Valuing Insurers with RIM
Insurance companies present a unique valuation challenge: their economic models are rooted in book value (the insurance float, plus invested capital), yet their intrinsic value depends entirely on whether they generate returns above the cost of equity. The residual income model is particularly well-suited to insurance because it values the business as an extension of its equity base. Underwriting profit and investment income are the residual income drivers; the quality and sustainability of these returns determine how much premium shareholders should pay above book value. Understanding how to calibrate RIM for insurers—accounting for float dynamics, underwriting cycles, and the peculiar economics of leveraged balance sheets—unlocks disciplined valuations in an industry notorious for misleading traditional metrics.
Quick definition
RIM values an insurer as book value plus the present value of all future residual income, where residual income equals the spread between return on equity (ROE) and cost of equity, applied to the float (insurance liabilities) plus invested capital. For insurers, book value is the starting point because the float is a liability (claims owed), not revenue.
Key takeaways
- Insurance float (unearned premiums, claims reserves) is leverage; if the insurer generates investment income and underwriting profit, the float amplifies ROE
- Sustainable ROE for an insurer depends on underwriting profitability (loss ratios), investment returns (bond yields, equity dividends), and expense ratios
- Book value growth is driven by retained earnings and float growth; if premiums grow faster than claims, float accumulates and equity expands
- Terminal ROE for insurers typically converges to the cost of equity or slightly below (reflecting commoditizing underwriting and investment returns)
- Cyclical underwriting (hard markets vs. soft markets) makes multi-stage RIM models essential; terminal assumptions should reflect normalized, mid-cycle profitability
- Interest rate environment directly impacts bond portfolio returns and discount rates, creating sensitivity across the valuation
The insurance economics foundation
Before applying RIM, understand how insurance profitability works:
Underwriting profit/loss = Premiums Earned − Claims Paid − Operating Expenses
If earned premiums are $10 billion, incurred claims $7 billion, and expenses $2 billion, underwriting profit is $1 billion. This profit margin depends on the accuracy of claims reserving, loss ratio management, and expense control.
Investment income = Investment portfolio return × float and invested capital
An insurer with $50 billion in invested assets earning an average 3.5% yields $1.75 billion in investment income. This is separate from underwriting profit and often substantial.
Net income = Underwriting Profit + Investment Income − Interest Expense − Taxes
Return on Equity (ROE) = Net Income / Average Shareholders' Equity
The key insight: insurance float is not revenue; it's a source of leverage. If the insurer uses premiums received (float) to invest and generate returns, the float is a form of interest-free capital. If the insurer underwrite-to-loss (combined ratio >100%), the float becomes expensive leverage, destroying value.
Building an insurance RIM model
Step 1: Estimate normalized ROE
Pull three to five years of ROE data, adjusting for underwriting and investment cycle:
Current Period: ROE = 11% (peak underwriting market, high investment returns)
Historical Average: ROE = 9.5% (normalized across cycles)
Industry Median: ROE = 10% (peers' long-term average)
For the explicit forecast (5 years), assume ROE declines gradually from 11% toward 10% as the underwriting cycle normalizes. For terminal value, assume terminal ROE = 10% (industry average).
Step 2: Project equity and float growth
Float growth is driven by premium growth minus claims and expenses. If the insurer writes 8% more premiums, claims grow in line with premium growth (loss ratios stabilize), and expense ratios improve slightly, float grows at ~6% and equity grows via retained earnings.
Year 1 Equity: $50 billion (current book value) Assumed ROE: 11% Assumed payout ratio: 40% Assumed retention: 60%
Net Income: $50B × 11% = $5.5B Retained Earnings: $5.5B × 60% = $3.3B Year 1 Ending Equity: $50B + $3.3B = $53.3B
Repeat for Years 2–5, adjusting ROE and payout assumptions as the cycle normalizes.
Step 3: Calculate residual income
Residual Income (Year t) = Equity(t) × (ROE(t) − Cost of Equity)
Year 1 RI: $53.3B × (11% − 8.5%) = $1.43B
Where cost of equity (8.5%) reflects the insurer's beta (typically 0.7–1.0) and market risk premium, adjusted for financial leverage.
