Fidelity National Financial, Inc. (NYSE: FNF) — The Best Title Franchise Money Can Buy, Stapled to a Spread Lender at Par

Independent Equity Research | Report date: 2026-07-19 | Prices as of 2026-07-17 close (FNF $52.04; FG $32.41)

⚡ Kimi’s Take

Kimi’s own subjective opinion; general information only, not investment advice. The analytical body below is deliberately position-free — it carries no recommendation, no price target, and no buy/sell language; everything opinionated lives in this block alone.

Call: HOLD — fairly priced at ~$52. Accumulate only on a title-cycle washout toward the low $40s — roughly the March-2026 low of $42.79, where the $2.08 dividend yields ~4.9% and the Title engine would be changing hands at ~10x mid-cycle earnings with the listed FG stake and holdco net debt as ballast. That is the zone where the margin of safety actually appears; at $52 it does not.

The framing is quality-at-a-fair-price / income vehicle, not contrarian value and not momentum. FNF is the best title-insurance franchise in America — industry-best pretax margins every single year 2021–25, ~1,000bp above First American at the 2024 trough, #1 family share at ~32%, a conservative mechanical 4.5% loss provision — stapled to F&G, a Blackstone-armed annuity spread lender earning ~8.2–8.4% adjusted ROE ex-AOCI that the market itself prices at 0.70x book ex-AOCI, i.e. below its cost of equity. The whole package trades at ~4% above our base SOTP: Title at ~$12.1B (FAF’s own 11.2x multiple), the ~70% FG stake fully marked at $3.04B, holdco net debt fully deducted. There is no conglomerate discount left to harvest — the two in-specie distributions did their job. What remains is a narrow debate: does the demonstrably superior title franchise deserve more than the token premium to FAF’s multiple the market currently pays (~11.7x implied vs FAF 11.2x on flattered earnings), and does F&G’s fee-based pivot ever lift ROE above ~9%? The tape agrees this is a stalemate: a range-bound, dividend-anchored value financial (beta 0.56, ~4% yield, negative momentum loading, Value/DividendYield regime in favor) whose falling-knife phase ended at $42.79 in March 2026 and which has since reclaimed all three EMAs without being re-rated. Not a short: the title franchise is genuinely superior, the dividend was raised through the 2023 trough, and SOTP shows no froth. Not a buy here: no discount to SOTP, F&G is growing AUM at returns below its cost of equity (funded partly by a 2025 equity raise at ~0.7x book that FNF itself joined for $151M), zero insiders bought FNF in 24 months — including the $43 dip — and the incentive plan pays 200% on title-only targets that were set below the prior year’s actuals.

Conviction: medium. Bull-flip trigger: a purchase-market reopening (mortgage rates durably below ~6%) driving title orders toward ~1.3M+ with adjusted margin holding ≥13.5–14% — volume outrunning refi-mix dilution would make the base-case ~$5.5–6.0 EPS path look conservative and the ~10.2x adjusted multiple cheap. Bear-flip trigger: evidence that F&G’s product-margin compression (1.88% → 1.66% annualized in Q1-26) is a trend rather than one-off timing, confirmed in the Q2-26 supplement, especially if paired with the first real credit-vintage impairments on its ~50% structured/private-credit book.

Tag: Best house on the title street — but the F&G annex is priced at full cost.

📈 Stock Price Action — Five-Year Event Map

Factual price history, not a call. Price moves are FACT; attributed drivers are INTERPRETATION, labeled as such. Prices are split- and dividend-adjusted (the two F&G stock distributions make the adjustment material); unadjusted noted where relevant. Source: AZI 5-year adjusted series through 2026-07-17, cross-referenced to 8-K filings, earnings prints, and the peer event map from a companion analysis of First American.

Arc. Over the trailing five years FNF round-tripped a full housing/rate cycle at roughly half the market’s beta: from ~$33 (adj, Jul 2021) through two separate −25%+ drawdowns into a double bottom ($27.34 on 2022-06-16 and $27.38 on 2023-03-17), a two-year grind to an all-time adjusted high of $60.03 (2025-04-02; $66.59 unadjusted), then a choppy 2025–26 producing a −19% flush to the 52-week low of $42.79 (2026-03-20) and a V-shaped recovery. The 52-week range is $42.79–$56.37; the stock closed 2026-07-17 at $51.90, −13.5% off the five-year high and +4.2% above its 200-day EMA. Five-year cumulative price return: +58%.

Cycle narrative. Events 1–3 are the rate shock and its double bottom: a geared title-volume play repriced for mortgage rates doubling, compounded by F&G’s fixed-income marks, with the SVB panic over insurer bond portfolios producing the second −25% leg on mark-to-market fear rather than title fundamentals (INTERPRETATION; 10-K FY22 recognized-losses line, SEC-sweep 8-K timeline). Events 4–5 show the stock trading as a rate-duration proxy, absorbing idiosyncratic events (the November 2023 ransomware disclosure cost it −1.5% on the day) when the macro wind was favorable (INTERPRETATION; 8-K 2023-11-21; FAF peer event map for the 2024-04-10 CPI sympathy). Events 6–8 complete the pattern: every major up-move since 2022 has coincided with falling-rate windows or F&G separation news, every major down-move with rising-rate windows or F&G mark-driven prints — the dominant macro exposure is rates, not the equity market. The February–April 2026 flush-and-recovery is the cleanest illustration: a sector-wide rate risk-off took the stock to $42.79 with zero company filings, and the title order inflection visible in the Q1-26 10-Q brought it back — though the tape faded FNF’s own Q1 print in May, having priced the beat in April (FACT: price moves; INTERP: attribution).

1. Executive Summary

Fidelity National Financial is two insurance businesses under one roof. The first is the strongest title-insurance franchise in the United States: a five-underwriter family (Fidelity National Title, Chicago Title, Commonwealth, Alamo, National Title of New York) holding a ~32% family share — #1 for five straight years — that earned pretax margins of 18.6% at the 2021 peak and never below 12.0% at the trough, beating First American’s margin every single year 2021–25, including a ~1,000bp gap at FAF’s 2024 trough (4.3% vs FNF’s 14.2%). Title generates most of consolidated pretax profit on ~$8B of segment assets, books a conservative flat 4.5% loss provision, and today sits cyclically low on volume (956K closed orders in 2025 = 44% of the 2021 peak) but near-peak on margin (15.9% adjusted in FY2025).

The second business is F&G Annuities & Life (NYSE: FG), a ~70%-owned, separately listed spread lender: $74.5B of AUM before flow reinsurance, record $14.6B gross sales in 2025, a Blackstone-managed investment book that is roughly half structured/private credit, 93% surrender-charge-protected liabilities, RBC ~430% — and an adjusted ROE ex-AOCI of just 8.2–8.4%, below any reasonable cost-of-equity estimate, which is why the market prices FG at 0.70x book ex-AOCI. F&G’s GAAP earnings are dominated by LDTI fair-value noise (segment pretax of −$35M in 2023, +$778M in 2024, +$323M in 2025) and must be read on the adjusted basis. A Corporate & Other segment is a structural drag (~$24–153M pretax losses per year).

The valuation picture is unusually clean because the market marks ~22% of FNF’s value every day via the FG quote. Our SOTP — Title standalone at FAF’s own 11.2x multiple (~$12.1B), the FG stake at market ($3.04B), less holdco net debt ($1.66B) — produces ~$13.5B, roughly $50/share against a $52.04 price: the market prices the structure correctly and there is no conglomerate discount to harvest. TTM adjusted EPS is ~$5.12 (GAAP TTM $2.81 is distorted by the $471M deferred-tax charge on the December 2025 F&G distribution and LDTI marks — the operative multiple is ~10.2x adjusted, not the 18.5x GAAP print). Normalized EPS scenarios run bear ~$3.6–4.1 / base ~$5.5–6.0 / bull ~$6.8–7.3, with normalization running up on title volume and down on title margin simultaneously.

Capital allocation is genuinely disciplined on the dividend (raised every year including the 2023 trough at a 52% payout) and the F&G round trip created value (~1.4–1.5x over six years), but buyback timing has been pro-cyclical, the separation was executed as taxable drips that burned the §355 tax-free option and crystallized $471M of deferred tax, and the incentive plan pays executives on title-only metrics with 2025 targets set below 2024 actuals — a real governance cost atop the Foley-network related-party lattice. The balance of this memo argues that FNF is a high-quality cyclical franchise attached to a mid-quality spread lender, priced at roughly the sum of those parts.

2. Business Overview

2.1 What the company is

Fidelity National Financial, Inc. (NYSE: FNF, CIK 0001331875) is a Nevada-domiciled (since June 2025) insurance holding company with three reportable segments. Title is the core: the #1 U.S. title-insurance family, issuing policies through five underwriters — Fidelity National Title Insurance Company, Chicago Title, Commonwealth Land Title, Alamo Title, and National Title of New York — which “collectively issue more title insurance policies than any other title company in the United States” (10-K FY2025, Item 1). Around the underwriters sit escrow and closing services, ServiceLink mortgage transaction services, loan sub-servicing, valuations, default services, and a home-warranty business. Distribution runs through ~1,300 direct offices and ~5,100 independent agents, “among the largest in the United States.”

F&G is the ~70%-owned (post-December-2025 distribution; ~85% is a stale figure) F&G Annuities & Life, Inc. (NYSE: FG): fixed indexed annuities (FIA), registered index-linked annuities (RILA, entered early 2024), fixed-rate/MYGA annuities, immediate annuities, indexed universal life, funding agreements (FABN + FHLB Atlanta), and pension risk transfer, written through FGL Insurance (Iowa-domiciled) and FGL NY Insurance out of Des Moines with ~1,100 employees, ~778,000 retail policyholders, and ~145,000 PRT participants. Corporate & Other holds the parent holding company, real-estate technology subsidiaries, remaining real-estate brokerage businesses, unallocated overhead, and eliminations.

2.2 Segment economics through the cycle

The five-year segment record shows two engines with opposite cycle signatures (FACT — 10-K FY2022/FY2025 segment footnotes and MD&A):

Segment assets at YE2025: Title $8,093M; F&G $98,432M; Corporate $2,489M (goodwill: Title $2,799M; F&G $2,180M). The structure in one sentence: a transaction-driven title oligopolist generates most of the profit on ~$8B of assets, while a ~$98B-asset spread lender contributes a minority of economic earnings but dominates the balance sheet and GAAP volatility (INTERPRETATION — F&G’s pretax swung from −$35M to +$778M to +$323M across 2023–25 on market-risk-benefit marks of $95M / −$25M / $167M; LDTI accounting noise, not spread deterioration; see §6). The counter-cyclical claim is real but partial: F&G’s GAAP pretax peaked in FY2024 as title troughed, and title peaked in FY2021 as F&G was rate-battered — but the smoothing is untested in a hard landing that hits title volumes and F&G credit simultaneously, and no operating synergies between title escrow and annuity manufacturing exist or are claimed.

2.3 Revenue mix: who keeps the premium dollar

Title’s FY2025 revenue of $8,490M decomposes as: direct premiums $2,574M (44.2% of title premiums), agency premiums $3,250M (55.8%), escrow/title-related/other fees $2,381M, and interest & investment income $363M (FACT — 10-K FY2025 Title MD&A). The agency channel is the critical economic detail: independent agents retained $2,518M of the $3,250M of agent premiums — 77.5% (77.4% in 2024, 76.9% in 2023) — so FNF keeps only ~22.5% of agent-originated premium dollars. FNF is materially more agency-weighted than First American (~40% agent share of revenue, per FAF’s own disclosures), which cuts both ways: the agency channel is variable-cost, so FNF’s cost base flexes down automatically in a downturn (agent commissions fall with volume), explaining its shallower margin troughs; but it captures less of each premium dollar on over half its book, capping peak economics.

Operating metrics (direct operations): orders closed 956K in FY2025 (879K FY2024, 837K FY2023 — against a 2021 peak of 2,169K); orders opened 1,411K; average fee per file $3,948 ($3,742 in 2024, $3,617 in 2023, $2,467 in 2021); closed-order mix 72.2% purchase / 27.8% refinance (FACT — 10-K MD&A). Fee-per-file gains are mix and home-price appreciation, not pricing power — residential fee per file was flat year-over-year in Q1-26 per management, with commercial carrying the upside (FNF Q1-26 call). Personnel costs run ~60% of direct premiums plus escrow fees; average Title headcount 22,248. Commercial title revenue — disclosed only in quarterly operating statistics — was $1,494M in FY2025 (+27%; third-best year ever) and $338M in Q1-26 (+15%).

2.4 The F&G machine

F&G is a spread lender: it sells long-duration, mostly surrender-protected annuity liabilities and earns the spread between asset yields and crediting rates. FY2025 gross sales were a record $14.6B (mix ~46% FIA, 26% fixed-rate/MYGA, 15% PRT, 12% funding agreements, 1% IUL); AUM before flow reinsurance reached $73.1B at YE2025 and $74.5B at Q1-26; retained AUM was $57.6B at YE2025, dipping to $56.4B at Q1-26 after a $1.8B in-force cession tied to the F&G Life Re sale (FACT — F&G Q4-25/Q1-26 releases and supplement). Liability quality is the strong suit: ~93% of the $36.1B FIA + fixed-rate account value is surrender-charge protected; funding agreements, PRT and immediate annuities are non-surrenderable; average crediting rate on fixed-rate annuities 5%; products reset annually, allowing active spread management. Blackstone ISG-I Advisors manages ~78% of the $69B portfolio — a purchased origination capability that is F&G’s one genuine supply-side edge, and a single-manager concentration. Distribution runs through ~300 IMOs and ~26 banks/broker-dealers representing ~187,000 agents, plus an owned-distribution build-out (Syncis, DCMT, Roar JV, PALH; $89M revenue in FY2025) and the new Fort Greene Re sidecar (~$1B of Blackstone-backed third-party capital) intended to make growth fee-based rather than balance-sheet-based.

