What Is Net Interest Margin? How Banks Really Make Money

Net interest margin (NIM) is a bank's net interest income, the interest it earns on loans and securities minus the interest it pays on deposits and borrowings, divided by its average earning assets. It measures how much a bank earns on the money it lends and invests, after paying for that money. US banks earned a NIM of about 3.32% in the second quarter of 2026, according to the FDIC, and net interest income makes up 60–80% of revenue for most banks.

NIM is the single most-watched ratio in banking because it shows the core of the business: borrowing short and lending long. It can also mislead. In 2022 Silicon Valley Bank's income statement looked healthy while rising rates had already wiped out roughly its whole equity on paper. To read NIM well, you need to know what drives it and what it leaves out.

How do banks make money from deposits?

A checking balance is not money sitting in a vault. Legally it is a loan from the customer to the bank, repayable on demand and insured by the FDIC up to $250,000 per depositor, per bank, per ownership category. The bank pays little or nothing for that loan and uses the money to make loans and buy securities that pay more.

The difference is net interest income. For most US banks it is 60–80% of revenue, and for many community banks 80–90%. The rest comes from fees: deposit service charges, card interchange, treasury management, wealth management and mortgage banking. Our guide to how financial companies make money sets the bank spread alongside the other revenue engines in finance.

How is net interest margin calculated?

The formulas are short:

Net interest income = interest earned on loans and securities − interest paid on deposits and borrowings

Net interest margin = net interest income ÷ average earning assets

Earning assets are loans, securities and interest-bearing cash, excluding buildings, goodwill and other assets that pay nothing.

A simple illustration: a bank has $10 billion of assets, of which $9.8 billion earn interest. It earns $410 million of interest and pays $100 million to depositors and lenders. Net interest income is $310 million, and NIM is $310 million divided by $9.8 billion, or about 3.2%.

NIM then feeds the rest of the income statement. Subtract noninterest expense and add fees to get pre-provision net revenue, the earnings available to absorb credit losses. Subtract provisions to get pretax income. Return on equity follows from an identity: ROE equals return on assets times leverage. A bank earning 1.2% on assets with eleven times leverage earns about 13% on equity. The FDIC reported industry ROA of about 1.37% in the second quarter of 2026.

What is a good net interest margin?

There is no single good number, because NIM depends on what a bank lends.

A high NIM is not automatically good. Capital One's margin is high because card yields are high, and so are card losses: US card charge-offs were 4.11% of balances in the fourth quarter of 2025. A bank that reaches for a higher NIM by taking more credit risk improves this year's return and damages a later year's. That is why the book's Chapter 5 on banks separates a high return earned from margin, fees and efficiency from one earned from leverage or credit risk.

What moves net interest margin?

Four things, according to the book's commercial and retail banking study:

1. Asset mix. Cards yield more than mortgages, which yield more than Treasuries.
2. Funding mix. Noninterest-bearing deposits cost nothing in interest, though they cost a lot to gather and service. Wholesale funding, brokered deposits and CDs cost close to market rates.
3. Repricing. Floating-rate loans, such as most business loans and credit cards, reprice with the policy rate. Fixed-rate mortgages and bonds do not. Deposits reprice at a rate the bank chooses, subject to competition.
4. The yield curve. Banks lend long and borrow short, so a steep curve helps and an inverted curve, as in the US in 2022–24, hurts.

What is deposit beta and why does it matter?

Deposit beta is the share of a change in market rates that a bank passes on to depositors. If the Fed raises rates by five percentage points and a bank's average deposit cost rises by two, its cumulative beta is 40%.

Across the industry, the cumulative beta in 2022–24 was roughly 40–55%, as reported. A bank whose depositors do not notice rates earns more when rates rise. A bank whose depositors shop for rates, or can leave easily, earns less and is more exposed to a run.

This is the idea of the deposit franchise: the value of being able to pay depositors less than market rates for years. It does not appear on the balance sheet, but it is the main reason a bank is worth more than its book value. Economists at the New York Fed have argued that it works like a hedge, because when rates rise and fixed-rate assets lose value, the benefit of paying depositors below market rises. The hedge works only if depositors stay.

Do rising interest rates help or hurt banks?

Both, depending on the balance sheet. A stylized example in the book's research compares two made-up $10 billion banks through a rate rise from 0.25% to 5.25%, roughly the shape of 2022–23.

Riverside gains because two-thirds of its loans are floating or reprice, and its depositors pass on only about 40% of the rise. Harbor's net interest income halves because more than half its assets are locked into 1.6% bonds, while its depositors demand more interest and move money out of free checking. The headline betas are not far apart. The damage comes from the combination of a rate-sensitive liability side and an asset side that barely reprices. Analysts who watched only betas in 2022 missed this.

What does net interest margin not tell you?

NIM is an income-statement measure. It ignores what has happened to the value of the assets.

The book's stylized $200 billion bank in Chapter 1 holds $90 billion of securities yielding 1.8% with a duration of six. After a 300 basis-point rate rise, those securities lose about 15.8% of their value, an unrealized loss of about 109% of the bank's equity. Yet its NIM falls only from 2.82% to 2.25%. Earnings barely flicker while the bank becomes insolvent on paper.

That is the story of Silicon Valley Bank. It reported net interest income growing strongly through 2021–22 while its held-to-maturity bonds lost about $15 billion of value. US banks as a whole carried about $684 billion of unrealized securities losses at the peak in the third quarter of 2023. Most survived them because their depositors stayed. Our case study on why Silicon Valley Bank failed shows what happened when SVB's did not.

Other businesses that look nothing like banks run on the same spread. Charles Schwab earns much of its income as a bank funded by client cash swept from brokerage accounts, and that income fell in 2022–23 when clients moved idle cash to higher-yielding money funds. Their deposit beta rose, and the margin shrank.

How should you use NIM?

Read NIM alongside three other things:

• The deposit base. What share is noninterest-bearing, insured and from operating accounts? How concentrated are the largest depositors?
• The securities book. How large are unrealized losses relative to equity, including on held-to-maturity bonds?
• Credit costs. Is the margin high because the bank takes more credit risk? Check charge-offs through a whole cycle.

Our step-by-step guide on how to analyze a bank turns these into a checklist. In money-equation terms, NIM is the visible output of two factors working together: whether the bank priced its loans for their risk, and whether its balance sheet, especially its funding, can survive the bad year.

The takeaway

Net interest margin measures how much a bank earns on its earning assets after paying for its funding, and at about 3.32% in mid-2026 it is the engine behind most US bank profits. It rises when a bank has sticky, cheap deposits and assets that reprice, and falls when depositors demand market rates and assets are locked into old yields. But it is a measure of income, not of value, and a bank can report a steady margin while its balance sheet quietly loses its equity. The book explains how banks earn, and how the strongest turn balance-sheet strength into market share, in Chapter 5, "Banks".