Step 4: Terminal value
Year 5 Equity: $62.5B (estimated after retained earnings growth) Terminal ROE: 10% (converged from peak) Terminal Growth: 3% (GDP growth, assuming float grows with insurance market) Cost of Equity: 8.5%
Terminal RI = $62.5B × (10% − 8.5%) = $937.5M
Terminal Value = $937.5M × (1.03) / (0.085 − 0.03) = $937.5M × 1.03 / 0.055 = $17.57B
Step 5: Present value and intrinsic value
PV(Explicit Period RI) = Sum of discounted residual income Years 1–5 = ~$5.2B (example) PV(Terminal RI) = $17.57B / (1.085^5) = ~$11.8B
Intrinsic Equity Value = $50B (starting book) + $5.2B + $11.8B = $67B
Per-share value = $67B / shares outstanding
Adjusting for float dynamics
Float is unique to insurance and can be a source of competitive advantage (if the company can invest it profitably) or a liability (if underwriting is unprofitable).
Favorable float: Insurer operates at underwriting profit and invests float at market returns. Examples: Berkshire Hathaway, GEICO. Float is valuable and can justify valuations well above book.
Unfavorable float: Insurer writes business at a loss (combined ratio >100%). Float is toxic; the company is borrowing at the loss margin. Values can fall below book as losses accumulate.
Neutral float: Insurer breaks even on underwriting (combined ratio = 100%) and earns market returns on invested assets. ROE approximates the investment return minus financing costs.
Adjust terminal ROE assumptions based on float quality:
- Favorable float: assume terminal ROE stays 200–300 basis points above cost of equity
- Neutral float: assume terminal ROE ≈ cost of equity + 100 basis points
- Unfavorable float: assume terminal ROE converges toward cost of equity or below
Handling underwriting cycles
Insurers operate in cyclical markets. When underwriting becomes unprofitable (soft market), loss ratios rise, combined ratios exceed 100%, and ROE falls. When underwriting becomes profitable (hard market), the opposite occurs. A point-in-time ROE may not reflect normalized profitability.
Solution: Multi-stage model
Stage 1 (Years 1–2): Assume current market conditions (hard market with strong ROE of 12%)
Stage 2 (Years 3–5): Assume transition to mid-cycle (normalized ROE of 10%)
Terminal: Assume perpetual mid-cycle (terminal ROE = 10%, cost of equity = 8.5%)
This avoids the mistake of assuming peak profitability persists indefinitely.
Interest rate sensitivity
Insurance valuations are highly sensitive to interest rates, which affect both the discount rate (cost of equity) and the investment portfolio returns.
Rising rates:
- Cost of equity increases (higher risk-free rate, higher equity premium)
- Valuation formula denominator (COE − g) increases, reducing terminal value
- But future bond yields rise, potentially supporting higher investment income
- Net effect: typically negative short-term (rates rise faster than future income adjusts)
Falling rates:
- Cost of equity decreases
- Terminal value increases
- But future yields fall, pressuring investment income
- Net effect: typically positive short-term (discount rate fall outweighs income pressure)
Include interest rate scenarios in sensitivity analysis, showing how valuations change with ±100 basis point shifts in rates.
Real-world examples
Berkshire Hathaway: Renowned for favorable float (insurance operates at underwriting profit, even in hard markets). Terminals ROE assumptions can exceed 15%, justified by decades of evidence. Book value of $600k per A share has supported valuations at 1.5–2.0x book, with RIM justifying the premium through high sustainable ROE and compounding growth.
Progressive Insurance: Builds float aggressively through growth in auto insurance premiums. ROE has ranged 10–14% across cycles. Mid-cycle RIM model assumes terminal ROE of 11%, cost of equity 8.5%, supporting valuations at 1.2–1.4x book.
Legacy carriers (State Farm, Allstate model): More mature, with declining float in traditional segments offset by profitability. Terminal ROE assumptions cluster around 10%, cost of equity 8–9%, implying modest valuations (0.9–1.1x book) reflecting commoditized underwriting markets.