Verdict (Business Overview): a two-engine insurance conglomerate. The title franchise is a genuine, cash-generative oligopoly asset — 18.6% peak / 12.0% trough pretax margins over five years — and F&G is a leveraged spread lender whose asset growth is real but whose reported earnings obscure economics that, on the adjusted basis, are adequate rather than attractive. Corporate & Other is a drag plus optionality on real-estate-tech ventures. Per-share analysis must net out the ~30% F&G minority interest, which became first-order after the December 2025 distribution (NCI took $78M of Q1-26 consolidated net earnings vs ~$0 in Q1-25).

3. Industry Dynamics

3.1 Title insurance: a regulated oligopoly taxing real-estate transactions

This subsection draws on the title-industry analysis from a companion analysis of First American Financial (FAF) published alongside this report; key numbers were spot-verified against FNF’s own 10-K with no contradictions found.

The U.S. title industry wrote ~$18.5B of premiums in 2025, +13.8% vs $16.2B in 2024, and paid $667M in claims — a ~3.6% loss ratio, extraordinarily low because title risk is retrospective (claims arise from pre-existing record defects). Fitch puts sector operating margin at 11.4% in 2025 vs 10.3% in 2024 (FACT — ALTA via HousingWire; Fitch via Beinsure). Structure is a Big-4 family oligopoly: FNF, First American, Old Republic, and Stewart wrote 80% of 2024 net premiums (FNF 10-K; ~40 independents split the rest), the top seven ~85%. ALTA 2025 individual-underwriter shares: First American Title 23.1%, Fidelity National Title 14.5%, Old Republic 14.0%, Chicago Title 13.1%, Stewart 10.9% — and because FNF owns three of those plus Commonwealth and Alamo, the FNF family is #1 at ~32% (company-stated ALTA share through Q3 2025; the companion FAF analysis’s ~27–30% roll-up uses a different manual methodology, direction identical).

Demand is a transaction tax on the mortgage origination cycle. The 2022–24 trough was historic — 30-year rates ~3% → ~6–7%, refi collapse, existing-home sales near 4M units for three straight years, among the lowest in three decades. The 2026 recovery forecasts, all hostage to a 6.0–6.5% mortgage-rate assumption (FACT — MBA/iEmergent/NAR): MBA 2026 originations $2.19–2.2T, +7–8% with refis revised up to +11%; iEmergent $2.235T (+15%); NAR existing-home sales +4.0% to 4.22M. Commercial title is the cycle’s differentiating leg: MBA forecasts CREF volumes +24% in 2026. Regulation is state-centric and double-edged — rate setting caps pricing power but suppresses price wars and deters entry; RESPA §8/CFPB referral-fee enforcement is a perennial overhang; ALTA notes title costs have fallen ~7.8% nationally since 2004. The structural verdict: a decent, not great, industry — genuine Greenwald barriers protecting a low-loss, transaction-driven profit pool, with the agency channel leaking ~80% of agent premiums downstream and inescapable beta to originations. 2025–26 is an early-to-mid-cycle recovery; nothing in the structure makes current margins permanent (INTERPRETATION).

FNF-specific industry angles (FACT — 10-K FY2025 unless noted): the multi-brand strategy is explicit — “we must operate our strongest brands in a given marketplace independently of each other” — which is how one parent holds ~32% while no single FNF brand exceeds ~15%: the brands compete against each other for agents and shelf space that a single-brand rival must win head-on. FNF is also the industry’s most aggressive vertical integrator into the agency channel’s consolidation wave, taking “minority and majority stakes in various network marketing groups and other distribution consolidators” — an attempt to recapture part of the agent-side economics the industry cedes (INTERPRETATION: distinctive vs FAF, but it blurs the underwriter/agent line and will attract RESPA-adjacent affiliated-business scrutiny). Commercial title runs through direct operations in all major markets (Atlanta, Boston, Chicago, Dallas, Houston, LA, NY, Philadelphia, Phoenix, Seattle, DC), one of the largest commercial networks in the industry, with >$100M face-amount capability and coinsurance for oversized risks. Geographic concentration: Texas 14.1% and California 12.2% of 2025 title premiums (CA = 25.7% of direct premiums) — state-level regulatory shock in either state is an FNF-specific exposure.

3.2 U.S. annuities: record demand, transformed supply, rising regulatory heat

Built from F&G releases, LIMRA, and trade press (F&G files separately, CIK 0001873863; figures here are from the FNF 10-K segment and F&G public reporting).

Market size and growth. U.S. individual annuity sales set a fourth consecutive record in 2025: $461.3B preliminary, +6% (finalized $464.1B, +7%), with nine consecutive $100B+ quarters (FACT — LIMRA, Feb 2026). The series: $313B (2022) → $385B (2023) → ~$432B (2024) → $461–464B (2025). Fixed indexed annuities — F&G’s core — hit a category record of ~$116B, +12% (one LIMRA-derived aggregator series shows $127.9B; the gap is survey-coverage methodology, unresolved — FNF’s own 10-K states the FIA market was $130B in 2024, closer to the higher series). RILAs reached $79.6B, +20%; indexed products (FIA+RILA) are now 45% of total sales vs 24% a decade earlier. LIMRA projects >$450B in 2026 despite gradual Fed cuts, leaning on “money in motion” — maturing 2021–23 MYGAs rolling at re-market rates. Demand drivers are genuinely secular: “Peak 65” demographics (4.1M Americans turning 65 per year), the decline of defined-benefit pensions, elevated-but-falling rates, equity volatility. Against a $461B industry, F&G’s $14.6B of gross sales is ~3% of the flow — a mid-tier player in a market where Athene alone wrote ~$33B of fixed annuities (FACT; INTERPRETATION on share).

Structure: the private-credit-backed model vs. traditional. Asset-manager-owned insurers have consolidated the industry: Apollo/Athene (~$331–355B assets, #1 in retail annuities and #1 in FIA for the third straight year, $15.0B of 2025 FIA sales), KKR/Global Atlantic (~$150–158B), Brookfield, Ares/Aspida, Carlyle/Fortitude; McKinsey estimates private-capital-backed insurers gathered ~$700B of AUM through 2023 — ~13% of the industry vs 1% in 2012. The model’s economics: annuity deposits provide long-duration sticky liabilities; the affiliated asset manager originates higher-yielding private credit against them; the parent earns insurance spread and asset-management fees. Traditional mutuals and independents — F&G among them — compete for the same IMO/agent shelf without a captive origination engine (FACT — LIMRA/trade press aggregations). F&G’s strategic response is to shift toward “a more fee-based, higher margin and less capital intensive business model”: the Fort Greene sidecar, the sale of its Bermuda reinsurer F&G Life Re (closed 2026-03-01), the Peak Altitude vehicle, and third-party flow reinsurance (> $15B of cumulative new business reinsured). INTERPRETATION: F&G is a scale-disadvantaged spread lender trying to convert itself into a capital-light manufacturer/distributor before regulators raise the cost of being a spread lender; selling Bermuda is a deliberate reduction of regulatory surface area.

Regulation is where the real action is. The NAIC life RBC overhaul is in flight: Proposal 2025-16-L (new capital factors for CLOs/private credit), comments closed January 2026 on RBC changes targeting “asset-intensive” reinsurance, a credit-rating due-diligence framework, and new economic scenario generators for stress-testing PE-backed portfolios finalized at the March 2026 Spring National Meeting. Federal escalation: the U.S. Treasury convened regulators on private credit and offshore reinsurance in April–May 2026; the Bermuda long-term reinsurance sector manages ~$1.52T of assets tied mostly to U.S. policyholders. Exposure metrics: U.S. insurers held $276.8B of CLOs at YE2024 (doubled since 2018), life insurers hold 82% of that, and 137 PE-owned insurers hold ~21% of insurer CLO exposure at materially lower average quality; ALIRT counts life-industry Level-3 bond fair value up from $183B (2016) to $464B (FACT — trade press; not verified against regulator originals). INTERPRETATION: regulation is moving against the PE model’s two pillars — cheap capital treatment of private assets and offshore reinsurance arbitrage — a slow-motion equalizer that directionally helps independents like F&G at the margin while raising compliance costs for everyone.

Spread economics and the capital cycle. The post-2022 rate rise transformed industry economics: new-money yields rose faster than crediting rates reset, producing record spreads and the 2022–25 sales boom (MYGA rates >6% since 2023); the same rate rise raised surrender incentives — Oliver Wyman found lapse rates rose ~5x in 2023 vs 2021 for high-guarantee policies past their surrender-charge period. The industry’s P&L is now short-rate-path dependent in both directions: falling rates compress new-money spreads; a re-rise re-ignites surrender risk on the 2022–24 low-crediting in-force block (FACT; INTERPRETATION). Through the Marathon capital-cycle lens: four consecutive record sales years with capacity flooding in via PE entrants, sidecars, and offshore reinsurance; 2026 looks like a plateau at record levels — late in the boom phase, not early in a bust. Returns on new money are wide but past peak; the marginal competitive weapon has shifted from crediting-rate generosity to distribution breadth and product innovation (RILA). The stored-up tail risk is the 2021–24 private-credit vintage bought at peak competition — concentrated in PE-backed writers; F&G’s book (97% investment-grade, 7bp average annual impairments over three years) is cleaner than that cohort but not untested (INTERPRETATION).

Industry verdict (argued): Title — a decent, not great, industry in an early-to-mid-cycle recovery, with FNF holding the best structural seat (#1 family share, largest commercial network, unique owned-distribution strategy). Annuities — an industry with genuinely secular demand and contested economics, late in its capital cycle, whose profit pool is migrating toward asset-manager-integrated spread lenders; regulation is slowly moving against that model’s capital arbitrage. F&G sits in the defensible-niche camp — mid-tier scale, no captive origination, exiting offshore reinsurance, pivoting to fees — a position that can earn its spread but does not set industry economics.

4. Competitive Position

4.1 Title: the industry’s best seat, evidenced by the trough

The title moat mechanisms, mapped to Greenwald and each tied to a metric (FACT — 10-K; INTERPRETATION — the mapping): (1) Title plants and proprietary data — decades-built indexed record compilations that cannot be replicated quickly (management: its data “cannot be replicated by simply digitizing public records,” Q1-26 call); the cost advantage shows up in the stable 4.5% loss provision. (2) Regulatory barriers — 50-state licensing, rate filings, statutory capital; caps pricing power but blocks entry. (3) Scale economies — the largest direct-office plus agent network spreads plant and technology fixed costs over ~40% more premium volume than FAF (~32% vs ~23% family share). (4) Distribution captivity — weak: agents represent multiple underwriters and keep ~77.5% of agent premiums; captivity is real but contestable on splits and service. Switching costs are distribution-level friction (relationships, escrow integration, lender panels), not contractual; end consumers do not choose the insurer. No network-effect claim survives scrutiny — the advantages are supply-side only.

The multi-brand strategy deserves a direct verdict, because it is the structural choice that distinguishes FNF from FAF’s single-flagship model. Five years of evidence say it works: FNF’s family share held 30.7–32.6% through a full cycle — answering the longitudinal share-defense question a companion First American analysis left open — while its pretax margin beat FAF every year:

(FACT — FNF 10-Ks; FAF figures per its 10-K filings, via the companion First American analysis. FNF 2025 margin is exactly 14.45%; shown rounded.) The mechanisms (INTERPRETATION): multiple brands hold shelf space with agents and local customers that a single brand would lose, defending volume without price competition; the heavier agency mix (55.8% of premiums) flexes costs down automatically in downturns while FAF’s direct-heavy model eats fixed costs; and scale spreads fixed plant/tech costs. The 2024 trough is the proof — FNF held 14.2% while FAF fell to 4.3%. Candor on counterpoints: multi-brand operation duplicates some local overhead; FAF’s model captures more margin per direct order in good times; agent brand preference is partly relationship artifact, not technology. On balance, demonstrated margin defense — not management narrative — carries the verdict.

The owned-distribution strategy (stakes in agency consolidators and network-marketing groups) is the most distinctive forward positioning difference versus FAF: an attempt to own a slice of the ~80% of agent-retained economics rather than accept pure wholesale status. It is small today, blurs the underwriter/agent line, and will draw RESPA-adjacent scrutiny, but if executed it converts the industry’s biggest structural leak into partial capture (INTERPRETATION). Commercial title — one of the industry’s largest direct networks, >$100M face-amount capability, coinsurance — is the current cycle’s strong leg, and FNF’s scale and balance sheet win large national accounts.

Verdict (Title): durable but narrow competitive advantage, demonstrably the strongest franchise in the industry. The moat protects share and relative profitability; it does not set price (state-regulated rates; ALTA: title costs down ~8% since 2004), does not capture agent economics, and does not de-link earnings from the mortgage cycle.

4.2 F&G: a defensible niche, not a moat

F&G’s competitive position must be stated plainly. The product is commodity: FIA/MYGA pricing competes on credited rates against Athene, Corebridge, MassMutual and the mutuals; F&G’s own risk factors concede competitors have “greater financial resources and higher financial strength ratings.” There is no demand-side moat. The one genuine differentiator is supply-side and purchased: Blackstone ISG-I Advisors manages ~78% of the $69B portfolio with sub-delegation to Blackstone Credit, Real Estate Debt and Asset-Based Finance, delivering origination (e.g., ~$5B of middle-market corporate lending, 89% investment grade, disclosed near-zero credit losses to date) that F&G could not source alone — for IMA fees, and with single-manager concentration risk. The Fort Greene sidecar (~$1B of Blackstone-backed capital) extends the model off-balance-sheet.

What F&G does have: liability quality (93% surrender-charge-protected account value; non-surrenderable funding agreements/PRT/immediate annuities; matched ALM), capital (FGL Insurance RBC estimated >410% at YE2025 per the 10-K, ~430% per F&G’s own release, vs a 400% target), scale in a niche (~3% of industry flow, historically a top-10 FIA writer — 2025 rank unverified), and a credible capital-light pivot (flow reinsurance > $15B cumulative, opex ratio 60bps → 48bps with a ~45bps target, fee-based earnings 15% → 25% target by YE2028). What it lacks: captive origination (the Athene/Global Atlantic model’s core advantage), pricing power, and a demonstrated through-cycle credit record — 7bp average annual impairments were compiled entirely in a benign credit environment.

