Two investors look at the same commercial bank. One reads the profit figure at the top of the quarterly report, sees a big number, and buys. The other scrolls to two smaller numbers the bank is forced to publish every month, the base rate and the spread, and quietly decides the dividend is about to shrink. A year later the second investor looks smarter. This piece is about why.
If you own a bank share on NEPSE, the bank base rate and spread in Nepal are not technical trivia for loan officers. They are the two dials that decide how much a bank earns on every rupee it lends, and one of those dials is held down by Nepal Rastra Bank. Understand the base rate and the spread and you understand the ceiling on Nepali bank profits better than any single quarter of earnings will tell you. Here is the position this article takes before the numbers arrive. The regulated spread is a hidden cap on how much banks can make from their core business, that cap has been biting harder as base rates fall, and the fat bank profits reported for 2025/26 were not built on lending at all. They were borrowed from somewhere else, and that source runs out.
What a bank actually sells
Strip a commercial bank down to its engine and it does one simple thing. It takes your deposit, pays you a little, lends that money to someone else, and charges them more. The gap between what it pays depositors and what it charges borrowers is where almost all of a traditional bank’s money comes from. Everything else, the card fees, the remittance commission, the trade finance charges, sits on top of that spread as a garnish.
So the question “how does a Nepali bank make money” has a boring, central answer. It makes money on the difference between two interest rates. That is exactly why the two numbers the regulator forces every bank to disclose, the base rate and the spread rate, are the ones that matter. They describe the engine, not the paint.
The base rate: the floor under every loan
The base rate is the lowest rate at which a bank is allowed to lend. Think of it as the bank’s own cost of doing business expressed as a percentage, the break-even line below which lending loses money. NRB requires every bank to calculate it by a set formula and publish it monthly, so nobody can hide it.
The formula stacks up a few costs. Start with the cost of funds, which is mostly the interest the bank pays on deposits. Add the cost of the money NRB makes the bank park and not lend: the cash reserve ratio held at the central bank and the statutory liquidity kept in government securities, both of which earn little or nothing, so they push the effective cost of lendable money up. Add the bank’s operating cost, staff, branches, technology, spread across its loan book. Then add a small, capped return on assets, the sliver of profit NRB permits inside the base rate itself. Sum those and you get the base rate.
A real example makes it concrete. Rastriya Banijya Bank, the large state-owned commercial bank, published an average base rate of 4.13 percent for the three months ending in mid-August 2026 (Bhadra 2083), built on a cost of fund of just 2.88 percent, according to the bank’s own base rate disclosure. Read that again. The raw money cost the bank under 3 percent, and by the time reserve costs, operating expenses and the permitted return were added, its lending floor sat near 4.1 percent. Any loan RBB writes has to price above that floor to make sense.
The base rate is not fixed. It moves with the cost of funds, which moves with deposit rates, which move with liquidity in the system. When banks are flush with deposits and short of borrowers, they cut deposit rates, the cost of funds falls, and the base rate falls with it. That is precisely what has happened. The average base rate of commercial banks slipped below 5 percent, to around 4.97 percent, in the final month of fiscal year 2025/26, likely the lowest in Nepal’s banking history, according to figures reported by myRepublica. Individual banks ranged widely: Standard Chartered Bank Nepal near the bottom at about 4.21 percent, NIC Asia at the top around 6.04 percent, per the same reporting. A high base rate is not a badge of strength. It usually means expensive deposits or a bloated cost structure, which forces the bank to charge borrowers more just to break even, which makes its loans less competitive.
The spread: where NRB puts its thumb on the scale
Now the second dial, and the one that turns this from an explainer into an argument.
The spread rate is the gap between the average interest a bank earns on its loans and the average it pays on its deposits, calculated by an NRB formula. If a bank earns 8 percent across its loan book and pays 4 percent across its deposits, its spread is roughly 4 percentage points. That spread is the bank’s gross margin on money. The wider it is, the more the bank keeps.
Left alone, a bank would happily widen that spread forever, squeezing depositors and borrowers at both ends. NRB does not leave it alone. The central bank caps the spread. For commercial banks the ceiling was tightened in stages, cut from 4.4 percent down to 4.0 percent, a reduction NRB pushed through in the review of monetary policy earlier this decade, as reported by The Himalayan Times and others. Four percent is the wall. A commercial bank may run its spread up to 4 percent and no further.
This is the single most important fact for a bank shareholder to hold in their head, and it is the one ShareSansar and Merolagani will print as a line item without drawing the conclusion. The spread cap is a hard ceiling on how profitable the core lending business can be. A Nepali commercial bank cannot decide to earn a 6 percent margin on money the way a bank in a lightly regulated market might. Its gross margin is legally boxed. However clever the management, however strong the brand, the main engine is throttled at 4 percent by rule. That is the hidden ceiling in the headline of this piece. To see how the spread sits alongside the other numbers that define a bank’s health, our explainer on how to read a bank’s NPL, CD and CAR ratios walks through each one.
Why the cap is biting harder now
A ceiling only hurts when you are pushing against it. For a while, that was not the problem. The problem now is the opposite, and it is arguably worse for earnings.
