Net Interest Margin: What It Measures, How It Cycles, and Why Screeners Miss the Turn

A bank's income statement looks different from almost every other type of company. There is no revenue line labeled "sales" and no cost line labeled "cost of goods sold." Instead, the top of the income statement shows interest income — what the bank earned on loans, securities, and other interest-bearing assets — and interest expense — what it paid on deposits and borrowings. The gap between those two numbers, divided by total interest-earning assets, is net interest margin.

NIM is not a ratio that shows up in most screeners by default. But it is the mechanism that explains most of a community bank's profitability cycle, and understanding it is the prerequisite for making sense of why banks look bad on trailing earnings screens at exactly the wrong time.

What NIM measures

The arithmetic is straightforward. If a bank holds $1 billion in interest-earning assets (loans and securities), earns $45 million in interest income, and pays $12 million in interest expense on its deposits and borrowings, its net interest income is $33 million and its NIM is 3.3% ($33M / $1B).

The two sides of that equation reprice at different speeds, and that asymmetry is the entire source of NIM volatility.

The asset side — primarily loans — reprices on the schedule set by loan terms. Variable-rate loans adjust within days or weeks of a rate change. Fixed-rate loans reprice only when they mature and are renewed. A bank with a portfolio of five- and ten-year fixed-rate commercial real estate loans will still be earning the same yield it locked in years ago, regardless of what the Federal Reserve does today.

The liability side — primarily deposits — reprices partially and with a lag. Savings accounts and money market accounts can be repriced quickly, but depositors have alternatives: if a bank lowers deposit rates too far, depositors move their money. In practice, deposit costs respond to competitive pressure from other banks, money market funds, and Treasury bills, not directly to the Fed Funds rate. This competition means deposit rates often move further and faster in rate-rise environments than a bank would like.

The interaction of these two sides is what creates the NIM cycle.

How the NIM cycle works

When the Federal Reserve raises rates rapidly, the immediate effect on bank NIM is complicated. Fixed-rate assets on the loan book don't reprice — they keep earning the same yield. But deposit costs start rising as depositors demand higher rates and competition for deposits intensifies. In the early stages of a rapid rate hike cycle, a bank can actually see NIM compress: the liability side reprices faster than the asset side.

Over time, as older fixed-rate loans mature and are replaced with new loans at higher rates, the asset side catches up. NIM expands. For banks with shorter loan durations or more variable-rate exposure, this catch-up happens quickly. For banks with long-duration fixed-rate portfolios — common in commercial real estate lending — it can take several years.

The reverse happens when rates fall. Variable-rate loans immediately start earning less. Fixed-rate loans again don't change. But deposit costs are now sticky in the other direction: depositors who locked into higher-rate CDs aren't repricing down until maturity; competitive dynamics mean the bank can't cut deposit rates as fast as market rates fall. NIM compresses.

The result is a NIM cycle that tends to lag the Federal Reserve rate cycle by six months to two years, depending on a bank's asset and liability mix.

Why this creates a screening problem

Most stock screeners use trailing earnings — typically the last twelve months of reported results, sometimes the last fiscal year — as their baseline filter. A bank at the bottom of a NIM compression cycle will show:

• Declining return on assets
• Declining return on equity
• Declining earnings per share, potentially for two or three consecutive years
• Declining net income, often for the same period

All of these are screener inputs. A bank that passes every qualitative test — sound credit quality, experienced management, conservative underwriting — will be automatically removed from institutional screening universes because its recent earnings are depressed by a rate cycle the screener cannot see.

The screener removes the bank precisely when the setup for the next phase is in place. The earnings are at or near the trough. The rate environment has shifted or is shifting. Fixed-rate loans are beginning to mature and replace at higher rates. NIM is recovering or about to. A bank that earns $12 million in a compression trough might be on a path to earning $35–45 million as rates normalize — but the screener sees only the $12 million, removes the bank from consideration, and the institutional buyer never gets there.

By the time the trailing earnings look good again — when NIM has expanded, EPS has recovered, and ROA is back to historical levels — the stock has moved. The opportunity that existed at the trough is priced in.

A concrete illustration

Hingham Institution for Savings (HIFS) provides a clear illustration of this mechanism in the recent rate cycle. HIFS is a Massachusetts savings bank that lends primarily against commercial real estate — a business with inherently long-duration assets. When rates rose sharply in 2022–2023, deposit costs at HIFS rose quickly while the fixed-rate CRE loan portfolio repriced slowly.

The result: NIM compressed to 1.04% in 2024. At that level, the bank's core net income collapsed. GAAP earnings were volatile from equity securities gains and losses. A screener looking at the 2024 income statement would have found a bank earning far less than its historical norm, with apparently erratic results and no analyst coverage. It would not have appeared on any trailing-earnings-based screen.

By 2025, as older loans matured and were replaced at market rates, NIM recovered to 1.70% for the full year and 1.89% in the fourth quarter — still well below the bank's pre-compression levels, but moving in a clear direction. Core net income recovered from $12.3M in 2024 to $32.1M in 2025. The trailing screen now shows improving earnings, but the steepest part of the recovery had already occurred.

A similar dynamic affected Investar Holding Corporation (ISTR), a Louisiana community bank whose NIM and earnings were further complicated by acquisition-related costs during its consolidation phase. In both cases, the metric the screener uses — trailing profitability — was at its lowest point when the business was transitioning toward recovery.

What to look for instead

Screening for banks at a NIM trough requires replacing trailing earnings with forward-looking indicators:

NIM relative to history. A bank reporting NIM 100–200 basis points below its five-year average, with a balance sheet positioned to reprice upward, is not in structural decline — it is at a cyclical low. This requires reading the earnings call or the NIM sensitivity section of the 10-K, not running a screener.

Asset-liability repricing schedules. Most bank 10-Ks include a table showing what percentage of the loan portfolio reprices within one year, within three years, and beyond. A bank with most of its fixed-rate loans in the one-to-three year bucket is in the early stages of an asset repricing tailwind. A bank with most of its fixed-rate loans in the five-to-ten year bucket will see the benefit much later.

Deposit behavior. If a bank has a stable core deposit base — checking accounts and savings accounts with low rate sensitivity — it is better positioned for a rising-rate environment than a bank funded heavily with CDs and wholesale borrowings that must be repriced at market rates. This shows up in the bank's funding cost disclosure, not in any standard screener field.

Credit quality as a separate question. NIM cycles are not credit cycles. A bank can have improving NIM and deteriorating credit quality simultaneously. The correct analysis holds them separate: NIM compression is a rate phenomenon; credit deterioration is an underwriting phenomenon. Confusing them — assuming that a bank with low recent earnings must have bad loans — is the specific error that creates the misclassification.

The structural problem that the community banking screening blind spot describes runs through every layer: the filing structure, the size thresholds, the liquidity minimums, and now the earnings metric. NIM is the mechanism that ensures banks disappear from institutional screens at the moment when the forward-looking case is most interesting. Understanding it doesn't remove the analytical work required — it removes the structural reason not to start.