Verdict (F&G): weak/narrow advantage — a mid-quality spread lender with one borrowed edge. Competent, Blackstone-armed, well-matched ALM, improving fee mix, solid capital; but commodity liabilities, heavy reinsurance counterparty leverage, and earnings hostage to the rate cycle and LDTI accounting. Economic franchise: mid-quality; low-teens through-cycle ROE at best, and currently ~8% adjusted. Not a Buffett-grade float compounder; not a distressed lender either.

4.3 The consolidated entity: a waypoint, not a destination

Does the combination create value? Modestly, and shrinking (INTERPRETATION). Pro: counter-cyclical earnings mix; capital diversification; FG dividend upstreaming; FNF sponsorship lowered F&G’s early cost of capital (the $250M FG preferred purchase in 2024 is visible support). Con: two regulated insurance balance sheets create consolidated-leverage optics and a ~30% minority-interest deduction that confuses per-share analysis; zero operating synergies exist or are claimed; and the market gets to price ~22% of FNF’s value every day via listed FG. Management’s own actions — two in-specie distributions in three years, taking the stake from ~100% to ~70% — concede the conglomerate is a legacy of the Foley holding-company playbook rather than an industrial logic. The governance overlay is a Foley-network affair: historically accretive deal-making (FIS, Black Knight, F&G itself), genuine conflict-of-interest surface area, and a board independent in form more than in origin (see §7, §9).

5. Growth History and Forward Opportunities

5.1 The historical record: cyclical, not secular

Title’s five-year arc is a cycle, not a trend: revenue $11,497M (2021) → $7,038M (2023) → $8,490M (2025); closed orders 2,169K → 837K → 956K. The 2025 print — premiums +13% to $5,824M, closed orders +9%, fee per file +5.5%, commercial revenue +27% to $1,494M — is recovery off the worst purchase market in decades, not a new run-rate: closed orders are still 44% of the 2021 peak and existing-home sales have sat near 4M units for three consecutive years (FACT — 10-K MD&A; FNF calls). F&G’s arc is real growth: AUM before flow reinsurance from ~$27B AAUM at the June 2020 acquisition to $74.5B (an 18% CAGR since 2019 per the FG Q1-26 call), gross sales $9.6B → $14.6B (2021→2025, record), with $9.0B of 2025 sales in “core” products (FIA $6.7B, IUL, PRT $2.1B) and $5.6B opportunistic (MYGA $3.8B, funding agreements $1.8B).

5.2 Forward lanes, ranked by quality

Lane 1 — Title cyclical volume recovery (high-quality cyclical). The MBA/iEmergent/NAR complex projects 2026 originations +7–15%, refis +11–38%, existing-home sales +4% — every figure hostage to 6.0–6.5% mortgage rates (§3.1). Q1-26 delivered the first hard evidence: opened orders +13%, closed +16%, refi orders opened at 2,000/day as rates touched the low-6s (moderating to 1,600/day in April), purchase orders +2% (April +4%). Q1-26 also showed the arithmetic caveat: refi at 37% of closed orders pulled fee per file down 3% and the adjusted margin to 13.1% from FY25’s 15.9% — volume returns on dilutive mix first. The upside case is volume: at 956K closed orders vs a 2,169K peak and a plausibly mid-cycle ~1.2–1.4M, title revenue has $2–3B of headroom before any pricing. Management guides to a 15–20% annual adjusted margin range “even if total residential volumes remain at current levels over the near term” (Q1-26 call) and delivered 15.9% through a 30-year-low purchase market (FACT; the margin range is management’s framework, not ours).

Lane 2 — Commercial title expansion (best structural lane, lumpy). Commercial revenue $1,494M in FY2025 (+27%, third-best year ever), Q1-26 +15% with national accounts +22%; deal inventory diversified across industrial, data centers, multifamily, affordable housing, retail, energy (Q1-26 call); MBA forecasts +24% CREF volumes in 2026. Commercial is less rate-sensitive, carries higher fees, and FNF’s scale wins national accounts — but it flatters margins near peaks and mean-reverts when CRE finance slows; 18% of title revenue is a cyclical-high mix (INTERPRETATION). The October 2024 acquisition of First Nationwide Title Agency’s commercial operations from AmTrust (terms undisclosed, immaterial — a Northeast talent/book bolt-on) and management’s signal that title-agent M&A will be “more active this year and next” (Q1-26 call, Park) extend this lane.

Lane 3 — F&G’s fee-based pivot (the one genuinely secular element, unproven). The strategy: shift from balance-sheet spread lending to fee-based, capital-light manufacturing — Fort Greene sidecar (“$1 billion of on-demand third-party capital”), flow reinsurance (> $15B cumulative new business reinsured; $928M of Q1-26 sales ceded), owned distribution (~$700M deployed, ~$80M annual EBITDA, now in a formal strategic-alternatives process at Peak Altitude), opex ratio 60bps → 48bps toward ~45bps by YE2027, fee-based share of adjusted earnings 15% → 25% by YE2028 (FACT — FG Q1-26 call/supplement). If executed, this raises ROE quality and reduces the capital intensity that currently forces dilutive equity raises. Today it is small: owned-distribution revenue $89M against $5.7B of segment revenue. Proof points to watch: net flows (~71% retention of gross sales in Q1-26 — healthy), new-money spreads as the Fed cuts, and the Peak process outcome.

Lane 4 — Technology/AI (cost-side, back-loaded). Over half the title workforce uses AI tools regularly; named platforms include SoftPro and inHere (~80% of residential sale transactions engaged, ~2.8M unique users in 2025); management frames “smaller wins… then bigger wins as we move into 2027” and hints the 15–20% margin range could migrate toward the middle if volume forecasts materialize (FACT — Q4-25/Q1-26 calls). Directionally margin-accretive; currently immaterial in the numbers (INTERPRETATION).

Verdict (Growth): ADEQUATE — two good cyclical positions with credible share/scale logic, no secular growth story. Treat 2025–26 title gains as mid-cycle recovery, not run-rate; treat F&G’s asset growth as quantity-over-quality until the fee pivot proves it can lift ROE above the cost of equity; Corporate ventures are immaterial optionality. Better than a melting ice cube; well short of a compounder.

6. Financial Quality

6.1 The two-engine framework: why consolidated GAAP EPS is nearly uninterpretable

FNF’s consolidated GAAP diluted EPS ran $8.44 → $4.10 → $1.91 → $4.65 → $2.21 across 2021–25, with TTM at $2.81. Almost none of that path is economics. The distorters (FACT — 10-K MD&A, releases): F&G’s LDTI fair-value machinery (market-risk-benefit remeasurement of $95M/−$25M/$167M across 2023–25, indexed-annuity and reinsurance embedded derivatives), the “recognized gains and losses” line (+$334M in 2021, −$1,493M in 2022 on rate-shock derivative and bond marks, −$164M, +$83M, −$60M), and in 2025 a $471M non-cash deferred tax charge — the outside-basis DTL crystallized when the December 2025 F&G distribution dropped FNF below 80% ownership (FY2025 effective tax rate 53.9%; ~$1.73/share; Q4-25 alone was a GAAP net loss of −$117M/−$0.43 against adjusted earnings of +$382M/$1.41). The operative lens is the company’s adjusted (non-GAAP) series:

(FACT — 10-K consolidated statements; releases for adjusted. The adjusted framework is management-defined.) The FY2025 adjusted build: Title $1,054M + F&G at FNF share $412M + Corporate $3M − eliminations $117M = $1,352M ($4.97). Q1-26: $197M + $80M + $0 − $28M = $249M ($0.93). The minority-interest step-up is now first-order: NCI took $78M of Q1-26 consolidated net earnings ($74M of it F&G) vs ~$0 in Q1-25, because the December 2025 distribution cut the F&G stake to ~70% — per-FNF-share F&G economics nonetheless improved (adjusted F&G contribution at FNF share was $80M in both Q1-26 at 70% and Q1-25 at ~84%).

One unusual QoE note cuts against the usual adjusted-EPS skepticism (INTERPRETATION): F&G’s alt-investment income has run persistently below management’s long-term expected return inside adjusted earnings — $123M ($0.45/sh FNF share) below in 2024, $228M ($0.84/sh) below in 2025, $31M below in Q1-26. If alts merely hit expectation, TTM adjusted EPS would be ~$5.4–5.6, not $5.12 — the adjustments currently penalize rather than flatter. The reverse risk is symmetric: if the underperformance is structural (“it has been five years since we have seen meaningful realizations” — Blunt, FG Q1-26 call), the normalization is a mirage.

6.2 Title engine: high-quality cyclical economics

(FACT — 10-K MD&A, 10-Q, quarterly operating statistics in releases.) Three quality observations. First, provisioning is conservative and mechanical: a flat 4.5% of title premiums every year 2021–25 and Q1-26, not trended to experience — actual FY2025 claims paid of $269M vs a $262M provision; reserve for title claim losses stable at ~$1.70–1.71B. Contrast FAF’s ~3.0% with releases (FAF’s 2025 EPS included a $39.8M release; FNF’s has none — per FAF’s 10-K and the companion First American analysis). FNF’s approach is the steadier, more conservative QoE signal: no release-driven EPS flattery in the recovery. Second, margin quality is genuinely improved: the 15.9% FY2025 adjusted pretax margin was earned at 44% of peak 2021 order volume — only ~1pt below the 16.7% earned in 2022 at 28% higher volumes — evidence of structural cost takeout plus commercial strength, not volume flattery. Third, the mix caveat is arithmetic, not hypothetical: Q1-26 refi at 37% of closed orders (vs 25% a year ago) pulled fee per file down 3% and the adjusted margin to 13.1%; residential fee per file was flat year-over-year — the decline is mix, not price-cutting (Q1-26 call). Division-level margins (Q1-26, CFO Park): direct ~20%, agency ~7% of gross, national commercial 27%, ServiceLink centralized refi/default 23%, subservicing ~20%, home warranty 16%.

6.3 F&G engine: spread economics, asset quality, and the reinsurance stack

(FACT — F&G releases and Q1-26 supplement; FNF 10-K segment MD&A.) The economic read (INTERPRETATION): F&G earns a ~1.9% product margin on $57B of retained AUM, converting to ~0.87% adjusted ROA and ~8.2% adjusted ROE ex-AOCI — adequate, below best-in-class annuity peers, and below any reasonable cost of equity for a levered spread lender (the market’s 0.70x ex-AOCI price-to-book on FG solves the same equation). Q1-26 showed product margin compression at the margin: cost of funds 3.20% → 3.42% annualized as crediting rates reset, product margin 1.88% → 1.66% annualized; fixed-income yield −16bps to 4.77%. Management attributes ~10bps of the yield decline to timing/one-offs and claims core spread held flat with Q4 — but this is one annualized quarter, and the trajectory needs the Q2-26 supplement to confirm (open item).

The asset book is designed-aggressive-but-so-far-clean (FACT — 10-K FY2025 Note D; FG supplement): $69.0B of investments — corporates ~$24B (37%), CLOs/loan-backed private obligations $10.9B (16%, up from 9%, partly a NAIC bond-definition reclassification), ABS $7.8B (11%), CMBS $5.2B (8%), non-agency RMBS $2.6B (4%), limited partnerships $4.6B (6%), commercial mortgage loans $3.0B (4%; 57% weighted LTV, 2.3x DSC, one delinquent loan), residential mortgage loans $4.4B (6%). NAIC 1: 62%; NAIC 2: 34%; below-investment-grade ~4%. Structured/private credit is roughly half the portfolio — this is the Blackstone-originated spread model. Credit-related impairments averaged 7bps/year over three years, below pricing assumptions — a pristine record compiled entirely in a benign credit environment. Gross unrealized losses on AFS of $3.18B are rate-driven. The alt sleeve was redefined effective 2026-01-01 (~$6B of debt-like assets moved to fixed income; alts now ~$4B, ~7% of the retained portfolio) with the LT expected return revised from 10% to 12–14%; Q1-26 annualized alt return 8.3% — still below expectation, and hitting expectation would add +34bps of ROA / +3.4pts of ROE (FG Q1-26 call).

The reinsurance counterparty stack is material and growing. Reinsurance recoverables of $17.6B at YE2025 rose to $20.0B at Q1-26, with funds-withheld reinsurance liabilities of $14.2B → $16.5B. Top-5 recoverables: Aspida ~$8.5B, Somerset ~$5.6B, Everlake ~$1.9B, Ancient Re ~$1.7B, Wilton ~$1.0B — collateralized via secured trusts/funds-withheld/LOCs, but a real counterparty concentration that grows with every flow-reinsurance cession (FACT — 10-K/10-Q balance sheets; FG supplement). Capital: RBC estimated >410% (10-K) / ~430% (F&G release) at YE2025 vs a 400% target — but FGL Insurance’s maximum ordinary dividend capacity for 2026 is $0 under the Iowa earned-profits test (no dividends paid to the FG holdco in 2025); FG’s $137M of 2025 common+preferred dividends were funded above the opco. Surrenders/withdrawals/death ran ~16% of beginning account balances in FY2025 (~15.6% annualized Q1-26) — the 93% surrender-charge protection is doing its job so far.

6.4 Balance sheet: two regulated insurers, one parent

(FACT — 10-K/10-Q balance sheets. NCI rose in 2025 on F&G’s equity offering and the 12% distribution reclass.) The consolidated balance sheet is ~92% insurance float and liabilities; the analytically meaningful view is the parent: ~$2,150M of senior notes at cheap fixed coupons (4.50% $450M due 2028, 3.40% $650M due 2030, 2.45% $600M due 2031, 3.20% $450M due 2051), an $800M revolver undrawn (matures Feb 2029), holdco cash and liquid investments of $659M at YE2025 declining to $495M at Q1-26, and ~$437M of 2026 ordinary dividend capacity from the title insurers without regulatory approval (FACT — 10-K Note F; Q1-26 release). F&G’s debt is standalone and explicitly non-recourse to FNF: $2,270M principal (7.40% $500M 2028; 6.50% $550M 2029; 6.25% $500M 2034; 7.95% $345M 2053; 7.30% $375M junior subordinated 2065), a $750M revolver undrawn, debt/capitalization ex-AOCI 26.2%, ~$165M of annual interest at a ~7% blended coupon — against a ~5.1% asset yield, an uncomfortable funding-cost vs asset-yield juxtaposition (INTERPRETATION). Parent fixed-charge coverage is comfortable: holdco cash + title dividend capacity + ~70% of FG’s ~$120M common dividend + $17M FG preferred ≈ $1.03B available against a $556M run-rate common dividend and ~$70M of parent interest — with the caveat that F&G’s $0 opco dividend capacity in 2026 means the FG leg currently contributes no new upstream cash.

