Financials When the Rate Path Is a Coin Flip
Hike, hold, cut — and the winners that split inside the sector
① "Financials go up when rates rise" is only half right. The sector isn't one thing — it contains sub-groups that react to rates in opposite directions.
② The last year shows it — the financials ETF (XLF) returned +9.5%, well behind the market (SPY +19.0%), while inside it Morgan Stanley was +47% and S&P Global was −13%. The spread within the sector is over 60 points.
③ The debate now is not "cut or hold" but "hold or hike." Rather than bet on direction, it's more useful to split, per scenario, who is favored and who is hurt inside financials.
🏦 Why "financials = a rate-rise trade" is only half right
The common logic goes like this. When rates rise, a bank's net interest margin (NIM) widens — loan rates reprice up fast while deposit rates lag. That part is true.
The problem is that close to half the sector isn't banks. Just the top holdings of XLF, the main US financials ETF, look like this.
| Holding | Weight | Nature |
|---|---|---|
| JPMorgan | 11.7% | Commercial + investment bank |
| Berkshire Hathaway | 11.3% | Insurance / conglomerate (not a bank) |
| Visa + Mastercard | 13.5% | Payment networks (essentially rate-agnostic) |
| BofA · Goldman · Wells Fargo · Morgan Stanley · Citi | 17.1% | Large banks (very different mixes) |
| Charles Schwab | 2.2% | Brokerage / wealth management |
So buying XLF is not "betting on banks." Berkshire plus Visa plus Mastercard alone is 36%, and those move largely independent of NIM. Approach it with the one-liner "rates up, so financials" and a third of what you're buying doesn't run on that logic.
📊 The last year — same sector, a 60-point spread
Instead of generalities, look at the actual returns. Below are trailing one-year returns from the data marketbrief tracks, as of September 4, 2026.
| Sub-group · representative name | 1Y return | P/E |
|---|---|---|
| Investment banking / trading — Morgan Stanley | +47.0% | 17.6 |
| Large bank — Citigroup | +44.4% | 14.8 |
| Investment bank — Goldman Sachs | +40.7% | 16.0 |
| Commercial bank — Bank of America | +25.9% | 14.5 |
| Commercial bank — JPMorgan | +21.8% | 15.4 |
| Payment network — Visa | +9.3% | 31.9 |
| Asset manager — BlackRock | +2.1% | 26.9 |
| Payment network — Mastercard | −0.9% | 31.9 |
| Insurance — Progressive | −10.2% | 11.0 |
| Financial data — S&P Global | −12.9% | 27.0 |
| Alternative assets — KKR | −20.5% | 34.4 |
Same "financials sector," yet the top-to-bottom spread is well over 60 points. Bundle it into a sector ETF and that spread washes out to an average — XLF at +9.5%, which per the sector rotation analysis is near the bottom of the 11 sectors, above only utilities and staples. Strip out just the bank sub-group and the score is different.
| ETF | What it holds | 1Y return |
|---|---|---|
| KRE (regional banks) | Small / mid regional banks | +15.0% |
| KBE (banks) | Large + regional banks | +13.0% |
| XLF (all financials) | Banks + payments + insurance + managers | +9.5% |
Not the banks — the non-bank parts. Payment networks (Visa + Mastercard, 13.5% combined) went sideways, Berkshire (11.3% weight) managed only +1.3%, and asset management, financial data, and alternatives were negative. The large banks' surge only just offset that drag to produce +9.5%.
🔀 How each sub-group reacts to rates
Why the split? Because each sub-group's link to rates is different.
| Sub-group | Link to rates | Reading the last year |
|---|---|---|
| Commercial banks (JPM·BAC·WFC) | Higher rates + a steeper curve → wider NIM. But a sharp hike can reverse it via slowdown and loan-loss fears | Reflected the rate-peak regime; steady |
| Investment banking / trading (GS·MS) | Driven less by the direction of rates than by volatility, volumes, and deal activity | IB and trading revenue surged amid Q2 volatility → best in the sector |
| Regional banks (KRE) | Same direction as large banks. Plus a separate risk from commercial real estate (CRE) loan exposure | Recovered since the 2023 stress; +15%, ahead of the large banks |
| Asset managers (BLK) | Fees track assets under management (AUM) → if rising rates drag bonds and stocks down together, AUM and fees are squeezed | Weak (+2%) |
| Insurance (PGR·CB) | Investment income on the bond float benefits from higher rates. But loss ratios and catastrophes dominate | Mixed — Progressive −10% on loss-ratio issues |
| Payment networks (V·MA) | Nearly rate-agnostic; tied to consumer spending and transaction volume. But high P/E means valuation compresses when rates rise | Sideways — a growth premium held down by rates |
| Exchanges / financial data (ICE·CME·SPGI) | Treated as structural growth stocks with high P/E. Rising rates bring valuation de-rating pressure | Weak (SPGI −13%) |
NIM does best with a gradual rise plus a steepening curve. If short rates spike and the curve inverts, bank margins actually get squeezed. And if hiking fast cools the economy, loan losses eat the margin gains. In fact US bank NIM eased from 3.30% in Q4 2025 to 3.22% in Q1 2026 — asset yields repriced first.
