← Files Equity CouncilARCHIVED FILE

references/sector-lenses.md

3.53 KB · Oct 5, 2026 · 18:35 UTC

↓ Download file

# Sector lenses and tangential research

Use only relevant lenses, sourced freshly for the actual industry. These are research questions, not fixed universal valuation rules.

| Business type | Operating/financial questions | Appropriate valuation cross-check |
|---|---|---|
| Banks | Deposit concentration/cost, uninsured funding, credit losses, duration, liquidity, capital and distributions | Equity/distributable capital, normalized ROE and tangible book; avoid generic industrial net-debt/FCF treatment |
| Insurers | Reserve development, combined ratio, investment assets, catastrophe and concentration risk, solvency capital | Equity value, sustainable ROE, excess/distributable capital |
| REITs | Same-property NOI, occupancy, lease roll, tenant concentration, recurring capex/leasing costs, debt ladders | NAV and reconciled FFO/AFFO; FFO alone is not owner cash |
| Software | Cohort retention, gross vs net retention, pricing, sales efficiency, customer budgets, capitalization, SBC and dilution | Cash flow/per-share economics with explicit reinvestment and growth fade |
| Semiconductors/hardware | Cycle, inventory, utilization, customer concentration, node/yield roadmaps, capex commitments, export controls | Mid-cycle earnings/FCF, capacity economics, scenarios for customer capex |
| Commodities | Cost curve, reserves/depletion, sustaining capex, commodity cycle, hedging, jurisdictions and closure liabilities | Mid-cycle NAV/DCF across commodity prices; never extrapolate peak spot margins |
| Utilities/infrastructure | Rate base, allowed vs earned returns, regulatory lag, financing needs, load, outages and stranded assets | Regulated cash flows/distributions and funding dilution |
| Biotech/pharma | Asset-specific evidence, trial endpoints, stage, indication/modality reference class, patents, commercialization and runway | Conditional event-tree/risk-adjusted value with financing; broad FDA phase statistics are not asset odds |
| Industrials | Order quality/cancellations, capacity, replacement cycle, customer capex, aftermarket, working capital | Through-cycle cash generation, installed-base economics |
| Consumer/retail | Unit volumes vs price/mix, repeat purchase, inventory, store/cohort economics, channels and substitution | Unit-level cash returns, mature margins, reinvestment economics |
| Platforms/marketplaces | Liquidity/network effects, multihoming, take rate, subsidies, worker/supplier economics, antitrust | Cohort/unit cash flow with competitive fade and realistic monetization |

## Causal tangent filter

For every adjacent topic record: external development -> operating driver -> cash-flow/balance-sheet effect -> shareholder return; exposed candidates; plausible magnitude; timing; evidence strength; research question; rank-changing condition.

Investigate high-impact channels such as power/water limits for compute infrastructure, supplier bottlenecks, customer financing, trade restrictions, substitution, legal remedies or technology transitions. Evaluate both benefits and costs; industry demand growth can benefit customers or suppliers while destroying returns for new entrants.

Prioritize by expected decision value: plausible impact on rank × uncertainty that can be resolved × feasibility of obtaining reliable evidence. Use qualitative high/medium/low bands unless a quantitative value-of-information model is defensible. Expand one causal link at a time. Stop when the next link has negligible plausible effect or no resolvable evidence; park it with a reason. The goal is broad relevant coverage, not a claim to analyze literally everything.

SHA-256: a11ee27fcc420aff8cabc567b8abe36c029d8f2fa03739a9121e1976f7df5cac