Verdict (Financial Quality): Title — high-quality cyclical franchise economics: conservative mechanical provisioning, demonstrated trough-margin defense, cash-generative with minimal reinvestment; quality flags are mix-driven fee-per-file inflation and arithmetic margin dilution as refi returns. F&G — a leveraged spread lender earning below-COE returns on an untested credit book, with sound headline capital but zero opco dividend capacity in 2026 and a building reinsurance counterparty stack; GAAP earnings structurally noisy, the adjusted framework reasonable but embedding an alt-return normalization that has persistently cut against reported results. Consolidated — current EPS is cyclically LOW, not high: TTM adjusted ~$5.12 is earned at trough-to-early-recovery title volumes and a still-building F&G book; GAAP TTM $2.81 is meaningless. And per-share math must run on FNF-attributable earnings at the ~70% F&G share — the NCI is first-order, not rounding.

7. Capital Allocation

7.1 The deployment record FY2019–2025

(FACT — 10-K cash-flow statements, dividend notes, DEF 14A CD&A, 8-Ks. Five-year 2021–25 totals per the company: ~$2.5B dividends + ~$1.3B buybacks.) The deployment priority is unambiguous and consistent (INTERPRETATION): (1) fund the growing dividend, (2) fund F&G growth (2020–24) then un-fund it via separation (2022– ), (3) buybacks strictly as the residual swing factor — $1.0B deployed in 2021–22, ~zero in 2023–24 when title cash fell, $252M resumed in 2025. The dividend is the sacred cow; the buyback is the shock absorber. That is textbook cyclical-industry capital discipline.

7.2 The dividend: the one unambiguous strength

DPS has been raised every year of the window: ~$1.20 (2019) → $1.35 → $1.56 → $1.77 → $1.83 → $1.94 → $2.02 (2025) → $2.08 annualized ($0.52/qtr declared February 2026, ~4.0% yield at $52). The trough test is the proof, not the projection: in 2023, through the worst purchase market in decades, the dividend was raised ($1.77 → $1.83) at a 52% payout of adjusted ANE ($498M vs $962M) and 66% of title-only adjusted earnings, while buybacks were cut to $4M (FACT). Coverage today: holdco cash $495M + ~$437M title dividend capacity + ~$84M parent share of FG common dividends + $17M FG preferred ≈ $1,030M against a $556M run-rate dividend and ~$70M parent interest — sustainable through a repeat of 2023 even with zero F&G upstreaming, with roughly two years of coverage from title capacity plus holdco cash alone (FACT inputs; INTERPRETATION — the policy reads as a deliberate covenant with the income investor base, and the ~4% yield is a valuation floor management actively defends).

7.3 The F&G round trip: value created, modestly — and the tax criticism

In: ~$2.7B for 100% of F&G in June 2020 ($12.50/share; ~$1.8B cash + ~24M FNF shares; annuity stocks depressed pre-rate-hike — FGLH had repurchased its own shares at ~$8 in 2019) + $250M FG preferred (Jan 2024, 6.8%) + $151M into the March 2025 equity offering ≈ ~$3.1B. Out/held: the ~70% stake worth $3.04B at market + ~$850M distributed in specie at carrying value ($301M Dec 2022 + $551M Dec 2025) + the $250M preferred + roughly $350–400M of cumulative FG dividends to the parent (inferred from the NCI dividend line — labeled inference) ≈ ~$4.4B ≈ 1.4–1.5x MOIC over six years, a high-single-digit IRR (INTERPRETATION — arithmetic on verified inputs). Value-creating but unspectacular; the cheap 2020 entry and the 2022–24 rate repricing did the work. The asset now being distributed earns ~8% adjusted ROE and trades at 0.70x book ex-AOCI — the market’s own verdict that F&G’s retained-earnings growth is worth less than book — so distributing it is the right exit regardless of tax friction.

The separation structure deserves genuine criticism (INTERPRETATION, from disclosed facts; tax mechanics inferential). Both distributions were taxable dividends. A §355 tax-free spin was plausibly unavailable in December 2022 (control acquired in a taxable transaction within five years) — but the five-year window ran in June 2025, and by dripping out 12% in December 2025 instead of spinning the whole ~84%, FNF dropped below 80% and (a) crystallized $471M of deferred tax and (b) foreclosed a future tax-free spin of the remainder, since FNF no longer holds the 80% “control” §355 requires. Management has been remarkably candid: “yes, we could have spun the entire company to FNF shareholders tax-free. Clearly, we didn’t do that. And by dropping below 80%, that option is off the table” (Park, Q3-25 call — FACT, verbatim). The counterpoints are real: the drips created float, liquidity, index inclusion (S&P SmallCap 600, May 2026), and a daily SOTP reference that has mechanically arbitraged away the holdco discount; and full immediate separation would have ended FG dividend upstreaming and the counter-cyclical earnings mix. Meanwhile the glide path is not even linear — management openly contemplates FG’s own buybacks pushing FNF’s stake back up: “if that means that we close that gap from 70% to 80%, or wherever it might be, that is very possible. But again, there is no target” (Park, Q1-26 call). The honest read: FNF wants FG’s market value to perform more than it wants to exit; full separation is off the table near-term and structurally taxable in any case (INTERPRETATION).

7.4 Buybacks, and F&G’s internal capital allocation

Buyback timing is poor-to-mixed ex post (FACT prices; INTERPRETATION judgment): 2021’s $461M @ $45.22 preceded the 2022 slide to ~$28; the $549M @ $41.05 in 2022 was well-timed; zero buybacks in 2023–24 with the stock mostly $30–50 was the cheapest window missed; 2025’s $252M @ $56.80 looks expensive against the Q2-26 dip to ~$45 and the current ~$52. Cumulative mark-to-market on the $1.27B of 2021–25 repurchases is roughly flat — buybacks have been a return of capital, not per-share value creation beyond the ~6% share-count reduction (287M → 269M diluted).

Inside F&G, the capital-allocation problem is sharper: AUM is growing ~12%/yr at 1.88% product margin compressing to 1.66%, producing 8.2–8.4% adjusted ROE ex-AOCI — almost certainly below F&G’s ~10–11% cost of equity (FG’s 0.70x ex-AOCI price-to-book is the market solving the same equation). Growing AUM at below-COE returns destroys per-share value at the unit level, and the March 2025 equity raise at ~0.7x book to fund that growth is the single most value-dilutive capital action in the window — one FNF itself joined for $151M (INTERPRETATION). F&G’s own $100M buyback program and Q1-26 repurchases at $24.14 (“a real significant, almost a silly discount” — Park) are the rational response to the same math.

7.5 Incentives and governance: the alignment discount is real

Comp structure (FACT — DEF 14A filed 2026-04-29): the annual cash incentive for all NEOs is 75% adjusted pretax title margin + 25% adjusted title revenue; the same title-margin metric is the sole performance-vesting condition on performance restricted stock. F&G is in no metric — despite contributing 30% of 2025 consolidated adjusted net earnings (the company’s own CD&A figure) and ~22% of market value. Worse, the 2025 targets were set below 2024 actuals (revenue target $7.0B vs 2024 actual $7.7B, −9%; margin target 13.0% vs actual 15.1%, −210bp) and then paid at the 200% cap — the second consecutive year at maximum. The committee’s conservative-forecast rationale is disclosed and plausible (April 2025’s macro setup was grim), and margin did land above 2024’s actual — but a plan that pays maximum on targets set under prior-year results transfers macro-volatility risk from executives to shareholders (INTERPRETATION). The defense — line-of-sight: title is what FNF executives control, F&G has its own listed-co comp system — is legitimate but incomplete: no metric anywhere addresses the holdco discount, the FG distribution glide path, or F&G’s below-COE growth, all first-order to FNF shareholder value right now.

Foley’s package is the largest governance-cost datapoint: an 830,152-share restricted grant in November 2024 (~$50M derived grant-date value at ~$60 — labeled estimate; the Form 4 reports no price) to a non-executive chairman already holding 10.0M shares (3.7%, ~$500M), vesting 25% immediately, atop a Director Services Agreement with full-vesting-on-exit protection and $253K of personal aircraft use in 2025 (FACT — DEF 14A; Form 4). The related-party lattice is dense: three “independent” directors hold co-investments in Foley’s sports vehicles (BKSE/Vegas Golden Knights, Black Knight Football Club/AFC Bournemouth); Ammerman chairs Cannae (Foley was Chairman/CEO/CIO until May 2025, remains Vice Chairman); Ammerman and Rood sit on F&G’s board; Gravelle is GC of both FNF and F&G; a Dayforce payroll relationship; the Ayers v. Foley derivative suit over 2022–24 director compensation (pending; MTD hearing March 2026, outcome unknown); the settled City of Miami F&G-acquisition suit; and the June 2025 Nevada redomestication, which shareholders approved but which weakens the shareholder-litigation forum. The market noticed: Foley was re-elected in June 2026 with 21.5% of votes withheld — the highest of his class — even as say-on-pay passed at 95.3% and board declassification (full annual elections by 2029) passed near-unanimously (FACT — 8-K 2026-06-11). Institutional holders are targeting Foley personally, not the pay program. Counterweights on the record: Cannae below 5%, declassification underway, 5.7% insider ownership, formal ownership/retention guidelines with teeth.

Insider transactions corroborate neutral-to-mildly-negative alignment (FACT — 95 Form 3/4/4A/5 filings parsed): zero open-market purchases of FNF common by any officer or director in 24 months — including the Q2-2026 dip to ~$43 — while the company itself was buying ($252M in 2025, $82M in Q1-26). Discretionary sales are small (~145K shares total; largest: CLO Sadowski, 69,196 shares @ $45.70 = $3.16M on 2026-06-26, near the lows, not flagged 10b5-1; director Dhanidina a serial small seller — his 2026-06-24 Form 4 showing $418.25/share is a probable filer error with no correcting amendment and is not cited as a real price). Grants over the window (~1.4M shares) dwarf sales. Nobody at FNF buys the stock with their own money; they are paid in it (INTERPRETATION).

Verdict — Capital Allocation grade: B−. Deployment record genuinely good: dividend raised every year including the 2023 trough at 52% payout; buybacks used correctly as the swing factor; F&G bought cheap and grown 2.7x in AUM; separation glide path mechanically shrinking the holdco discount with ~$850M distributed in specie. What holds the grade at B−: buyback timing destroyed as much as it created; F&G’s internal growth runs below its cost of equity, funded partly by a dilutive 2025 equity raise FNF joined; the separation chose taxable drips over a plausibly available tax-free spin, crystallizing $471M of deferred tax and burning the §355 option; and incentives pay on title-only metrics with below-prior-year targets at 200% payout plus a $50M special grant to a non-executive chairman. Capital deployment: B/B+. Incentive alignment: C−. Net: B−.

8. Changes and Headwinds — Last Two Years

A dated timeline of the material changes, July 2024 – July 2026 (FACT — 8-K filings, 10-Q, DEF 14A, PR Newswire originals, earnings-call transcripts; filing-level facts cross-validated across sources):

• 2024-10-02 — FNTA commercial acquisition. FNF acquired the commercial operations of First Nationwide Title Agency from AmTrust; NYC-based commercial title agency; leadership team joined; terms undisclosed; no 8-K (immaterial talent/book bolt-on in Northeast commercial title — resolved open item: real but small).
• 2025-01-13 — F&G hybrid debt. $375M 7.300% junior subordinated notes due 2065 — part of F&G’s standalone, explicitly non-recourse debt build-out ($2.27B raised 2023–25).
• 2025-03-24 — F&G equity offering. 8.0M shares @ $33.60 (~$269M gross); FNF purchased 4.5M of them (~$151M) to limit dilution — a parent-support action that also made FNF a buyer of below-COE growth at ~0.7x book (§7.4).
• 2025-06-11/12 — Nevada redomestication approved. Delaware → Nevada, part of the 2025 wave of DE→NV exits; reduces Delaware Chancery exposure — relevant context: two Delaware derivative suits in the window (INTERPRETATION).
• 2025-06-10 — Ayers v. Foley filed. Stockholder derivative suit (Del. Ch., C.A. No. 2025-0650-LWW) against non-employee directors alleging breach of fiduciary duty re: 2022–24 director compensation; motion-to-dismiss hearing rescheduled to 2026-03-09; outcome not found in filings or news as of today (open item). FNF: “does not believe the lawsuit will have a material impact.”
• 2025-11-06/07 — Q3-25 beat + second F&G distribution approved. Adjusted pretax title earnings $410M (+27%), adjusted title margin 17.8%; same morning the board approved the second special distribution of ~16M FG shares (~12%); FNF raised the quarterly dividend 4% to $0.52; F&G raised its dividend 14% to $0.25.
• 2025-12-31 — Second distribution completed. 16,280,204 FG shares (6 per 100 FNF); stake ~84% → ~70%; FG public float roughly doubled; taxable dividend; dropping below 80% triggered the $471M non-cash deferred-tax charge (Q4-25 GAAP net loss −$117M vs adjusted +$382M/$1.41).
• 2026-02-19/20 — Record FY2025 print. Title adjusted pretax $1.4B / 15.9% margin; commercial revenue ~$1.5B (third-best ever); F&G record gross sales $14.6B, AUM before flow reinsurance $73.1B (+12%).
• 2026-03-01 — F&G Life Re sold. Bermuda reinsurer sold to Ancient Financial Holdings for ~$102M cash + 19.9% LP interest; +$14M pretax gain; deconsolidated (renamed Ancient Re); ~$900M of statutory liabilities recaptured and a $200M asset dividend taken at YE2025 ahead of the sale; Blackstone retained asset management of the in-force FIA assets; new forward-flow MYGA reinsurance with Ancient. A deliberate reduction of regulatory surface area (§3.2).
• 2026-03-16 — FG buyback program. New three-year $100M authorization; FG repurchased ~1.2M shares for $29M @ $24.14 in Q1-26 under the prior program — buying its own stock at roughly half ex-AOCI book.
• 2026-05-06/07 — Q1-26 title inflection. Title revenue $2,004M (+13%); adjusted pretax $268M (+27%); adjusted margin 13.1% (+140bps); opened orders +13%, closed +16%; refi mix 33% of opened / 37% of closed; commercial +15%. Consolidated adjusted EPS $0.93 vs $0.78 — framed by Zacks as a “miss” vs $1.10 consensus, which management attributed to analysts mis-normalizing F&G alt income; treat sell-side estimate comparisons as structurally unreliable while alts underperform (INTERPRETATION).
• 2026-05-12 — Nolan employment agreement. First A&R agreement with the CEO: 3-year term, base $1.1M, incentive target 200% of base; concurrent 39,542-share grant. Removes FNF CEO succession risk.
• 2026-05-13 — S&P SmallCap 600 inclusion for FG (effective 2026-05-18) — a direct consequence of the enlarged post-distribution float.
• 2026-06-10/11 — First Nevada annual meeting. Board declassification approved (phased, full annual elections by 2029); Foley re-elected with ~21.5% withheld — highest of the class (§7.5).
• 2026-06-16 — F&G leadership transition. Chris Blunt retires as F&G CEO to run Peak Altitude Equity (remains an F&G director); Conor Murphy (President & CFO) named CEO; Michael Bailey CFO effective 2026-08-03. Not telegraphed on the May 7 call. Read with the Peak Altitude strategic-alternatives process (~$700M deployed, ~$80M EBITDA; deconsolidation/partner/sale; update promised “the next couple of quarters”) — the owned-distribution value-unlock is now the former CEO’s full-time project (INTERPRETATION).

