The banks thesis
A bank earns the spread between what it pays for deposits and earns on loans, on a highly leveraged balance sheet where debt is the raw material, which is why it looks permanently cheap: the market is pricing the tail risk that leverage plus a bad credit cycle can wipe the equity out.
Banks are where several of the school's threads converge: they break the standard valuation machine (Other Intrinsic Methods), they are the purest play on rates and the credit cycle (Macro, Fixed Income), and they perpetually screen "cheap," which makes them the best teaching ground for the difference between cheap and undervalued. This guide frames the sector the way a financials analyst does.
Why banks need their own analysis
A bank borrows money as its raw material (deposits and wholesale funding) and lends it as its product, so its debt is not financing to be netted out: it is the business. That inverts the usual toolkit. Enterprise value is meaningless (there is no "operations minus financing" to isolate); free cash flow is undefinable when loans are inventory; and the analysis runs on equity directly, via the bank methods from Other Intrinsic Methods. Everything below assumes that reframing.
The earnings engine, decomposed
| Component | What drives it | The analyst's read |
|---|---|---|
| Net interest income | Net interest margin x earning assets | The core; NIM expands when the bank re-prices assets faster than deposits |
| Fee income | Cards, wealth, investment banking, deposit fees | The diversifier; less capital-intensive, valued at a higher multiple |
| Credit costs (provisions) | The credit cycle; underwriting quality | The swing factor: benign for years, then it is the whole story |
| Operating leverage | Costs vs revenue growth (efficiency ratio) | Well-run banks grow revenue faster than costs; the efficiency ratio tracks it |
| Capital / leverage | Regulatory capital ratios (CET1) | How much equity cushions the assets; sets both safety AND return on equity |
The tension at the center: leverage amplifies ROE (a thin slice of equity under a large asset base), but the same leverage is what makes a credit downturn existential. A bank optimizing ROE by minimizing capital is maximizing fragility, which is why regulators set floors and why the capital ratio is read as a safety gauge and a return driver at once.
The "always cheap" trap
Banks routinely trade at low P/E and near or below tangible book, and screens flag them as bargains constantly. Sometimes they are; often the market is correctly pricing that (a) leverage means a bad enough credit cycle impairs the EQUITY, not just earnings, and the downside is not -30% but -100%; (b) reported book value is an estimate that is least reliable exactly when it matters (loan losses cluster at troughs, and mark-to-market gaps on held-to-maturity bonds sank several banks in 2023); and (c) the business is opaque, a black box of loans whose quality outsiders verify slowly. A low multiple on a levered black box is the market's honest discount for tail risk, not a free lunch. The justified P/TBV formula from Other Intrinsic Methods is how you test whether THIS bank's discount is deserved.
The credit cycle is the master variable
Bank earnings look smooth and high in the back half of an expansion, precisely when risk is building: loans made in the good times default in the bad ones, with a lag. This produces the sector's cruelest trap, a cousin of the cyclical case study: banks look cheapest (low P/E on peak earnings, benign provisions) right before the cycle turns, and most expensive (losses, suspended dividends, dilutive capital raises) near the bottom, when the survivors are the buys. The professional reads provisions and underwriting standards through the cycle, treats a long stretch of unusually LOW credit costs as a warning rather than a virtue, and weights balance-sheet strength most when optimism is highest.
The current debate, framed
- The bull case. Higher-for-longer rates (see the macro thesis) mean structurally wider net interest margins than the ZIRP decade allowed; the strongest deposit franchises fund cheaply and reliably; valuations already embed heavy pessimism; and the best-capitalized banks return large amounts of capital via buybacks below book, which is accretive.
- The bear case. The same higher rates stress commercial real estate and floating-rate borrowers, seeding the next credit cycle; deposit competition (and the 2023 reminder that deposits can flee at digital speed) squeezes funding costs; regulation ratchets capital requirements higher, capping ROE; and fintech/private-credit disintermediation chips at the franchise's edges.
- The synthesis professionals use: banks are not one trade. The great deposit franchises with fortress capital are a different risk object than the thinly-capitalized, CRE-heavy regional. The sector rewards discrimination (which bank, at which point in the credit cycle, at what capital level) far more than a directional "banks up or down" call.
What professionals watch
- Net interest margin trend and deposit betas: how much of rising rates the bank keeps versus passes to depositors.
- Credit-quality leading indicators: non-performing loan formation, net charge-offs versus reserves, and criticized-loan trends, especially in commercial real estate.
- Capital ratios (CET1) and the buyback pace: the cushion and the capital-return story, in one place.
- Deposit stability: uninsured-deposit share and funding mix, the 2023 crisis's permanent addition to the checklist.
- The yield curve: banks borrow short and lend long, so the curve's shape (inversion vs steepening) directly shapes the core spread.
- P/TBV
- Price to tangible book value: market price over book value with goodwill and intangibles stripped out. The standard bank multiple, read against ROE, since a bank earning above its cost of equity deserves more than tangible book and one earning below deserves less.
- Book value
- Shareholders' equity as the balance sheet states it: assets minus liabilities. Meaningful where assets are financial or tangible (banks, insurers), nearly meaningless where the real assets are brands and code.
- Return on equity
- Net income divided by shareholders' equity: what the owners earn on the capital they have in the business. High ROE sustained for years is the signature of a strong business or heavy leverage; the analysis is telling those apart.
- Residual income
- Valuing equity as book value plus the present value of earnings above the required return on that book. Earning exactly the cost of equity adds nothing; only the excess creates value. The academic backbone of bank valuation.
- Cost of equity
- The annual return shareholders require for holding the stock instead of alternatives of similar risk. Unobservable, so it is estimated, most commonly with the CAPM.
- Credit spread
- The extra yield a borrower pays over the government rate: the market's live price of default risk plus illiquidity. Spreads widening is markets losing faith, sector by sector or all at once.
- Yield curve
- Government yields plotted against maturity. Upward-sloping is the normal state; its shape is the bond market's forecast of growth, inflation and policy.
- Regime
- A market era with its own rules: the inflation seventies, the QE 2010s. Backtests assume tomorrow is drawn from the sample's regime; report performance per named regime instead of one blended number.
- Mid-cycle earnings
- A cyclical company's earnings averaged across the cycle rather than taken at the current peak or trough. Valuing cyclicals on spot earnings buys high and sells low by construction.
- Seniority
- The bankruptcy queue: secured lenders, unsecured bondholders, subordinated debt, preferred, then equity last. The same company's different claims can deserve opposite verdicts because they stand in different places.