Analyzing Banks and Financial Institutions
Why standard fundamental ratios mislead for banks, and how SurgeFlow's sector-aware scoring reads leverage, funding structure, asset quality, and book value across four markets.
Category: Fundamentals
Why banks break the standard playbook
For most companies, the income statement tells the story: revenue comes in, costs go out, and the balance sheet is supporting machinery. For a bank, that picture is inverted — the balance sheet IS the business. A bank's core product is leverage: it borrows short from depositors and wholesale markets, lends long to households and firms, and earns the spread between the two. Metrics that anchor analysis elsewhere lose their meaning here. Gross margin is undefined when there is no cost of goods sold. Free cash flow is close to meaningless when lending out a deposit looks like a cash outflow. Debt-to-equity, a warning flag for an industrial firm, is the operating model for a bank — leverage of ten-to-one or more is normal, not distress. Getting this wrong is one of the most common failure modes in quantitative screens: a naive ratio engine will flag every healthy bank as over-levered and cash-flow-negative. Analyzing financials properly means switching to a different set of lenses — capital, asset quality, funding, and the durability of spread income.
The mechanics: five lenses, in plain terms
Each lens below replaces a standard industrial ratio with one that actually describes a lender. The formulas are simple; the judgment is in knowing what can distort each one. - Earnings quality — net interest margin: NIM = net interest income / average earning assets, where net interest income = interest earned on loans and securities minus interest paid on deposits and borrowings. NIM measures how profitably the balance sheet is deployed — but a rising NIM can also signal rising risk appetite, not just skill. - Leverage and returns: ROE = ROA * (average assets / average equity), where ROE is return on equity and ROA is return on assets. Because assets/equity commonly runs above ten for banks, a small return on assets becomes a large return on equity — and small asset losses become large equity losses. The multiplier works in both directions. - Capital adequacy intuition: the simplest leverage lens is equity / total assets; regulators refine this into capital / risk-weighted assets, where riskier assets consume more capital. The intuition is that capital is the loss-absorbing cushion standing between asset write-downs and depositors. - Asset quality: NPL ratio = non-performing loans / gross loans, and coverage ratio = loan-loss reserves / non-performing loans. Reserves are management estimates, so provisioning choices can flatter or depress reported earnings long before losses are realized. - Funding and liquidity: loan-to-deposit ratio = total loans / total deposits. Deposit-funded banks are generally more stable than those leaning on wholesale funding, which tends to vanish exactly when it is needed most.
How SurgeFlow applies it
SurgeFlow's fundamental scoring on /fundamentals applies sector-aware treatment for banks (and, on similar logic, REITs) behind a sector-fundamentals mode. When a name is classified as a financial, the engine stops treating standard margin and cash-flow ratios as informative: operating margin, free-cash-flow measures, and generic debt ratios are down-weighted or set aside, because for a lender they mostly measure noise. In their place, the scoring leans on leverage and funding structure, book-value-based measures, and the earnings-quality signals described above. An honest note on limits: these are descriptive screens computed from published financial statements. SurgeFlow does not build regulatory capital models — it does not compute risk-weighted assets, does not replicate supervisory capital ratios, and does not run stress tests. A bank can screen well here and still fail a regulator's assessment, and vice versa. The sector-aware mode reduces one specific, well-documented error — mis-reading financials through industrial ratios — but it does not turn the site into a bank examiner.
What SurgeFlow data this uses
The fundamentals behind these screens cover four markets: the United States, China, Japan and Hong Kong. Provenance differs by market: FMP supplies US statements; Tushare covers China and Hong Kong; and Japan combines J-Quants with EDINET filings. That mix matters more for banks than for most sectors: accounting standards, provisioning conventions, and the granularity of loan-book disclosure differ across jurisdictions, so a Chinese bank's NPL ratio and a US bank's are not computed under identical rules. SurgeFlow normalizes fields where it can, and price and turnover history is maintained for every covered ticker even where some fundamental line items are missing — a gap in a bank's disclosed loan detail shows up as an absent input, not a silently imputed one.
Why P/B and E/P read differently for financials
Price-to-book carries more signal for banks than for most sectors because a bank's book value sits unusually close to economic reality: its assets are largely financial claims carried at or near fair value, not decades-old factories at historical cost. The textbook link is justified P/B ≈ (ROE − g) / (r − g), where ROE is return on equity, g is sustainable growth, and r is the cost of equity — a bank earning above its cost of equity merits a premium to book, and one earning below it merits a discount. That is precisely why a very low P/B is not automatically a bargain: it is often the market's statement that reported book overstates recoverable value — that loans are under-provisioned or capital will need rebuilding. Earnings yield (E/P) is similarly slippery: bank earnings are smoothed through provisioning discretion, and cyclically strong earnings near the top of a credit cycle can precede the write-downs that erase them. For financials, book-value and earnings measures should be read together, against capital and asset quality — never in isolation.
Explore it yourself
Every lens in this article can be rebuilt from SurgeFlow's public research API. A free (beta) API key is available at /account, with a limit of 2,000 requests per day, and the full endpoint catalogue is documented in the OpenAPI specification at /api/public/openapi.json. Pull bank fundamentals across the four covered markets, recompute the leverage and book-value ratios yourself, and compare how the same lender screens under industrial ratios versus financial ones. This is research data for studying how banking metrics behave — nothing on SurgeFlow is investment advice, and no output here is a recommendation to buy or sell anything.
Sources and further reading
Primary, academic, and SurgeFlow data references for the claims above: - Diamond, D. W. and Dybvig, P. H. (1983), "Bank Runs, Deposit Insurance, and Liquidity", Journal of Political Economy — why funding structure and liquidity are existential questions for banks. - Diamond, D. W. (1984), "Financial Intermediation and Delegated Monitoring", Review of Economic Studies — the economics of why banks exist as levered lenders. - Merton, R. C. (1977), "An Analytic Derivation of the Cost of Deposit Insurance and Loan Guarantees", Journal of Banking and Finance — capital as an option-like loss-absorbing cushion. - Fama, E. F. and French, K. R. (1992), "The Cross-Section of Expected Stock Returns", Journal of Finance — the classic precedent of treating financials separately because their leverage means something different. - Beatty, A. and Liao, S. (2014), "Financial Accounting in the Banking Industry: A Review of the Empirical Literature", Journal of Accounting and Economics — evidence on provisioning discretion and earnings smoothing. - SurgeFlow routes: /fundamentals (sector-aware fundamental scores), /research/valuation-ratios-for-beginners (ratio basics), /membership and /api/public/openapi.json (public research API).