Do the changes strengthen or weaken the thesis? Net strengthen, with two governance asterisks (INTERPRETATION). The operating changes are all in the thesis direction: title margins delivered at the top of guidance through a 30-year-low purchase market and now inflecting on refi + commercial; F&G’s capital-light pivot executing (sidecar, flow reinsurance, opex 60→48bps, F&G Life Re exit); capital return accelerated (~$800M in 2025, ~$500M in-specie distribution). The genuine fundamental drags: F&G spread compression (product margin 1.88% → 1.66% annualized, one quarter so far) and five years of alt underperformance. The governance asterisks: Nevada redomestication plus a director-comp derivative suit plus 21.5% withheld on Foley, against the simultaneous declassification and FG insider-buying positives (CEO Blunt bought 5,000 FG shares @ $29.56 in January 2026). The two-year story in one line: FNF used a cyclical title trough to raise margins, return ~$1.3B, and convert its F&G stake into a listed, index-included, 70%-owned asset — while governance normalized a step behind the capital actions.

9. Risk Analysis (Risk Matrix)

Catastrophic-loss assessment: a true wipeout is unlikely — two regulated insurers with 93% surrender-protected liabilities, ~$1.0B of parent liquidity plus ~$437M annual title dividend capacity, cheap fixed-rate parent debt (2.45–4.50%, maturities 2028–2051), and an undrawn revolver. The realistic bear path is a double compression — title re-trough plus F&G spread/credit stress — that breaks the counter-cyclical smoothing narrative precisely when it is needed; the two engines’ correlation in a hard landing is untested (INTERPRETATION). The 2008-era legacy matters for calibration: lifetime max drawdown −75% predates the window — read “low beta” (0.56) as low market sensitivity, not low tail risk.

10. Valuation Discussion (Embedded Expectations)

No price target; no recommendation. This section asks what the current price must assume. SOTP multiples and scenario inputs are ASSUMPTION-based and labeled as such; market data FACT (ROIC, 2026-07-17 close).

10.1 The multiple stack

At $52.04 (269M shares; market cap ~$14.0B): P/E TTM GAAP 18.5x — depressed denominator, not expensive price: the $471M deferred-tax charge and LDTI marks crushed TTM GAAP EPS to $2.81, and AZI’s 93.6th-percentile GAAP P/E ranking is the same artifact and should not be cited as evidence of expensiveness. The operative figures: P/E TTM adjusted ~10.2x ($52.04 / ~$5.12, company-defined non-GAAP); P/B GAAP 1.93x (BVPS $26.97); P/B ex-AOCI 1.53x (BVPS ex-AOCI $34.01); dividend yield ~4.0% on the $2.08 run-rate. ROIC’s enterprise values for FNF (−$10.6B) and FG (−$23.2B) are negative and economically meaningless (insurer float counted against insurance liabilities) — EV multiples are not used anywhere here; adjusted P/E, ex-AOCI P/B, yield, and SOTP are the appropriate tools for a two-engine insurer.

10.2 SOTP — the mandatory frame, and its central fact

(FACT inputs: Title TTM adjusted after-tax contribution ~$1,078M = $1,416M TTM adjusted pretax × (1 − 24%); FG stake $3.04B = 134M shares × ~70% × $32.405 — coincidentally ≈ FNF’s share of FG GAAP book, since FG trades at 0.99x GAAP book / 0.70x ex-AOCI book; holdco net debt per 10-K Note F / Q1-26 release. The multiples are judgment.)

The section’s central fact (INTERPRETATION): at $52.04 the market prices FNF ~4% above base SOTP — there is NO holding-company discount. Inverting: market cap $14.0B + holdco net debt $1.66B − FG stake $3.04B implies the market values Title standalone at ~$12.6B ≈ 11.7x TTM adjusted Title earnings — a token premium to FAF’s 11.2x, and FAF’s E is recovery-flattered while FNF Title’s E is cyclically low on volume. The classic “FNF trades at a conglomerate discount to its listed FG stake” argument has no cushion at the current price: the stake is fully marked, the two taxable distributions did their job, and the market already pays a FAF-or-better multiple for the Title stub. What remains is not a discount but a debate about the right multiple on each part.

10.3 Comps

(FACT — ROIC, 2026-07-17; FAF row per the companion First American analysis; annuity multiples on 3/31 quarter-end pricing, directionally valid. AEL excluded — taken private by Brookfield May 2024; Athene private inside Apollo. The listed annuity comp set is thin.) FAF is the only true title pure-play; ORI mixes P&C; STC carries real-estate services. None is clean — which is precisely why the SOTP, not the consolidated multiple, is the right frame (INTERPRETATION). Does FNF Title deserve a premium to FAF’s 11.2x? The evidence says yes, moderately: margin leadership every year 2021–25 (~1,000bp at the trough), #1 share at ~32% vs ~23%, and a conservative 4.5% provision with no release flattery vs FAF’s 3.0%-with-releases. Through-cycle quality argues 12–13x; the counterweight is that FNF’s agency-heavy mix caps peak economics and F&G’s noise comes attached to the equity. The market concedes ~11.7x — a token premium (INTERPRETATION). On the annuity side, the listed complex trades at ~0.7–1.3x book on LDTI-noisy GAAP earnings; FG’s 0.70x ex-AOCI book at 8.4% adjusted ROE is consistent with the sector’s pricing of ~8–9% ROE spread lenders — no comp says FG is mispriced in either direction with conviction.

10.4 Normalized earnings power (scenario framework)

Critical framing (INTERPRETATION): current TTM adjusted EPS ~$5.12 is cyclically LOW on title volume (956K orders = 44% of peak) but cyclically HIGH on title margin (15.9% adjusted, near-peak in absolute pretax on ~$3B less revenue) — and Q1-26’s refi mix is already diluting margin as volume returns. Normalization therefore runs in both directions at once on Title: volume up, margin down — roughly offsetting. The base case is not dramatically above the current print. (ASSUMPTION-based scenarios; 272M shares, 24% tax, −$117M Corporate/eliminations.)

Bear mechanics: margin →12% on stalled volumes (refi dilution without purchase recovery; commercial mean-reverts) AND F&G product margin compresses — a double compression, and note the correlation risk: a hard landing hits title volumes and F&G credit/surrenders together, breaking the smoothing story. Base mechanics: the MBA +6–8% 2026 recovery persists, volumes reach ~1.3M by 2027, margin dilutes to ~14%, commercial holds, F&G grows AUM high-single-digit at ~1.6–1.7% product margin — EPS ~$5.5–6.0, only modestly above the current $5.12 TTM because margin give-back absorbs the volume recovery. Bull mechanics: purchase market genuinely reopens (rates <6%), margin holds ~16% on ~$12B revenue, F&G sustains ~1.8% product margin with the fee pivot adding non-spread earnings. Quality flags (FACT→INTERPRETATION): alt income ran $0.84/sh below LT expectation inside FY25 adjusted earnings — if alts merely hit expectation, TTM adjusted is ~$5.4–5.6; if the underperformance is structural, base-case F&G is too high. No release tailwind is embedded in any scenario (FNF’s 4.5% provision, unlike FAF). Commercial at 18% of title revenue is a cyclical high; bear assumes mean reversion, base assumes hold.

10.5 What the price embeds

At $52.04 the market underwrites, reading across three independent frames (INTERPRETATION):

1. SOTP frame: Title at ~11.7x adjusted earnings (FAF’s multiple plus a token premium), FG fully marked at market, holdco net debt fully deducted → price ≈ SOTP + ~4%. The market prices the structure correctly and the FG stake fully. There is no conglomerate discount to harvest.
2. Earnings frame: ~10.2x TTM adjusted EPS vs FAF at ~11.2x on flattered earnings — the market pays less per dollar of FNF’s cyclically-low-volume earnings than per dollar of FAF’s cyclically-recovered earnings. Embedded: partial title normalization (volume recovery) but not full credit, and no re-rating of the annuity engine.
3. Justified-P/B frame: P/B ex-AOCI 1.53x with r≈9%, g≈3% implies a durable ROE of ~12% on the $34.01 ex-AOCI book ≈ ~$4.1 of “permanent” EPS — ~$1.0/sh below the current adjusted $5.12 print. The market is explicitly fading part of current earnings: it treats a slice of the near-peak title margin (and/or F&G’s spread level) as non-permanent and credits only part of the volume recovery. Cross-check: the ~4.0% dividend yield plus ~0.5–0.6%/yr buyback (~4.5% shareholder yield) does real work supporting the multiple.

Net: the embedded expectation is a soft-landing-with-fade — title margin gives back ~1.5–2pts as refi returns, F&G spreads drift down, volumes recover only partially, EPS settles ~$4.5–5.5 as the durable level — between our bear and base. The market is NOT pricing a 2021-style peak, a deep trough, an FG re-rate, or a separation catalyst. The single most fragile assumption in the whole valuation, both directions, is the mid-cycle title margin: bear 12%, market ~13% implied, base 14%, bull 16% — a ±2pt error on ~$10B of title revenue is ±$150M+ of after-tax earnings, ±$0.55/share (INTERPRETATION).

Priced correctly (evidence-supported): the two-engine SOTP with FG fully marked; the GAAP noise (informed capital is right to look through the DTL/LDTI artifacts — the adjusted ~10.2x is the operative number); some fade of current title margin (refi-mix dilution is arithmetic); FG at ~0.7x ex-AOCI book while ROE ex-AOCI is 8.2–8.4% and opco dividend capacity is $0 — the discount to book is earned, not an oversight. Priced incorrectly / debatable: the bull variant — the market concedes only ~11.7x for a Title franchise that out-margined FAF by ~1,000bp at the trough and assigns zero value to alt-income normalization (+$0.84/sh if alts hit LT expectation) and to an FG fee-pivot re-rate; its load-bearing assumption is that title volumes normalize toward ~1.3M orders while adjusted margin holds ≥13.5–14%, and its fragility is that the whole path rests on mortgage rates. The bear variant — a double-compression scenario in which F&G’s 7bp impairment record does not survive a real credit cycle and the reinsurance stack transmits the stress; its fragility is three years of clean credit and 93% surrender protection, which argue the tail is fatter in narrative than in near-term cash flows.

11. Variant Perception

Consensus. Formal sell-side consensus EPS was not pulled (no access; flagged). What the news feed shows instead: the Q1-26 print was framed by Zacks as a “miss” ($0.93 vs $1.10 consensus) — a comparison management attributed to analysts mis-normalizing F&G alt income, and one this memo treats as structurally unreliable while alts underperform (§8). Retail/blog sentiment (Seeking Alpha, four pieces Feb–Jun 2026) is uniformly bullish — “top pick for 2026,” “30% discount to intrinsic value,” upgrades after the 25% decline — opinion sources cited for sentiment only. The tape’s own positioning is more informative than the blogs: the factor data shows FNF held as a low-beta (0.56), high-yield (~4%), Value/LowVol-tilted financial with negative momentum loading and no quality tilt — factor-similar to FAF, STC, ACT, CNO, AGO, MET and high-dividend financial ETFs. There is no momentum crowd in this stock; the marginal holder is an income/value investor being paid ~4% to wait. The regime is mildly favorable (Value +1.70z, DividendYield +1.66z on the 1-year; high beta penalized), with rate direction the live headwind (FNF benefits when yields fall). Short-interest data was not pulled for FNF; FG short interest reportedly rose +21% in February 2026 to 1.2% of shares (news feed, unverified).

Bull variant (argued): the market is paying ~11.7x for a Title franchise that demonstrated ~1,000bp of trough-margin superiority over the peer whose own multiple is 11.2x on flattered earnings — a token premium for demonstrably superior through-cycle economics, conservative 4.5% provisioning, #1 share at ~32%, and the industry’s best commercial network hitting its strongest cycle leg (+24% CREF forecast). Volume is the coiled spring: 956K orders = 44% of peak, and every point of purchase-market recovery drops through at high incremental margin (ServiceLink centralized refi margins went 9% → 23% on modest volume). Add free options the market prices at zero: alt-income normalization (+$0.84/sh if alts merely hit the 12–14% LT expectation), the FG fee-pivot re-rate (fee earnings 15% → 25% target; opex 48 → ~45bps; sidecar capital), and any further separation/float catalyst. Bull EPS ~$6.8–7.3 puts the stock at ~7–7.7x. Load-bearing assumption: title volumes normalize toward ~1.3M orders while adjusted margin holds ≥13.5–14% — volume outrunning refi-mix dilution. Fragility: the entire path rests on mortgage rates; if 6.0–6.5% persists, volumes stall and the bull path collapses to base.

