First American sells the insurance policy that guarantees a buyer (owner's policy) and their lender (lender's policy) that the property title is clean, and runs the escrow and closing that moves the money. Every U.S. home purchase, most refinances, and most commercial property deals need one. It is one of the "Big 4" title underwriters (with Fidelity National, Old Republic and Stewart) that write roughly 80% of U.S. premiums. FY25 revenue was $7.45B; TTM through June 2026 is ~$8.0B.
Title Insurance & Services (~94% of revenue). Direct premiums and escrow fees (~34% of title revenue), premiums written through independent agents (~41%, low margin because the agent keeps ~80%), information and other services (~15%: data, title plant, appraisal, default services), and investment income (~8% of revenue but roughly two-thirds of title pretax profit). Commercial (~$1.2B run-rate, on pace for a record 2026) is the highest-margin slice, with ~$20K revenue per order vs ~$3.9K for a purchase and ~$1.05K for a refi. Home Warranty (~6%): ~$445M of one-year home-system service contracts, ~20% pretax margin. Corporate: holding-company debt and venture investments.
How it makes money: it collects the premium up front, holds escrow cash and claim reserves, and pays claims of only ~3–4% of premiums over many years. Its federal savings bank holds customer escrow and 1031-exchange deposits and invests them, so FAF earns a spread on ~$13B of cash and investments. Key sensitivities: mortgage originations (especially purchase dollars and home prices), the refi share (a refi pays ~a quarter of what a purchase does), commercial real-estate deal volume, short-term rates (investment income), and title-claims experience.
The screen's +116.9% operating-income growth is mostly a one-off. FY24 GAAP earnings were crushed by $401.6M of realized investment losses ($2.91/share), from selling low-yield bonds to reinvest at higher rates (the same tax-planning sales hit 2022 for $516M and 2023 for $206M). GAAP pretax went $165M → $826M and GAAP EPS went $1.26 → $6.00, so any growth rate measured on GAAP or "operating" income laps a loss year. On the company's adjusted basis (ex investment gains/losses and acquired-intangible amortization), EPS went $4.40 → $6.05 (+37%) and the title adjusted pretax margin went 10.3% → 12.2%. That +37% is a genuine cyclical recovery off the 2023–24 trough (closed orders +11%, commercial revenue to ~$1.0B with Q4 +35%, investment income +11%), not a structural step-change. TTM adjusted EPS is $7.09, about 16% above my normalized $6.11.
Normalized, it's worth ~$68. I excluded the 2020–21 refi boom (FY21 GAAP EPS $11.14, adjusted $8.16 on $4.4T of originations). Two lenses on mid-cycle earnings: (1) the driver engine at $2.0T originations, a 30% refi share and a 12% title pretax margin (the 2019/2022–25 average is 12.0%) gives $6.29; (2) the 2019/2022–25 adjusted-EPS average of $5.27, grown 4%/yr into 2026 dollars, gives $5.93. The blend, $6.11, at the 12x mid-point of the 11–13x title-insurer range gives $73.32. The excess-return model on $55.05 of book (ROE fading from 13.0% TTM to 11.0% normalized, cost of equity 10.26%) gives $63.26, because at today's 5.2% risk-free rate a title insurer earning 11% only just clears its hurdle. Blended 50/50: $68.29. Probability-weighted, with a refi-wave blue sky: $73.67.
The market prices FAF exactly as a book-value compounder. $63.68 is 1.16x book, 9.0x TTM adjusted EPS and 10.4x normalized. The reverse valuation says the price needs only $5.74 of normalized EPS (≈ $1.8T of originations at a 12% margin, or an 11.4% sustained ROE): about what 2019 earned, and 19% below TTM. That is a reasonable, not heroic, hurdle. Street targets ($79–96, mean $85; 7 Buy / 1 Hold) capitalize TTM earnings; I don't.
The hidden hedge is the float. Investment income is ~$630M TTM, roughly two-thirds of title adjusted pretax. At the 2023–24 trough, the 9.9–10.3% title margin was almost all investment income; the core title operation barely broke even. Higher-for-longer rates kill refis (MBA's 10/11 forecast: 30-yr above 7%, two more Fed hikes, 2027 originations −5% to $1.99T with refi −17%), but the same rates keep FAF's ~$13B book and bank deposits earning ~4.5–5%. The tornado shows it: ±$100M of investment income moves value ±$9, as much as ±$0.4T of originations. That is why the 2023–24 trough held at an adjusted EPS of ~$4, not ~$2.
Why it isn't cheaper: 2027 is set up as a down year for residential volume, the commercial record (Q2 commercial ARPO $19,980, +31%) is cyclical and lumpy, refi ARPO is a quarter of purchase so refi-led order growth flatters counts more than dollars, and the FHFA/title-waiver debate is a slow-burn threat to the lender's-policy economics.
