How do I tell if a bank stock is overvalued or undervalued?

Compare what a bank stock costs to what it should be worth based on its profitability and book value. The most reliable approach uses several valuation methods together (justified P/B, P/E analysis, Graham Number, and peer comparison) and looks for agreement between them. When most methods point in the same direction, the signal is strongest.

No single number tells you whether a bank stock is cheap or expensive. Bank balance sheets are complex, earnings are cyclical, and accounting book value can diverge from economic value for a variety of reasons. The most reliable way to assess valuation is to apply several independent methods and see whether they agree.

Start with the ROE-P/B Framework

The ROE-P/B framework is the most theoretically grounded approach for banks. It calculates a justified P/B multiple based on the bank's return on equity (ROE), its sustainable growth rate, and an estimate of its cost of equity. The justified P/B represents what the stock should trade at if the market priced it purely on fundamentals.

Compare the justified P/B to the bank's actual P/B ratio. If a bank trades at 0.85x book value but the framework says 1.25x is justified, that gap suggests the market is underpricing the bank relative to what its profitability warrants. Flip the scenario: if a bank trades at 1.6x book but the justified multiple is only 1.1x, the stock appears expensive relative to what its returns can support.

The key inputs matter enormously here. Use a normalized ROE rather than a single quarter's figure, since bank earnings fluctuate with credit cycles and interest rate movements. An ROE inflated by unusually low loan loss provisions will overstate the justified P/B, making a fairly valued stock look cheap.

Check P/E in Context

Price-to-earnings gives you a different lens on the same question. Bank P/E ratios typically fall between 8x and 15x, though this range shifts with interest rate cycles and credit conditions. Compare the bank's current P/E to three benchmarks: its own five-year or ten-year average, the average for its peer group, and the P/E implied by dividing its P/B by its ROE.

If all three comparisons suggest the P/E is below normal, the stock is likely priced at a discount. But be careful with P/E during benign credit periods. When a bank books very low provisions for loan losses, reported earnings look inflated, which pushes P/E down.

The low P/E might reflect temporarily high earnings rather than a genuinely cheap stock. Looking at pre-provision net revenue (PPNR) as an alternative earnings measure can help you see through provision noise.

Use the Graham Number as a Sanity Check

The Graham Number estimates a maximum fair price based on earnings per share (EPS) and book value per share (BVPS). For banks with stable, high-quality earnings and reliable book values, a stock trading well below its Graham Number adds another data point supporting undervaluation.

The Graham Number works best as a confirming indicator rather than a primary tool. It doesn't account for growth, competitive advantages, or franchise value, so it tends to be conservative. A bank that consistently trades above its Graham Number isn't necessarily overvalued if its profitability and growth justify a premium. But a bank trading 30% or more below its Graham Number while maintaining strong fundamentals deserves a closer look.

Run an Income-Based Valuation

A dividend discount model (DDM) approaches valuation from the income side. If you can reasonably estimate a bank's future dividends, growth rate, and an appropriate discount rate, the DDM produces a fair value estimate based on the present value of expected cash flows to shareholders.

The DDM is most useful for banks with established, consistent dividend histories. If the model's fair value exceeds the current share price by a meaningful margin and the underlying assumptions are conservative, this provides independent confirmation from the other methods. Be honest about the sensitivity of your assumptions, though. Small changes in the growth rate or discount rate can swing the DDM value by 20% or more, so treat it as directional rather than precise.

Benchmark Against Peers

Peer comparison is the reality check on all of the above. Valuation doesn't exist in a vacuum. A bank trading at 0.9x book value sounds cheap until you realize its closest peers trade at 0.75x because the entire sector is under pressure.

The right comparison group matters. Match banks by size, geography, business mix, and growth profile. A community bank in a fast-growing market shouldn't be compared to a stagnant rural bank twice its size. Once you have a reasonable peer group, compare P/B, P/E, and dividend yield side by side.

If the target bank trades at a meaningful discount to peers while showing comparable or better profitability, efficiency, and asset quality, that discount may represent mispricing. If the discount exists because the bank's fundamentals lag its peers, the lower valuation is the market doing its job.

Convergence Is the Key

No single method is reliable enough on its own to drive a buy or sell decision. The value of running multiple approaches is convergence. When the justified P/B shows undervaluation, the Graham Number supports it, the DDM agrees, and the bank trades at a discount to fundamentally similar peers, you have a strong case that the stock is cheap.

When methods disagree, that's equally informative. A stock might look cheap on P/B but expensive on P/E, which could indicate the market is pricing in deteriorating earnings that haven't hit the book value yet. Conflicting signals mean you need to dig deeper before reaching a conclusion. Investigate which assumptions are driving each method's output and determine which set of assumptions better reflects reality.

Why Cheap-Looking Stocks Aren't Always Bargains

The hardest part of valuation isn't running the numbers. It's figuring out whether the market already knows something you don't. A bank trading at 0.6x book value with an ROE of 4% isn't mispriced. The market is telling you that the bank's returns don't justify a higher price, and the numbers back that up.

Genuine undervaluation tends to exist in a few specific situations:

The bank's recent earnings understate its normalized earning power due to temporary issues (a one-time charge, an unusual provision, a short-term margin squeeze)

The bank operates in a market or segment the investment community largely ignores, particularly smaller community banks with limited analyst coverage

The market is applying sector-wide pessimism to a bank whose individual fundamentals don't warrant the discount

A catalyst for improvement exists (management change, cost restructuring, balance sheet repositioning) that the market hasn't fully priced in

Conversely, be skeptical of apparent undervaluation when credit quality is deteriorating, when the bank has heavy commercial real estate concentration in a softening market, or when insider selling is elevated. These are situations where backward-looking metrics may paint a rosier picture than what lies ahead.

The most rewarding bank stock investments come from identifying the gap between where the market prices a stock today and where fundamentals suggest it belongs, then having the patience and conviction to wait for that gap to close.