Bear variant (argued): both engines compress at once. Title margin normalizes toward 12% (refi-mix dilution plus commercial mean-reversion from a cyclical-high 18% of revenue) on stalled volumes; F&G’s product margin keeps sliding (1.88% → 1.66% annualized is the start of a trend, not one-off timing) as the Fed cuts and crediting rates reset up; the 2021–24 private-credit vintages in a ~50% structured/private book finally impair (7bp history is a fair-weather record); the $20B reinsurance stack (Aspida/Somerset ~$14B) transmits counterparty stress; and the “counter-cyclical smoothing” narrative — never tested in a hard landing — fails exactly when needed, because a housing re-trough and a credit downturn arrive together. Bear EPS ~$3.6–4.1 puts the stock at ~13–14x normalized earnings with a ~4% yield as the only floor. Overlay: governance (Foley grant, title-only incentives paying 200% on below-prior-year targets, Nevada redomicile) argues the alignment discount is partially permanent, and zero insider buying at $43 says the people who know the asset best declined to underwrite the dip. Load-bearing assumption: F&G’s pristine impairment record does not survive a real credit cycle. Fragility: three years of clean credit, 93% surrender protection, matched ALM, and RBC ~430% argue the tail is fatter in narrative than in near-term cash flows.

Where this memo lands relative to consensus: the market’s embedded soft-landing-with-fade (§10.5) sits between our bear and base and is, on the evidence, the reasonable center of gravity. The genuine variant perceptions available are narrow: (a) a quality argument that the Title stub deserves 12–13x, not 11.7x — supported by five years of margin evidence but hostage to the mortgage cycle; (b) a normalization argument that adjusted EPS is understated by alt drag (+$0.84/sh) — supported by management’s own disclosures but five years stale as a promise; © a governance argument that the Foley/alignment discount is partially permanent — supported by the 21.5% withheld vote and the comp record. None of the three, alone or together, manufactures a mispricing large enough to override the SOTP arithmetic at $52 (INTERPRETATION).

12. Fact vs. Interpretation

13. Open Questions

1. F&G spread trajectory. Is Q1-26’s product-margin compression (1.88% → 1.66% annualized; cost of funds 3.20% → 3.42%) one-off timing, as management claims, or the start of a trend as the Fed cuts? The Q2-26 FG supplement is the test.
2. §355 timing. Was a tax-free full spin of F&G legally available in H2-2025 (five-year rule from the June 2020 taxable acquisition; §355(e)/device analysis)? Management’s own quote confirms the option is now gone; whether it had to be sacrificed is inferential. A tax-counsel memo would settle it; material to the capital-allocation grade.
3. Ayers v. Foley. Outcome of the 2026-03-09 motion-to-dismiss hearing is not found in filings or the news feed as of today. Likely immaterial either way (director-comp derivative suit), but unverified post-hearing.
4. Peak Altitude process. Deconsolidation, partner, or sale — outcome promised “the next couple of quarters” from 2026-05-07. Blunt’s full-time move to Peak suggests the process is live; the structure chosen will speak to whether the fee-pivot is strategy or narrative.
5. Alt-income normalization. Five years without “meaningful realizations” (Blunt). Is the 12–14% revised LT expected return on the redefined $4B alt book credible, or is the shortfall structural? Worth +$0.84/sh (FNF share) either way.
6. FG stake direction. Do FG buybacks push FNF back toward 80% (management: “very possible… no target”), or does a third distribution arrive despite its now-taxable character? The answer determines whether the daily SOTP reference shrinks or the separation thesis reactivates.
7. F&G Life Re post-closing adjustments (expected Q2/Q3-26 filings) and the forward-flow MYGA reinsurance economics with Ancient.
8. F&G’s exact 2025 FIA market rank/share (carrier-level Wink/LIMRA tables paywalled; top-10 historically, unverified for 2025) and the FIA industry total ($116B vs $127.9B aggregator discrepancy; FNF’s 10-K implies the higher series for 2024).
9. July 8–10, 2025 −11% drop — unattributed; no 8-K, no peer sympathy, inconclusive news searches.
10. Windacre Partnership (7.0%) — 13G filer with an activist-leaning reputation; no Schedule 13D on file. Worth monitoring as a potential governance catalyst given the Foley withhold vote.
11. Title underwriter statutory surplus/RBC analog — not disclosed (the same gap noted in the companion First American analysis); trough-year title dividend capacity beyond 2026’s ~$437M not pulled.
12. F&G standalone filings (CIK 0001873863) not mined — adjusted spread-margin series, surrender experience detail, and statutory detail live there; F&G figures here are from FNF’s segment reporting and F&G releases/supplements.
13. ServiceLink and home warranty economics — not separable from Title disclosure; any sale/spin would change the Title-multiple story. Corporate & Other real-estate-tech ventures not inventoried by name.

14. What Must Be True

Bull case — what must be true, with the falsification test:

1. Mortgage rates break durably below ~6% and purchase volume genuinely reopens — title closed orders toward ~1.25–1.65M by 2027. Falsifier: 2026 order growth stalls at the Q1-26 pace (+13–16% annualizing to well under 1.1M) or purchase opened orders stay ~flat (+2% y/y in Q1-26 is not a reopening).
2. Adjusted title margin holds ≥13.5–14% as volume returns — volume outruns refi-mix dilution. Falsifier: two consecutive quarters of adjusted margin below ~13% on rising volumes (Q1-26’s 13.1% is already at the edge).
3. F&G’s product margin stabilizes ≥ ~1.6% and the fee pivot delivers (fee share of adjusted earnings marching toward 25% by YE2028; opex → ~45bps). Falsifier: Q2-26 supplement confirms compression is a trend (product margin <1.6% annualized, cost of funds still rising).
4. Alts normalize toward the 12–14% LT expectation. Falsifier: a sixth year of sub-expectation returns on the redefined book.
5. The market grants the Title stub a quality premium (12–13x) rather than FAF-parity-plus-a-token. Falsifier: none mechanical — but if FNF’s trough-margin advantage over FAF narrows materially in the next downturn, the premium argument dies.

Bear case — what must be true, with the falsification test:

1. A hard landing pairs a purchase-market re-trough with F&G credit stress — breaking the counter-cyclical smoothing narrative. Falsifier: title volumes keep recovering through a soft patch while F&G impairments stay near the 7bp run-rate.
2. F&G’s 2021–24 private-credit vintages (roughly half the book) impair materially, and/or the Aspida/Somerset reinsurance stack transmits counterparty stress. Falsifier: another clean credit year with NAIC scrutiny landing ≤ the ~5pt RBC impact F&G estimates.
3. Title margin mean-reverts toward 12% (commercial mix reverses from its cyclical-high 18% of revenue). Falsifier: commercial revenue holds ~$1.5B+ through 2027 while refi/purchase volumes rise.
4. The governance discount widens (Ayers surprises, further Foley awards, Nevada-forum abuse). Falsifier: declassification completes uneventfully by 2029 and the withhold vote on Foley shrinks.
5. The ~4% dividend floor fails — i.e., the payout is cut or frozen. Falsifier: the 2023 precedent itself — dividend raised at the trough at 52% payout with ~2 years of coverage from title capacity plus holdco cash. This is the bear case’s weakest leg.

15. Source Appendix

The full, itemized source list — every filing, release, transcript, data feed, and trade-press citation with dates — appears in Appendix B below. Principal sources: FNF 10-K FY2021–FY2025 (CIK 0001331875; FY2025 filed 2026-02-26), 10-Q Q1-2026 (filed 2026-05-08), ~50 8-Ks (Aug 2021–Jul 2026), DEF 14A (filed 2026-04-29), 95 insider filings (Forms 3/4/4A/5, Jul 2024–Jul 2026); FNF earnings releases and call transcripts (Q3-25, Q4-25, Q1-26); F&G Annuities & Life (FG) Q4-25/Q1-26 releases, Q1-26 financial supplement, and Q1-26 call transcript; AZI five-year price series and FactorsToday factor data (pulled 2026-07-17/19); ROIC.ai market data and news feed; LIMRA, MBA, iEmergent, NAR, ALTA, Fitch, Oliver Wyman, McKinsey and annuity-trade-press industry data (flagged where not verified against regulator originals); and a companion analysis of First American Financial (FAF), cross-read for the shared title-industry frame and the FAF margin/multiple comparisons, attributed inline where used. All sources used are external and public.

This report is the author’s independent analysis, published for general information only. The body of the report carries no recommendation and no price target; the single labeled exception is the “Kimi’s Take” block at the front, which is the author’s own opinion. Facts, interpretations, assumptions, and open questions are labeled as such throughout. Nothing here is investment advice or an offer to buy or sell any security.

APPENDIX A — Standard Diligence Questionnaire

Fidelity National Financial, Inc. (NYSE: FNF) — Report date 2026-07-19

Independent research — report dated 2026-07-19. Companion appendix to the FNF research report. Every answer is grounded in the public sources listed in Appendix B (22/22 primary-source verification checks passed; six rounding-level standardizations applied, noted where used). Labels: [F] Fact (filing/primary-source verified), [I] Interpretation (analyst judgment), [A] Assumption (scenario or derived value). No recommendation, no price target. FNF is a two-engine insurer — the #1 US title family plus ~70%-owned F&G Annuities & Life (NYSE: FG) — so several generic questions are answered with the correct insurance-sector analog and the mapping is stated where it changes.

1. General — the questions thoughtful investors are actually asking

Q: Is the sum-of-the-parts discount real, or already closed? [F] FNF’s ~70% FG stake is 134M shares × 70% × $32.405 (2026-07-17) = $3.04B, 21.7% of FNF’s ~$14.0B market cap — marked to market every day. [I] The base SOTP (~$50/sh: Title at 11.2x TTM adjusted after-tax + FG at market − holdco net debt $1.66B) sits ~4% below the $52.04 price: there is no holding-company discount left to harvest. Inverting the price, the market pays ~11.7x TTM adjusted Title earnings for the stub — a token premium to FAF’s ~11.2x. The two in-specie distributions (Dec 2022, Dec 2025) did their job; the classic “conglomerate discount” thesis is stale. [A] The SOTP multiple choice is judgment; the stake mark, net debt and share counts are verified facts.

Q: What are normalized title volumes and margins — and which way does normalization run? [F] Direct closed orders: 2,169K (2021) → 956K (2025) = 44% of peak; adjusted pretax Title margin 15.9% FY2025 (GAAP 14.45%), only ~1pt below the 2022 margin earned at 28% higher volumes; Q1 2026 refi = 37% of closed orders with fee per file −3% and adjusted margin 13.1%. [I] Volume is cyclically LOW, margin is near-peak — normalization runs in both directions at once (volume up, margin down on refi-mix arithmetic), roughly offsetting. [A] Base-case normalized adjusted EPS ~$5.5–6.0 vs TTM ~$5.12 — recovery is real but not dramatic; management’s stated 15–20% annual margin range held through a 30-year-low purchase market.

Q: Is F&G worth more dead (distributed to shareholders) than alive (inside FNF)? [F] FG trades at 0.99x GAAP book / 0.70x book ex-AOCI ($32.405 vs BVPS ex-AOCI $46.51). [I] The market prices F&G as a sub-cost-of-equity spread lender, so marking the stake at market in an SOTP is not generous, and distributing it has been the correct exit — the round trip (below) returned ~1.4–1.5x. But note the asymmetry management created: dropping below 80% ownership in Dec 2025 burned the §355 tax-free-spin option for the remainder (CFO Park on the Q3’25 call: “that option is off the table”), and FNF openly contemplates FG’s own buybacks pushing its stake back from 70% toward 80% (“no target”). The glide path is not linear; FNF wants FG’s market reference more than it wants a clean exit.

Q: Is F&G’s ~8% ROE above or below its cost of equity? [F] FG adjusted ROE ex-AOCI: 8.2% (FY2025) / 8.4% (rolling 4-qtr Q1’26); adjusted ROA 0.87% LTM; product margin 1.88% FY25 → 1.66% annualized Q1’26. [I] Almost certainly below a ~10–11% cost of equity for a levered spread lender — FG’s 0.70x ex-AOCI price-to-book is the market solving the same equation. Growing retained AUM ~12%/yr at below-COE returns destroys per-share value at the unit level, which is why the March 2025 equity raise at ~0.7x book (and FNF’s $151M participation) was the most value-dilutive capital action in the complex. Mitigant: management knows — the fee-based pivot targets ~25% of adjusted net earnings from fees by YE2028 (from ~15% in 2025), and alt income running below its 12–14% long-term expectation adds +3.4pts of ROE if it normalizes (Blunt’s own math).

Q: What does the Foley governance lattice cost the multiple? [F] Non-independent chairman (3.7% personal holder) with a Director Services Agreement, an 830,152-share special grant (Nov 2024, ~$50M derived value [A]), related-party co-investments (BKSE/BKFC) among three “independent” directors, F&G board interlocks, a Nevada redomicile (Jun 2025), a pending derivative suit over director pay (Ayers v. Foley), and 21.5% of votes withheld on Foley’s June 2026 re-election. Counterweights: board declassification approved near-unanimously (full annual elections by 2029), Cannae below 5%, say-on-pay 95.3%, 5.7% insider ownership. [I] The withheld vote is targeted at Foley personally, not the pay program. A durable governance discount exists and should be treated as partially permanent — but the trajectory is normalizing in form.

2. Cyclicality & Earnings Nature

Are earnings at a cyclical high or low? [I] Low — split by engine. Title: TTM adjusted pretax ~$1,416M earned at 44% of peak 2021 order volume — cyclically depressed revenue, near-peak margin; consolidated adjusted EPS ~$5.12 TTM is a trough-to-early-recovery print, not a peak. F&G: mid-cycle-to-early on AUM growth (+11–12%/yr) but late in the industry spread cycle. [I] (Marathon capital-cycle lens) The annuity industry wrote four consecutive record years ($313B→$461B, 2022–25) on capacity that flooded in from PE-backed writers (~$700B private-capital AUM, 13% of industry vs 1% in 2012); returns on new money are still wide but past peak, and the stored-up risk has migrated from the liability side (2023 surrenders) to the asset side (2021–24 private-credit vintages).