Call: SLIGHTLY CHEAP. Accumulate below ~$61 (justified P/B 1.1x = $60.90); a starter at today's $63–64 is fine for income accounts. Trim above ~$80 (≈11x TTM adjusted EPS, toward the bull case of $98). Holders: hold. The dividend was just raised 11% to $0.61/qtr ($2.44/yr, 3.8% yield, ~40% of normalized EPS), and buybacks continue near $59–62 (FY25: 2.1M shares at $58.54; Q2-26: 330K at $61.99). Q3 print: Oct 21 after close (consensus $1.86). Watch September refi orders after the rate back-up, commercial ARPO, investment income and deposit balances, and the claims ratio (Q2 3.0% with an $11M prior-year release). Options: after the print, sell a defined-risk 60/55 bull put spread or cash-secured $60 puts 30–45 DTE. The short strike sits at justified-P/B value, with the 3.8% yield behind any assignment. No naked shorts and no calls bought into the print.
| HIGH | Residential purchase volume & mortgage rates. Purchase orders fell to 1,072/day in Q2 (−3% Y/Y), and the MBA sees 2027 purchase volume flat at $1.42T with the 30-yr above 7%. Long: housing turnover and home prices. Short: rates, affordability. Every ±$0.4T of originations moves value about ±$12/share. |
| MED | Refi share. Refis were 34% of closed residential orders in Q2 but earn ~$1,050 each vs $3,900 for a purchase. MBA forecasts refi −17% in 2027. Long falling rates (the blue-sky case); a refi wave lifts counts far more than dollars. |
| MED | Commercial real estate cycle. Commercial is ~$1.2B run-rate and the highest-margin line (Q2 +34%, record pace). Short a CRE-transaction slowdown; large deals are lumpy, so quarters can swing ±$50M. |
| MED | Investment income / float. ~$630M TTM on ~$13.4B of cash and investments plus bank deposits. Long short rates and deposit balances. Short Fed cuts, deposit outflows, and AOCI marks on the bond book when long yields rise (these hit book value). The 2022–24 realized-loss sales show management will trade book value for yield. |
| LOW | Claims & operational. Claims at 3.0–3.75% of premiums, $1.15B reserve, wire-fraud exposure, and the Dec-2023 cyber incident that took systems offline. A long-tail line, but historically well reserved (Q2 prior-year release $11M). |
| LOW | Regulatory / disruption. FHFA title-waiver pilots and attorney-opinion letters threaten lender's-policy pricing at the margin. Industry concentration (Big 4) and FAF's data/title-plant moat slow any change. |
| Originations / refi share ↓ · title pretax margin → | 9.0% | 10.5% | 12.0% | 13.5% | 15.0% |
| $1.6T · 20% refi | $42 | $51 | $60 | $69 | $78 |
| $1.8T · 25% refi | $46 | $55 | $64 | $74 | $84 |
| $2.0T · 30% refi | $48 | $58 | $68 | $78 | $89 |
| $2.2T · 33% refi | $52 | $62 | $73 | $84 | $96 |
| $2.6T · 40% refi | $57 | $69 | $80 | $93 | $106 |
| $3.2T · 50% refi | $64 | $77 | $90 | $105 | $119 |
| Normalized ROE ↓ / cost of equity → | 9.06% | 9.76% | 10.26% | 11.06% | 11.76% |
| 9.0% | $54 | $49 | $45 | $40 | $37 |
| 10.5% | $69 | $61 | $57 | $51 | $47 |
| 12.0% | $84 | $75 | $69 | $62 | $57 |
| 13.5% | $100 | $89 | $82 | $73 | $67 |
| 15.0% | $116 | $103 | $96 | $85 | $78 |
| Year | Orig. $T | Closed orders (k) | Revenue $M | Inv. gains $M | GAAP EPS | Adj. EPS | Title adj. margin | In norm.? |
| 2019 | ≈2.4 | 796 | 6,202 | +66 | $6.22 | $5.76 | 15.2% | yes |
| 2020 | 4.1 | 1,044 | 7,087 | +105 | $6.16 | $5.45 | 14.6% | no (refi boom) |
| 2021 | 4.4 | 1,051 | 9,221 | +436 | $11.14 | $8.16 | 15.6% | no (refi peak) |
| 2022 | 2.3 | 696 | 7,605 | −516 | $2.45 | $6.35 | 12.3% | yes |
| 2023 | 1.6 | 456 | 6,004 | −206 | $2.07 | $3.80 | 9.9% | yes (trough) |
| 2024 | ≈1.8 | 481 | 6,128 | −402 | $1.26 | $4.40 | 10.3% | yes (trough) |
| 2025 | ≈2.0 | 532 | 7,452 | +21 | $6.00 | $6.05 | 12.2% | yes |
| TTM | 2.1E | 541* | 7,984 | +44 | $7.21 | $7.09 | 12.9% | — |