📅 The current regime — "hold or hike"
As of September 2026, the axis of debate has shifted from "cut or hold" to "hold or hike." Sticking to confirmed facts:
- The 10-year Treasury yield has risen to around 5.3%, near the highest since 2007.
- September FOMC (Sept 16) hike odds swung between the 30s and 60s (%) around Jackson Hole, whipsawed by Fed-official remarks.
- Meanwhile jobs are cooling — July payrolls −23K, August ADP +38K, both below expectations.
- August CPI on Sept 11 and the FOMC plus dot plot on Sept 16 will decide the direction.
How to read the data is in reading the economic calendar; how this regime got here is in July CPI & PPI, fully analyzed and the Jackson Hole / Warsh analysis. A cut is closer to a tail risk that materializes only if jobs break down sharply.
🎯 By scenario — winners and losers inside financials
Three paths, framed as relative advantage inside the sector, not sector direction. These are scenarios, not forecasts.
| Scenario | Relatively favored | Relatively hurt |
|---|---|---|
| Hike (September or later) |
Commercial and regional banks (wider NIM) · insurers with large short-bond books | High-P/E growth-type financials — payment networks, exchanges, financial data, asset managers (further valuation de-rating) |
| Prolonged hold (current base case) |
Banks broadly (margins hold at current levels) · investment banks if markets stay active | No clear loser. But if "rates have peaked" takes hold, there's rebound room in the growth-type financials that were held down |
| Cut (premised on a labor-market slump) |
Asset managers (AUM recovers if asset prices rebound) · large, defensive insurers | Banks — this is not a "good cut." Shrinking NII overlaps with recession-driven loan losses. Regional banks are especially exposed |
"Lower rates, more lending, good for banks" is a story for when the economy is fine. The cut being discussed now would come because the economy is cooling, so banks get margin compression and rising defaults at once. That's why you can't mechanically apply "cut = good" to bank stocks.
🛡️ For overseas investors — ETF, sub-group, or single names
- A single XLF position is not a "bank bet." Payment networks plus Berkshire are more than a third of it. For exposure to the NIM story, KBE (banks) or KRE (regional banks) is more direct — bearing in mind regionals carry added CRE risk.
- Read valuation by sub-group. Banks trade at P/E 12–16x; payments, exchanges, and data at 25–32x. The latter wobbles first in a further-hike scenario because of discount-rate sensitivity.
- Individual banks differ too. JPM, BAC, WFC are lending-heavy (NIM); GS and MS are trading- and IB-heavy. The latter react more to market volatility than to rate direction.
- Check portfolio concentration. Whether you're already overweight rate-sensitive assets (banks, REITs, utilities) can be checked with the portfolio diagnostic. FX timing is in the currency calculator.
Financials is not a sector you bet the "rate direction" on — it's one where the winners split inside it with each rate regime. Banks (NIM) benefit from a gradual rate rise and a steeper curve; high-P/E growth-type financials like payments, exchanges, and asset managers do the opposite. Approach it through XLF alone and that spread disappears into the average.
📌 What to watch
| When | Event | Why it matters for financials |
|---|---|---|
| 2026-09-11 | August CPI | The last inflation print before the hike/hold call — directly affects banks vs. growth-type financials relative strength |
| 2026-09-16 | September FOMC + dot plot | Less the decision than the hint at the 2027 path. Curve shape (steepening / inversion) drives the bank-margin outlook |
| Mid-October | Q3 bank earnings | NIM direction, loan-loss provisioning trend, whether IB and trading revenue holds up |
| Ongoing | 10-year yield and the 2s10s spread | If steepening continues, banks are favored; if it flattens or inverts again, the reverse |
| Ongoing | Regional-bank commercial real estate (CRE) loans | KRE's separate risk — any sign of rising delinquencies unwinds the regional-bank premium fast |
The "rate path" this piece deliberately left open in its title was decided on September 16: a hike, to 3.75–4.00%. And yet Goldman Sachs fell 3.96% that same day — the usual rule that hikes favor banks got overridden by Warsh's hawkish forward guidance and unrealized bond-portfolio losses. The "look at sub-groups, not the sector as a whole" principle from the table above held up in practice. The full macro read on that day is covered in Inside the September FOMC.
※ This is a sector analysis based on ETF and stock data marketbrief tracks and public market prices as of September 4, 2026, and is not a recommendation to buy or sell any security. The names, returns, and P/E ratios cited are illustrative for the data discussion and change over time. The scenarios reflect market consensus, not a fixed forecast, and can change with data releases. Investing in stocks carries a risk of loss of principal, and investment decisions and their outcomes are the investor's own responsibility.
- St. Louis Fed — Lower Asset Yields Squeeze Bank Interest Margins (June 2026)
- Yahoo Finance — Which Big Bank Stock Has Dominated in 2026
- NerdWallet — Best-Performing Bank Stocks: September 2026
- CME Group — Understanding the CME FedWatch Tool Methodology
- CNBC — Odds the Fed hikes in September tumble following big July jobs miss
- marketbrief internal data — ETF / stock returns, P/E, analyst consensus (as of 2026-09-04)
※ This report is provided for informational and educational purposes only and does not constitute a recommendation to buy or sell any security.
New research, when it lands
Subscribe and the next deep dive comes to you, along with the daily market brief — free, unsubscribe anytime.