External or internal drivers? [I] Predominantly external on both engines, with real internal offsets. Title: mortgage rates and the origination cycle set volume (MBA 2026 baseline $2.2T, +6–8%, refi +11% — hostage to a 6.0–6.5% rate assumption); internal levers are cost discipline (personnel 60% of direct premiums + escrow fees), commercial build-out ($1,494M, +27%, third-best year ever) and multi-brand share defense. F&G: the rate path drives spreads in both directions (falling rates compress new-money spreads; a re-rise re-ignites surrender risk on the 2022–24 low-crediting block — Oliver Wyman found lapses rose ~5x in 2023 vs 2021 for out-of-surrender-charge policies); internal levers are Blackstone origination yield, the fee-based pivot and opex (60bps → 48bps of AUM, target ~45bps by YE2027).

How stable is revenue, per engine? [F] Title revenue: $11,497M (2021) → $7,038M (2023) → $8,490M (2025) — transaction-driven, swung −39% peak-to-trough; title margins held 12.0–18.6% GAAP (agency mix flexes costs: agents absorb ~77.5% of agent premiums). F&G GAAP pretax: $(35)M → $778M → $323M (2023–25) — LDTI noise, not economics; adjusted (ANE at FNF share) $285M → $475M → $412M, materially smoother. [I] The two engines genuinely dampen consolidated volatility (F&G GAAP peaked exactly when title troughed) — but the smoothing narrative is untested in a hard landing that hits title volumes and F&G credit/surrenders simultaneously.

Market size and growth. [F] US title insurance: ~$18.5B premium pool 2025 (+13.8% y/y), domestic, state-regulated, claims paid only ~3.6% of premiums; Big-4 families hold ~80% of NPW; sector operating margin 11.4% (Fitch). Growth is a function of transactions and home prices, not secular adoption; ALTA notes title costs down ~7.8% since 2004. [F] US annuities: record $461.3B of 2025 sales (fourth consecutive record; LIMRA; finalized $464.1B), FIA ~$116B record (+12%), RILA $79.6B (+20%); indexed products = 45% of sales vs 24% a decade earlier; LIMRA forecasts >$450B through 2026–28 on Peak-65 demographics (4.1M Americans turning 65/yr) and DB-pension decline. [I] Title is a mature, cyclical domestic pool with the best structural seat (FNF ~32% family share, #1); annuities are a genuinely secular demand pool with contested economics — F&G is a mid-tier player (~3% of industry flow) in a market where Athene alone writes ~$33B of fixed annuities with captive origination F&G lacks.

3. Business Quality & Competitive Moat

Is the industry getting more or less competitive? [I] Title: no — shares have been stable for years (FNF family 30.7–32.6% through a full cycle; ALTA/Demotech). Annuities: more — private-credit-backed capacity set the industry’s marginal crediting rate; competition for IMO/agent shelf space is “vigorous” (10-K). Regulation (NAIC RBC overhaul on CLOs/private credit and asset-intensive reinsurance; Treasury scrutiny of the ~$1.5T Bermuda complex) is moving against the PE model’s arbitrage — a slow-motion equalizer that marginally helps independents like F&G.

How profitable is the business (ROE/ROIC decomposition)? [F] Title pretax margin 18.6% peak / 12.0% trough (2021–25), beating FAF every year — by ~200bps at the 2021 peak and ~1,000bps at the 2024 trough (12.5–14.2% vs 8.6%/4.3%). F&G: ~1.9% product margin on $57.6B retained AUM → 0.87% adjusted ROA → 8.2–8.4% adjusted ROE ex-AOCI, levered ~10x on insurance equity with a $2.27B non-recourse debt stack (debt/cap ex-AOCI 26.2%). [I] Title is a high-quality cyclical franchise; F&G is a mid-quality spread lender. Traditional ROIC is not meaningful for insurers (negative enterprise value); ROE-ex-AOCI and pretax margin are the correct decompositions.

Barriers to entry, per engine. [F/I] Title (Greenwald mapping): (1) title plants — decades-built indexed record compilations that “cannot be replicated by simply digitizing public records” (management, corroborated by the stable 4.5% loss provision); (2) 50-state licensing, rate filings, statutory capital; (3) scale economies — ~1,300 direct offices, ~5,100 agents, plant/tech fixed costs spread over ~40% more premium than FAF; (4) multi-brand shelf-space defense — FNF holds ~32% family share while no single brand exceeds ~15%. F&G: no demand-side moat — commodity FIA/MYGA liabilities competing on credited rates; the one supply-side edge (Blackstone ISG manages ~78% of the $69B portfolio) is genuine but purchased (IMA fees) and concentrates manager risk in one counterparty. Contrast Athene, whose captive origination is owned, not rented.

Can the business be understood? Undermined by foreign low-cost labor? [I] Economically, yes — two insurance engines; under GAAP, barely — LDTI fair-value marks (MRB, embedded derivatives) plus the 2025 tax charge make consolidated EPS nearly uninterpretable year-to-year ($1.91 → $4.65 → $2.21). Foreign labor is not a threat: state-licensed, relationship-distributed, domestically regulated businesses.

Do brands matter? Nature of competition? Switching costs? [F/I] Brands matter as shelf space, not consumer preference — end consumers don’t choose the underwriter; lenders, agents and brokers do. Competition is on service, quality, relationships and (state-capped) price. Switching costs are real but low: distribution-level friction (escrow integration, lender panels, splits — agents keep ~77.5% of agent premiums and represent multiple underwriters). No network-effect claim survives scrutiny; scale benefits are supply-side only.

4. Financial Condition & Balance Sheet

Assets not fully recognized on the balance sheet? [F/I] Yes. Title plants (decades of compiled records, effectively carried at cost/nothing yet central to the moat), the brand family, technology platforms (SoftPro, inHere — ~2.8M unique users in 2025), and Corporate & Other real-estate tech ventures carried at ~$0 in the SOTP. Goodwill is already substantial ($2.8B Title, $2.2B F&G).

Off-balance-sheet items? [F] The big one: reinsurance recoverables $17.6B (YE25) → $20.0B (Q1’26), top-5 concentration Aspida $8.5B / Somerset $5.6B / Everlake $1.9B / Ancient Re $1.7B / Wilton $1.0B — collateralized via trusts/funds-withheld/LOCs (funds-withheld liabilities $16.5B partially offset) but a real and growing counterparty stack as the capital-light strategy cedes more (>$15B cumulative new business reinsured). Also: FHLB/FABN funding agreements ($6.2B of reserves at Q1’26) are economically leverage inside contractholder funds; title claim reserves $1.70B; F&G’s $2.27B debt is explicitly non-recourse to FNF but consolidated.

How conservative is the accounting? [F] Title loss provisioning is a mechanical 4.5% of premiums every year 2021–2025 (no releases) vs FAF’s ~3.0%-with-releases — FNF’s is the more conservative, steadier QoE signal (FY2025 claims paid $269M vs $262M provision). [I] LDTI marks cut both ways and swamp GAAP (TTM GAAP EPS $2.81 vs adjusted ~$5.12; 2025 ETR 53.9% on the $471M distribution DTL). Unusually, the adjusted framework currently penalizes reported results: alt-investment income ran $0.84/sh (FNF share) below the 12–14% long-term expected return inside FY2025 adjusted earnings — the reverse of the usual adjusted-EPS skepticism. A “clean” TTM excluding true-ups would be modestly higher (~$5.4–5.6) [A].

How CapEx-hungry? [I] Not — the question doesn’t map to an insurer; the correct analog is capital intensity of growth. Title is float-driven, asset-light, minimal reinvestment (personnel is the cost base, 60% of direct premiums + escrow). F&G is capital-hungry as a spread lender — every dollar of AUM growth consumes statutory capital, which is why FGL Insurance’s 2026 ordinary dividend capacity is $0 (Iowa earned-profits test; no dividends to the holdco in 2025) despite RBC >410% (~430% per F&G) — and why management is pivoting to flow reinsurance, the ~$1B Fort Greene sidecar, and fee income.

5. Capital Allocation & Management

How much FCF, and how is it used? [I] “Free cash flow” doesn’t map cleanly to an insurer; the sector analog is distributable capacity: [F] ~$437M of title-insurer ordinary dividend capacity in 2026 + holdco cash $495M + ~$84M/yr parent share of FG common dividends + $17M FG preferred ≈ ~$1.03B vs a $556M run-rate common dividend + ~$70M parent interest. [F] Deployment priority is unambiguous: (1) the dividend — raised every year ($1.56 → $2.02, 2021–25; $0.52/qtr Feb 2026), including the 2023 trough (payout 52% of adjusted ANE; buybacks cut to $4M the same year); (2) F&G growth-then-separation; (3) buybacks as the pure swing factor ($461M/$549M in 2021–22, ~zero in 2023–24, $252M in 2025). [I] Textbook cyclical discipline on the what; poor-to-mixed on buyback timing ($56.80 avg in 2025 vs the ~$43–52 range since).

The F&G round trip. [F] In: ~$2.7B (Jun 2020, $12.50/sh into a depressed annuity sector) + $250M preferred (2024) + $151M (Mar 2025 offering) ≈ $3.1B. Out/held: ~70% stake at $3.04B market + ~$850M distributed in specie (carrying values) + ~$350–400M cumulative FG dividends ≈ ~$4.4B. [I] ~1.4–1.5x MOIC, high-single-digit IRR [A] — value created, unspectacularly; the cheap 2020 entry and the 2022–24 rate repricing did the work. The two taxable distributions crystallized a $471M deferred-tax charge and forfeited the §355 option — a genuine criticism (management’s own Q3’25 call confirms the tax consequence), countered by the float/liquidity/index-inclusion logic (FG joined the S&P SmallCap 600 in May 2026).

M&A history. [F] F&G 2020 (~$2.7B, approved by special committees of both boards; settled Delaware derivative suit); eight title tuck-ins for $66M (2025); FNTA commercial operations (Oct 2024, undisclosed/immaterial); F&G-level owned distribution (Roar JV 70% $316M, PALH $314M, 2024); abandoned Stewart deal (2019, paid reverse termination fee). [F] Management guides to more title-agent-side M&A in 2026–27 (Park, Q1’26 call).

SBC, comp design, insider behavior — the honest read. [F] NEO pay = 75% adjusted pretax title margin + 25% adjusted title revenue; F&G (~30% of consolidated adjusted earnings) is in no metric. 2025 targets were set below 2024 actuals (revenue $7.0B vs $7.7B actual; margin 13.0% vs 15.1%) and paid at the 200% cap — two consecutive max years (Nolan 2025 total comp $11.65M). [I] Defensible design (line-of-sight; cyclical-appropriate metrics), poor calibration — a pay-for-luck tilt that transfers macro risk to shareholders; flag. [F] Foley’s 830,152-share grant (~$50M derived [A]) to a non-executive chairman already holding ~$500M of stock is the largest governance-cost datapoint. [F] Insider tape: zero open-market FNF purchases by any officer/director in 24 months — including the ~$43 dip — while the company itself bought back $252M + $82M; discretionary sales are small (~145K shares vs ~1.4M granted); largest: CLO Sadowski 69,196 sh @ $45.70 = $3.16M (2026-06-26, near the lows, not 10b5-1-flagged). Director Dhanidina’s 9,543-sh sale at a recorded $418.25 (2026-06-24) is a probable filer error (no correcting 4/A through 2026-07-19) — do not cite the price as real. [I] Nobody at FNF buys the stock with their own money; they are paid in it.

Management motivations. [I] Foley-network deal-making DNA (buy/build inside FNF, then return the asset in specie — Cannae 2017, F&G 2022/2025); a deliberately defended “widow-and-orphan” dividend covenant (~4% yield as valuation floor); and a separation end-state that management pursues at its own pace — Nolan’s new 3-year agreement (May 2026, target incentive 200% of base) removes FNF succession risk while the Blunt→Murphy F&G handoff (Jun 2026) puts the capital-light narrative’s architect in the CEO seat.

6. Valuation & Market Data

Security type. [F] Common stock, Nevada corporation (redomiciled from Delaware Jun 2025). Not an ADR, not an MLP, no K-1. ~269M shares; $14.0B market cap at $52.04 (2026-07-17).

Dividend policy. [F] $2.02/sh declared FY2025 ($546M); $0.52/qtr declared Feb 2026 (~4.0% yield; ~4.5% shareholder yield with buybacks). Raised every year through the window; 2025 payout = 40% of adjusted ANE.

Profitability measure. [F/I] Use adjusted EPS (company-defined non-GAAP): $3.55 → $4.63 → $4.97 → TTM ~$5.12; GAAP ($2.81 TTM) is distorted by the $471M distribution DTL and LDTI marks — and AZI’s GAAP P/E percentile (93.6th) is a denominator artifact, not expensiveness. P/E ~10.2x adjusted; P/B ex-AOCI 1.53x ($34.01). [I] Does net income diverge from cash from operations? For an insurer, CFO includes premium float swings and is not a quality check; the right divergence test is GAAP-vs-adjusted earnings (documented above) and dividend-capacity-vs-payout (comfortable).

SOTP and embedded expectations. [A] Bear ~$45 / Base ~$50 / Bull ~$57 per share vs $52.04 — price ≈ base SOTP + 4%. [I] Embedded: Title at ~11.7x adjusted earnings (token premium to FAF), FG fully marked, and — via justified P/B — a durable ROE of ~12% implying ~$4.1 of “permanent” EPS, ~$1.0 below the current print: the market explicitly fades part of the near-peak title margin and credits only partial volume recovery. It prices no FG re-rate and no separation catalyst. The most fragile assumption in both directions is mid-cycle title margin: ±2pts on ~$10B revenue = ±$0.55/sh.