| Ex-peak average adjusted EPS (2019, 2022–25) → grown 4%/yr into 2026 dollars | $5.27 → $5.93 |
| Ex-peak average title adjusted pretax margin | 12.0% |
| Driver engine at mid-cycle: $2.0T, 30% refi → title revenue $7,190M × 12% + home warranty $90M − corporate $100M, 24% tax, 103.0M diluted shares | $6.29 |
| Normalized EPS (average of the two lenses) | $6.11 |
| TTM adjusted EPS (FY25 $6.05 − H1-25 $2.37 + H1-26 $3.41) · premium to normalized | $7.09 · +16% |
| Barchart "+116.9% operating income growth" decomposed: GAAP pretax FY24 → FY25 | $165M → $826M |
| … of which lapping FY24 realized investment losses (−$402M → +$21M) | +$423M (≈64% of the change) |
| … adjusted net income FY24 → FY25 (the real recovery) | $459M → $627M (+37%) |
FAF is a float business. Escrow and bank deposits, claim reserves and unearned premiums are operating liabilities that fund a ~$13B investment book, and investment income is core earnings (~two-thirds of title pretax), not "non-operating cash." There is no clean line between operating capital and financing, so EV/WACC is not meaningful. "Reinvestment" is the statutory and bank capital needed to grow, not capex. So equity is valued directly: normalized EPS × a fair title-insurer P/E (11–13x; 12x mid-point), cross-checked with an excess-return model on book value and a justified P/B of (ROE − g)/(CoE − g). Goodwill and intangibles ($1.9B, mostly data/title-plant and services acquisitions) are left in book because they produce the information revenue. TBV-based multiples are shown for reference.
| Year | ROE | Book (start) | EPS | Excess return | PV |
| 2027 | 12.6% | $55.05 | $6.94 | $1.29 | $1.17 |
| 2028 | 12.2% | $58.17 | $7.10 | $1.13 | $0.93 |
| 2029 | 11.8% | $61.37 | $7.25 | $0.95 | $0.71 |
| 2030 | 11.4% | $64.63 | $7.38 | $0.75 | $0.51 |
| 2031 | 11.0% | $67.95 | $7.50 | $0.52 | $0.32 |
| Terminal | (11.0% − 10.26%) × book $69.99 ÷ (10.26% − 3%) | TV $7.43 | $4.56 | ||
| Excess-return value = BVPS $55.05 + PV excess $3.64 + PV terminal $4.56 | $63.26 |
| Normalized EPS $6.11 × 12x | $73.32 |
| Base value (50/50 blend) | $68.29 |
| Cross-check: justified P/B (11.0% − 3%)/(10.26% − 3%) = 1.11x × BVPS | $60.90 |
| Cross-check: justified forward P/E at CoE 10.26% (payout 1 − g/ROE) | 10.3x → $63.11 |
| Reverse: normalized EPS the price implies (same blend) · sustained ROE (justified P/B) | $5.74 · 11.4% |
| Risk-free (10Y UST) | 5.20% |
| Equity risk premium | 4.40% |
| Beta (bottom-up title/mortgage-linked insurers; estimate, FAF regression ~1.1–1.3) | 1.15 |
| Cost of equity | 10.26% |
| Stable growth (nominal housing turnover × price) | 3.00% |
| Tax rate (Q2-26 effective 22.8%; normalized) | 24% |
| Stockholders' equity · goodwill & intangibles | $5,620.8 · $1,903.7 |
| Shares outstanding (10/15/26) · diluted (Q2-26) | 102.1M · 103.0M |
| Book / tangible book per share (my calculation; not company-reported) | $55.05 / $36.41 |
| P/B · P/TBV · P/E TTM adj. · P/E normalized | 1.16x · 1.75x · 9.0x · 10.4x |
| Cash & equivalents · investments | $2,624.8 · $10,733.7 |
| Notes & contracts payable · debt-to-capital (ex $1.0B bank secured financings) | $1,546.5 · 31.4% (21.5%) |
| Reserve for claim losses | $1,151.6 |
| Q2-26: revenue · adj. EPS · title adj. margin · investment income | $2,117.3 · $2.08 · 14.0% · $164.0 |
| Q2-26 closed orders/day: purchase · refi · commercial; refi share | 1,072 · 555 · 246; 34% |
| Dividend (raised 11% on 9/16/26) · yield | $0.61/qtr · 3.8% |
| Buybacks: FY25 2.1M sh ($122M at $58.54) · Q2-26 330K sh ($20M at $61.99) | ~2% of shares/yr |
| 52-week range · next earnings | $56.20 – $79.87 · Oct 21 (after close) |
| Street: 7 Buy / 1 Hold / 0 Sell, mean target $85 (range $79–96) | Base is -20% vs mean target |