gets a nudge downward: home loans, car loans, business loans. The news covered it as one afternoon's stock move. The real story plays out over three years, at three different speeds, and the order in which it reaches each business is the whole game.
Who reprices first?
Money got cheaper. That is the entire event. But businesses do not all feel that cheaper money at the same moment. A bank's loans can reprice before its deposits. A family's EMI can fall before they decide to buy. A builder may not see the resulting demand for years.
So the useful question is never whether a rate cut helps. It is who feels it first, and who waits. Answer that for each business and the whole signal falls into place.
They borrow from banks and markets, so their cost of funds falls. If that funding cost drops faster than the yields on the loans they have already made, their spread widens for a while. Whether it actually does depends on the mix of their borrowing and how fast their own loans reprice. So the thing to watch is the funding spread, not a guaranteed win.
Cars and flats are bought with EMIs. When the EMI falls, the showroom fills over the following quarters. Nothing shows in the very next results, which is exactly when impatient investors give up.
They sit at the end of the chain. A flat booked this year gets plastered and painted two or three years from now. The demand is real and it is coming, just not on the market's clock.
Any company carrying large debt gets its interest bill trimmed as loans reprice. The weaker the balance sheet, the bigger the relief, which is why low-quality stocks often bounce hardest on rate cuts.
This surprises people. Most bank loans are linked to the repo and reprice within weeks, but deposits are locked at old rates for months or years. So the bank's income falls before its costs do, and the spread gets squeezed for two or three quarters. The headline said banks win. Their next two results said otherwise.
Fixed deposit rates follow the repo down. Anyone living on interest income, including many retirees, takes a quiet pay cut that never makes the front page.
Companies that promised customers fixed returns years ago must now earn those returns in a cheaper-money world. Falling rates make old guarantees more expensive to keep.
Ask anyone who wins from a rate cut and they will say banks. Watch what actually happens to a bank's accounts and you see the opposite, at least at first.
Many newer floating-rate retail loans in India are linked to an external benchmark such as the repo rate (Business Standard), so those loans reprice relatively quickly, within about a quarter of a cut. But the bank's own cost, the interest it pays on your fixed deposit, is locked until that deposit matures. So for two or three quarters the bank earns less on those loans while still paying old, higher rates on deposits. The spread compresses.
Then the clock catches up. Old deposits mature, new ones are booked cheaper, the cost side falls to meet the income side, and the spread heals. Meanwhile cheaper loans have pulled in more borrowers, so the bank is now earning a normal spread on a bigger book. The rate cut does help banks. It just hurts them first. If you only looked at one quarter, you would learn the wrong lesson.
If the framework is right, the banks with fast-repricing loans and slower-repricing deposits should be the first to feel a squeeze, and it should show up in one number above all: the net interest margin, the gap between what a bank earns on its loans and what it pays on its deposits.
So the metric to watch is not revenue, which is noisy, but the NIM, because the mechanism is specifically about the spread between what a bank earns on assets and what it pays for funding. That makes the NIM the cleanest possible proof.
Then the banks. We have not ranked every lender by exposure, and we are not claiming these are the most exposed in the system. We just need two that clearly fit the mechanism, a large floating-rate loan book that reprices fast, funded by deposits that reprice slowly. State Bank of India and Bank of Baroda both fit, so we go into their filings and compare each with the very same quarter a year earlier.
Take SBI first, and rather than jump to the margin, take the spread apart into the two forces that make it, both on a domestic-operations basis. They moved in opposite, unhelpful directions.
| State Bank of India: the spread, taken apart | Q1 FY25 | Q1 FY26 | Change | What it means |
|---|---|---|---|---|
| Yield on advances | 8.89% | 8.78% | -11 bps | What SBI earns on loans fell, as repo-linked loans repriced down. |
| Cost of deposits | 5.00% | 5.21% | +21 bps | What SBI pays for deposits actually kept rising, as older high-rate deposits had not yet matured. |
| Net interest margin | 3.35% | 3.02% | -33 bps | Income down, funding cost up, so the spread was squeezed from both sides at once. |
This is the mechanism, not a guess. You can watch the income side fall while the funding side refuses to, and the margin give way in the middle. That is exactly why NIM, not revenue, is the honest place to look. Source.
Both banks showed the direction the framework predicted: their margins compressed in the first quarter after the rate-cut cycle began.
Not every bank compressed equally, and that is exactly the point. ICICI Bank, where low-cost current and savings deposits fund a much larger share of the book, saw its margin barely move, from 4.36% to 4.34% over the same year (ICICI Bank). A deeper cushion of cheap, sticky deposits means less of your funding reprices against you when rates fall. So the squeeze is not about being a bank; it is about the shape of the balance sheet. Fast-repricing assets sitting on slow, cheap liabilities feel it first and worst, which is why SBI and Bank of Baroda are the sharper test cases here.