Common mistakes
Extrapolating peak-cycle ROE indefinitely. In hard underwriting markets, ROEs spike to 15%+ as premium rates rise. Assuming this persists leads to absurdly high valuations. Model reversion to normalized (10–11%) over 5 years.
Ignoring float quality. A 15% ROE earned on negative-float underwriting (combined ratio >100%) is value-destructive. The float is a liability, not an asset. Compare combined ratios across competitors and model their sustainability.
Using nominal bond yields as terminal investment returns without adjusting for inflation. A 4% yield looks reasonable, but if inflation runs 2–3%, real returns are only 1–2%. Use realistic long-term real yields, not cyclical peaks.
Forgetting that float is a liability on the balance sheet. Equity includes float, but float is money owed to policyholders. Growth in float without profit growth doesn't increase book value per share—it increases liabilities. Focus on earnings power and ROE, not absolute float size.
Not stress-testing combined ratios. Underwriting profitability is uncertain. Model scenarios: combined ratio 95% (strong underwriting), 100% (break-even), 105% (soft market). Show how each scenario impacts terminal ROE and intrinsic value.
FAQ
Q: Is book value a valid starting point for insurance valuations?
A: Yes, more so than for most industries. Book value reflects the equity invested (and capital used to build reserves). However, don't assume intrinsic value = book value. Book value is the floor; intrinsic value = book value + PV(future residual income). Valuing above book requires evidence of sustainable ROE > cost of equity.
Q: How do I account for changes in insurance reserves?
A: Reserves are liabilities (claims owed). As new claims emerge and old reserves are paid, book value fluctuates. In the model, focus on normalized reserves (the industry standard for loss reserving, accounting for development). Use multiple-year adjustments to book value if there's evidence of under- or over-reserving.
Q: What cost of equity should I use for an insurer?
A: Start with CAPM: Risk-Free Rate + Beta × Market Risk Premium. Insurance companies typically have betas of 0.7–1.1 (depending on leverage and business mix). For a 2.5% risk-free rate and 6% market risk premium, a 0.9 beta yields cost of equity = 2.5% + 0.9 × 6% = 7.9%. Adjust for company-specific risks (loss reserves uncertainty, interest rate sensitivity).
Q: Should I model the catastrophe risk separately?
A: Yes, especially for property & casualty insurers. Catastrophic losses are tail risks (low probability, high impact) and can wipe out multi-year earnings. Either use higher cost of equity to reflect cat risk, or apply a catastrophe-adjusted loss ratio in terminal assumptions (e.g., assume one major cat event per five years and reflect expected losses in normalized profitability).
Q: How do dividend yields relate to insurance valuations?
A: Insurance companies often pay high dividend yields (3–5%) because they generate steady cash. RIM models this via payout ratio—high payouts reduce equity growth but return cash to shareholders. Don't double-count: model either RIM with payout policy explicitly, or use dividend yield as a reality check (if RIM implies 2% growth but company pays 4% dividend, that's incoherent).
Related concepts
- Return on equity (ROE) for financials: Accounting for leverage and float dynamics
- Insurance float quality and underwriting profitability: The source of residual income in insurer RIM models
- Underwriting cycles and normalized profitability: Multi-stage models addressing cyclical ROE
- Interest rate sensitivity in valuations: Discount rate changes and investment portfolio impacts
- Book value and equity-based valuation: The foundation for insurance RIM models
Summary
RIM is particularly powerful for valuing insurance companies because it centers valuation on the equity base and the spreads between ROE and cost of equity. Start with historical and normalized ROE, adjusting for underwriting and investment cycles. Model float growth as a driver of equity expansion, remembering that favorable float (profitable underwriting) is a source of competitive advantage while unfavorable float (underwriting losses) is a liability. Terminal ROE assumptions should reflect mid-cycle profitability, not peak-cycle peaks; most insurers converge toward 10–11% ROE as underwriting commoditizes. Interest rates impact both the discount rate and investment returns, warranting scenario analysis. Use multi-stage models to capture cycle transitions explicitly. Stress-test combined ratios and reserve adequacy. Finally, remember that valuations above book value require evidence of sustainable ROE exceeding cost of equity; without that evidence, book value is the ceiling on intrinsic value, not the floor.