Look at what the system’s spread has actually done. By mid-June 2026 the weighted average deposit rate had fallen to about 3.29 percent and the weighted average lending rate to about 6.64 percent, leaving an arithmetic system spread of roughly 3.35 percentage points, well under the 4 percent cap. RBB’s own published spread was 3.41 percent in mid-August 2026, down from 3.70 percent a year earlier and far below where it sat in 2023, according to the bank’s disclosure. Banks are not being squeezed by the cap right now. They are running below it, and the gap is shrinking on its own.
Why does that hurt? Because the same flood of deposits that dragged the base rate below 5 percent has nowhere to go. Banks are sitting on an estimated NPR 1.3 kharba (about NPR 1.3 trillion) of idle, lendable money, held back by weak credit demand rather than by any rule, with the overall credit-to-deposit ratio falling to around 72.42 percent against NRB’s 90 percent ceiling, according to figures reported by myRepublica. A bank with deposits it cannot lend still pays interest on those deposits. That dead weight pulls the realized spread down regardless of where the legal cap sits. The engine is not throttled by the regulator at the moment. It is starved of fuel, borrowers, and running lean.
So the spread cap and the spread reality bracket the bank from both sides. On a good day, when credit demand is strong, the 4 percent cap stops the bank earning more. On a bad day, like now, weak demand and idle deposits drag the real spread well below the cap. Either way, the core margin has a low ceiling and a soft floor. That is a structurally modest business, not a growth machine.
The profit that was not really lending
Here is where a lot of retail investors got the story backwards in 2025/26, and where the argument of this piece earns its keep.
The headline was cheerful. Combined net profit of the 20 commercial banks jumped to about NPR 69.78 arba in fiscal year 2025/26, up 32.3 percent from NPR 52.81 arba the year before, and 16 of the 20 banks reported positive distributable profit, up from 13, according to sector results reported by The Annapurna Express. A casual reader sees plus 32 percent and assumes the banks are lending like champions again.
They are not. That profit jump was driven mainly by lower impairment charges, meaning banks set aside less against bad loans than the year before, not by a surge in interest income. In fact core interest income at several large banks went sideways or shrank as the spread narrowed. Net interest income at Nepal Bank fell about 6.4 percent and at Rastriya Banijya Bank about 13.4 percent year on year, per results reported by Nepalnews. Read those two facts together and the picture flips. The engine, spread income, sputtered. The profit rose because the previous year’s provisioning bill was unusually heavy and this year’s was lighter by comparison.
That distinction matters enormously for anyone valuing a bank stock. Interest income from the spread is the recurring, repeatable part of a bank’s earnings. A one-year drop in impairment charges is not recurring. It can reverse the moment bad loans tick back up, and Nepal’s non-performing loans have been climbing for two years, a trend we cover in Nepal’s rising bank NPLs. A profit built on a lighter provisioning charge, sitting on top of a narrowing spread, is a profit that can shrink fast. The 2025/26 numbers looked like recovery. Underneath, the money machine got weaker, not stronger.
How to actually read the two numbers on a bank
None of this requires a finance degree. It requires reading two published figures the right way. Here is the practical method.
First, treat the spread as the profitability gauge, not the profit line. A bank running close to the 4 percent cap is extracting the maximum the regulator allows from its core business. A bank stuck well below the cap, like most of the sector today, is telling you its margin is being compressed, usually by expensive legacy deposits, weak lending, or both. Track the spread quarter over quarter. A steadily falling spread is a warning that core earnings are thinning even if the bottom-line profit holds up for other reasons.
Second, use the base rate as a cost and competitiveness signal. A bank with a base rate far above its peers is carrying a higher cost of funds or a heavier cost structure, which forces higher lending rates and makes it harder to win good borrowers. The bank with the lowest base rate in its class can undercut rivals on loan pricing and still make its margin. In a market where the average base rate is near 4.1 to 5 percent, a bank stuck above 6 percent is at a real disadvantage.
Third, and most important, separate spread income from everything else when you judge a profit. Ask what actually drove the year. If profit rose because interest income rose, the spread engine is working. If profit rose because impairment charges fell, be careful, because that lever only pulls once, and it pulls the other way when NPLs climb. This is exactly the skeptical read the reported 32 percent jump deserved. Since NRB’s monetary policy sets the liquidity conditions that push deposit rates, base rates and ultimately spreads around, it is worth learning to read NRB’s monetary policy like an investor so you can see the next move in bank margins before it lands in the accounts.
The verdict
Nepali commercial banking is a capped-margin business, and investors who treat it like an uncapped one will keep overpaying. The regulated spread, 4 percent for commercial banks, sets a hard ceiling on core profitability that no amount of good management can lift. Right now the sector is not even pressing against that ceiling. Deposits are piling up, credit demand is weak, base rates have fallen to record lows near 4 to 5 percent, and realized spreads are drifting below 3.5 percent. The strong-looking profits of 2025/26 leaned on lighter loan-loss charges, not on a healthier lending engine, and that support can vanish as bad loans rise.
None of this makes bank shares uninvestable. Banks still dominate NEPSE, still pay dividends, and a low base rate can eventually revive lending if confidence returns. But the two small numbers the regulator forces onto every bank’s website tell you more about the next two years of dividends than the profit headline does. Read the spread and the base rate first. Read the profit second. When the two disagree, trust the spread.
This is analysis, not financial advice.