Two notes on reading this fairly. First, we compare each bank with the same quarter a year earlier, not with the quarter just before, so that a bank's ordinary seasonal swings do not get mistaken for the rate-cut effect. Second, SBI's figure is its domestic-operations margin and Bank of Baroda's is its bank-wide global margin, because that is what each bank reports cleanly; since each bank is measured only against its own year-ago quarter, the different bases do not distort the one thing that matters here, which is that both margins fell. Bank of Baroda's own management named the cause: the cost of deposits, it said, would take more time to realign while the loans had already repriced down (Business Standard).
A fair challenge, and the one that could sink us if we skipped it: how do we know the repo cut squeezed these margins, and not something else wearing the same clothes? A bank's NIM can move for at least three other reasons at once. Its loan mix can drift toward lower-yielding corporate loans. Competition for deposits can heat up on its own. And the tail of the previous rate-hiking cycle can still be biting, as expensive deposits booked at the old peak keep maturing and drag the average funding cost up. That last one is not a hypothetical here: SBI's cost of deposits actually rose over the year, which is the old cycle unwinding, not the new one.
Two things keep the attribution honest. First, the asset side is mechanical, not a correlation we are hoping into being: a large share of these loans is linked to the repo by regulation and reprices down within a quarter of a cut, which is precisely why the yield fell first and fastest. Second, and stronger, the control case above. Same economy, same quarter, same rate cut, yet a bank built on cheap, sticky deposits barely moved. If a broad macro force were doing this, ICICI would have been dragged down too. What separated the squeezed banks from the unsqueezed one was the exact balance-sheet shape the framework named, which is what a cause is supposed to look like.
So here is the claim, stated at its honest edge: the cut was the primary driver of the asset-side squeeze, not the sole author of every basis point. Some of the funding-cost stickiness belongs to the old hiking cycle, and we say so. That is why we did not stop at one bank falling. We went looking for a bank that should not have moved, and checked that it did not.
And it was not a shock to the market either. Before these results printed, Fitch had already flagged that roughly 45% of system loans reprice down almost at once, and pencilled in about a 30 basis point margin contraction for banks over the year (Business Standard). We are not claiming any bank stock fell on a given day because of this. The claim is narrower and stronger: the specific number the framework pointed at, the margin, actually deteriorated, roughly as much as the people who model this for a living expected, for the reason the framework gave.
Now the second half of the clock, in the same number. SBI's domestic margin bottomed at 3.02% in the June quarter, then ticked back up to 3.09% in the September quarter as cheaper deposits finally began to replace the old expensive ones (Business Standard). The funding side was catching the loan side, exactly as the mechanism said it would, and S&P Global was by then expecting bank margins to keep improving through the second half of the year (Business Standard).
Behind the margin, the rest of the chain is only just getting going: cheaper loans pull in more borrowers, the loan book grows, and the demand works its way out to autos, housing and eventually cement and paint. Rate cut, then margin pressure, then funding costs catch up, then margin recovery, then loan growth, then downstream demand. The clock, proven one tick at a time.
The thing the framework predicted, a first-quarter squeeze in the margin, is sitting right there in the filings. That is what Fathom is for: make the prediction, then go check whether reality agrees.
Here is the honest part. Cheaper loans only create demand if the loan was the thing holding the buyer back.
The RBI cut rates hard in 2019 too, well before the pandemic, taking the repo down through the year. The textbook chain said auto sales should recover. They did not. Passenger-vehicle and two-wheeler sales kept falling through 2019 (Business Standard), the worst year for auto sales in over two decades, because the problem was not the EMI. Incomes were weak, insurance costs had jumped, and lenders in the shadow-banking system had just been burnt and did not want to lend at any rate. Money was cheaper, but nobody confident was asking for it, and nobody nervous was offering it.
The hidden assumption is confidence. The chain only works if a willing borrower meets a willing lender, and a rate cut can lower the price of money without making anyone more willing to use it. When the RBI is cutting because the economy is weakening, the cut and the weakness arrive together, and the weakness sometimes wins. So before trusting the chain, ask why the RBI is cutting. A cut into a healthy economy is fuel. A cut into a frightened one is a bandage.
Now reverse the signal. The RBI raises the repo rate by one percentage point to fight inflation. Do not repeat the terms above. Rebuild the transmission yourself, from the cut to the last business that feels it.
Where this signal plays out in depth: the sectors it moves and the companies that lived it.
A rate cut is a cost cut for borrowers and a pay cut for lenders and savers, delivered on four different clocks. Banks hurt before they heal, borrowers act over quarters, and the builders at the end of the chain wait years. And the chain only works if a willing borrower meets a willing lender: a cut into a frightened economy can vanish without creating a single new sale.
When money gets cheaper, ask who reprices first.