7. Risks & Downside

What would cause the stock to decline? [F/I] (1) Mortgage-cycle re-trough — every major down-move in five years coincided with rising-rate windows (the stock is a rate-duration proxy; beta 0.56 but negative rate-factor loading −0.24/−0.29). (2) F&G spread compression — cost of funds 3.20% → 3.42% annualized (Q1’26), product margin 1.88% → 1.66%; one quarter so far, unconfirmed trajectory. (3) F&G credit-vintage impairments — ~50% of the $69B portfolio is structured/private credit (CLOs $10.9B, ABS $7.8B, CMBS $5.2B, LPs $4.6B, mortgage loans $7.4B), originated at peak competition, tested only in a benign environment (7bp/yr impairments). (4) Reinsurance counterparties — Aspida/Somerset ~$14B combined recoverables. (5) Recurring F&G GAAP-mark earnings days (each recent mark-driven print sold off 5–7%). (6) Distribution/technical overhang on FG (float doubled; short interest +21% in Feb 2026 per news feed, unverified) and further taxable distributions. (7) Governance — Nevada forum, Ayers v. Foley, pay calibration.

Chance of a total loss? [I] Low. Regulated insurance assets on both engines; industry title claims ~3.6% of premiums; F&G RBC >410%, 93% surrender-charge protection, matched ALM, ~4% below-IG. The five-year max drawdown of −37% (2022–23 double bottom at ~$27.34/27.38) is the realistic bad template; the lifetime −75% drawdown is the 2008 housing-crash legacy, pre-dating the current balance-sheet structure. [I] Catastrophic scenario: a private-credit bust that impairs F&G’s ~50% structured book and breaks a major reinsurance counterparty while rates re-rise and re-ignite surrenders while the purchase market re-troughs — the double-compression that breaks the counter-cyclical-smoothing thesis in both engines at once. Fat in narrative; mitigated in near-term cash flows by surrender protection, collateralized reinsurance and a pristine (if untested) credit record.

8. Recent News & Events — Two-Year Timeline (Jul 2024 – Jul 2026)

[F] All items dated and sourced (8-K filings, PR Newswire originals, transcripts):

• 2024-10-02 — Acquired First Nationwide Title Agency commercial operations (AmTrust; terms undisclosed; talent bolt-on; below SEC-disclosure materiality).
• 2025-01-13 — F&G issued $375M 7.30% junior subordinated notes due 2065.
• 2025-03-24 — F&G equity offering: 8.0M shares @ $33.60; FNF bought 4.5M (~$151M) to limit dilution.
• 2025-06-10 — Ayers v. Foley derivative suit filed (director comp 2022–24; MTD heard 2026-03-09, outcome unresolved as of today).
• 2025-06-11/12 — Shareholders approved redomestication Delaware → Nevada.
• 2025-11-07 — Board approved the second F&G distribution (~12%); FNF dividend raised to $0.52/qtr; F&G dividend +14%.
• 2025-12-31 — 16,280,204 FG shares distributed (6 per 100 FNF); stake to ~70%; $471M non-cash deferred tax charge (Q4’25 GAAP loss $(117)M vs adjusted +$382M/$1.41).
• 2026-02-19 — Record FY2025: title adjusted pretax $1,359M / 15.9% margin; F&G record gross sales $14.6B; F&G Life Re sale announced.
• 2026-03-01 — F&G Life Re sold to Ancient Financial Holdings (~$102M cash + 19.9% LP; +$14M pretax gain; Blackstone retained asset management; forward-flow MYGA reinsurance with Ancient).
• 2026-03-16 — F&G authorized a new $100M three-year buyback (repurchased $29M @ $24.14 in Q1’26 — “almost a silly discount,” Park).
• 2026-05-06/07 — Q1 2026 title inflection: revenue +13%, adjusted title pretax +27%, orders +13–16%, refi mix 33–37%, commercial +15%; adjusted EPS $0.93 (Zacks “miss” vs $1.10 attributed by management to analyst alt-income mis-normalization).
• 2026-05-12/13 — Nolan employment agreement (3-yr, 200% target incentive); FG joined the S&P SmallCap 600 (May 18).
• 2026-06-10/11 — First Nevada annual meeting: board declassification approved near-unanimously (fully annual by 2029); Foley re-elected with 21.5% withheld (highest of class).
• 2026-06-16 — F&G CEO transition: Blunt → Murphy (CFO promoted; Blunt to run Peak Altitude full-time, remains director); formal strategic-alternatives process on Peak Altitude (~$700M deployed, ~$80M EBITDA) under way — un-telegraphed ~6 weeks after the Q1 call.

Accounting-policy / reporting changes in the window. [F] F&G updated ANE methodology post-LDTI (Jul 2023); NAIC Principles-Based Bond Definition reclassification (Jan 2025) moved part of the CLO/loan-backed book; F&G redefined “alts” (Jan 2026: ~$6B debt-like assets to fixed income; alts now ~7% of retained portfolio; LT expected return revised to 12–14%); F&G Life Re deconsolidation (Mar 2026) removes the Bermuda entity and adds the 19.9% Ancient LP pick-up. [I] Net read on the two years (analyst verdict): net strengthen — title margins delivered at the top of guidance through a 30-year-low purchase market and are now inflecting; the F&G stake was converted into a listed, index-included, 70%-owned asset; governance normalized a step behind the capital actions.

All load-bearing figures were verified against primary filings. External/vendor items (LIMRA, MBA, ALTA via the companion First American analysis, AZI, FactorsToday, ROIC.ai) carry source attribution and are labeled accordingly. This appendix carries no investment recommendation and no price target.

APPENDIX B — Source Appendix

Fidelity National Financial, Inc. (NYSE: FNF) — Report date 2026-07-19

Fidelity National Financial, Inc. (NYSE: FNF) — independent research initiation, 2026-07-19

All sources listed below are public: SEC filings (via EDGAR), company releases and public earnings-call transcripts, public data services, trade press, and regulatory references.

(a) SEC filings

FNF = CIK 0001331875; all filings below were pulled from SEC EDGAR (352 filings reviewed, window as noted).

FNF 10-K (annual): FY2021 (filed 2022-02-25) through FY2025 (filed 2026-02-26), via EDGAR. Primary workhorse: the FY2025 10-K (segment MD&A, Item 1 market share/brands, Note S income taxes incl. the $471M outside-basis DTL, legal proceedings incl. Ayers v. Foley, dividends/buybacks).

FNF 10-Q (quarterly): Q2-2021 through Q1-2026 (Q1-26 filed 2026-05-08; Q1-26 Title segment table, order-mix table, F&G Life Re deconsolidation, balance sheet inputs for BVPS).

FNF 8-K / 8-K-A: 50 filings, Aug 2021 – Jul 2026 — quarterly earnings shells plus material events: CEO transition (2022-01-05), first F&G distribution (2022-11-10/2022-12-01), F&G debt issuances (2023-01-13, 2024-06-04, 2024-10-04, 2025-01-13), ransomware (2023-11-21), consent solicitations (2024-04-23, 2025-06-04), F&G equity offering (2025-03-24), Nevada redomestication (2025-06-12), second F&G distribution (2025-11-07, 2025-12-18 — final ratio 16,280,204 shares), Nolan employment agreement (2026-05-12), board declassification + annual-meeting votes (2026-06-11). Caveat: earnings 8-Ks are shells; the EX-99.1 earnings-release exhibits were pulled separately from EDGAR.

DEF 14A / PRE 14A / DEFA14A: 2022 through 2026 (2026 proxy filed 2026-04-29; Summary Comp Table, CD&A incentive metrics/targets, ownership guidelines, related-party lattice, security ownership).

Insider filings (Forms 3/4/4A/5): 95 filings, Jul 2024 – Jul 2026, via EDGAR; parsed into a full transaction table.

Other forms reviewed: 10-K-A, 11-K, ARS, N-PX, S-3ASR, S-8 (indexed, not individually read).

F&G Annuities & Life, Inc. (NYSE: FG; CIK 0001934850) — separate registrant. Used: the FG Q4-25 and Q1-26 earnings releases and the FG Q1-26 financial supplement (AUM rollforward, product margin/cost of funds, ROE ex-AOCI, BVPS, capitalization, reinsurer concentration), all via EDGAR.

Earnings releases (FNF, EX-99.1 exhibits): Q4-25 and Q1-26 releases (adjusted EPS/ANE builds, segment adjusted pretax, commercial quarterly stats, holdco cash); plus Q4-22 and Q4-24 releases for prior-year adjusted figures.

(b) Transcripts

Read in full (public earnings-call transcripts):

• FNF Q1-2026 earnings call (2026-05-07)
• FNF Q4-2025 earnings call (2026-02-20)
• FNF Q3-2025 earnings call (2025-11-07)
• FG Q1-2026 earnings call (2026-05-07)

Not read in full: FG Q4-2025 and Q3-2025 standalone calls (color taken from FNF calls + FG releases); KBW Title Insurance Day (2025-12-03, Seeking Alpha) identified but not fetched. Known transcript artifacts (use filings instead): Park’s “$659M Q2 2025” holdco-cash misstatement; a garbled “up 25%” purchase-order figure in one vendor’s transcript text.

© Data feeds

• ROIC.ai — live quotes (FNF $52.04, FG $32.405 at 2026-07-17 close), multiples, TTM financials, news sweep (FNF + FG, ~48 items each back to Nov 2025; feeds thin before Nov 2025).
• AZI — split-and-dividend-adjusted FNF price series, 5,220 daily rows 2005-10-14 → 2026-07-17, with EMAs/beta/alpha; own-history valuation percentiles (GAAP P/E 93.6th pct — denominator-distorted, flagged).
• FactorsToday — factor loadings, leaderboard, specific vol, similar stocks, regime returns (pulls dated 2026-07-17/19). Vendor caveats noted (dividend-yield field capitalizes F&G stock distributions; rs_* basis ambiguous).

(d) News / trade press

Cited with dates below; none are filing-equivalent — attributed at the point of use:

• Annuity industry data: LIMRA (2026-02-12 record-sales release; 2026 outlook), Financial Advisor Magazine (2026-03-23, $464B final), MyAnnuityStore LIMRA aggregations (2026-05-21, 2026-06-24 FIA history), Athene LIMRA ranking via GlobeNewswire/Quiver Quantitative (2026-03-24). ACLI/Wink carrier tables referenced but paywalled — F&G’s 2025 FIA rank NOT verified.
• Title industry: ALTA (family share via FNF 10-K; underwriter shares via the companion First American analysis), ALTA news on the FNTA acquisition (2024-10-03), AM Best (2024-10-02), HousingWire (May 2026, 2025 premiums), Fitch via Beinsure (Mar 2026, sector margins), Demotech (via 10-K).
• Mortgage/housing forecasts: MBA (Oct 2025 + Apr 2026), iEmergent (Q2 2026), NAR — via the companion First American analysis’s citations.
• Company news (PR Newswire originals): FNF Q4-25 (2026-02-19), FNF Q1-26 (2026-05-06), FG Q4-25 (2026-02-19), FG Q1-26 (2026-05-06), second F&G distribution (2025-11-07, ratio 2025-12-18), FG dividend +14% (2025-11-07), FG $100M buyback (2026-03-16), FG S&P SmallCap 600 inclusion (2026-05-13), FG leadership transition Blunt→Murphy (2026-06-16), FNTA acquisition (2024-10-02), FGL Holdings merger announcement (2020-02-07).
• Regulatory/policy trade press: Capstone DC (2026-05-14), The Global Treasurer (2026-05-13), Trade Treasury Payments (2026-05-14), ABF Journal (2026-06-23), ALIRT via Morningstar/PRN (2026-04-29), Oliver Wyman annuity lapses (2024-07-08), McKinsey PE-insurer stat via ai-cio (2025-01-09).
• Sell-side/opinion (sentiment only, not fact): Zacks (2026-05-06 consensus comparison), Seeking Alpha (4 pieces Feb–Jun 2026), MarketBeat (2026-05-10), defenseworld/gurufocus 13F roundups (treated as noise).

(e) Regulatory / public references

• NAIC: life RBC overhaul (Proposal 2025-16-L on CLO/private-credit capital factors; asset-intensive reinsurance comment period closed Jan 2026; rating-provider due-diligence framework; economic scenario generator field test, Mar 2026 Spring National Meeting); NAIC Principles-Based Bond Definition (effective Jan 2025 — drove F&G’s CLO/loan-backed reclassification); NAIC CLO capital proposal (F&G estimate: ≤5pt RBC impact, per FG Q1-26 call). FGL Insurance domiciled in Iowa (earned-profits dividend test → $0 ordinary dividend capacity 2026).
• RESPA §8 / CFPB: referral-fee enforcement overhang on title affiliated-business arrangements (via 10-K risk factors and the companion First American analysis).
• US Treasury / Bermuda scrutiny: Treasury (Sec. Bessent) convening regulators on private credit/offshore reinsurance Apr–May 2026; Bermuda long-term reinsurance ~$1.52T assets (trade press, section d).
• Fed FSR Nov 2025: PE-linked insurer leverage in historical top quartile (as reported in trade press).
• Ayers v. Foley docket: Patrick Ayers v. William P. Foley II et al., C.A. No. 2025-0650-LWW (Del. Ch.), filed 2025-06-10; MTD filed 2025-08-01; hearing rescheduled to 2026-03-09 — per 10-K FY2025 legal proceedings. Post-hearing outcome not found in filings or news as of 2026-07-19 (UNRESOLVED). Docket itself not independently pulled.
• Prior Delaware action: City of Miami General Employees’ Retirement Trust v. Foley et al. — settlement stipulation (8-K 2022-04-06).
• §355 tax mechanics: analysis is inferential; the Q3-25 call confirms the sub-80% consequence. No tax-counsel source.

(f) Companion analysis used

• A companion analysis of First American Financial (FAF), published alongside this report — cross-read extensively: the shared title-industry frame (ALTA/HousingWire/Fitch/MBA/NAR citations), the FAF margin series for the FNF-vs-FAF through-cycle comparison, the FAF comps table, and event-map peer-sympathy dates. Attribution noted at each reuse point; all FNF-specific figures were derived from FNF’s own filings.

Claims not primary-sourced (register)

1. LIMRA annuity market data ($461.3B/$464.1B 2025 total; FIA ~$116B vs $127.9B discrepancy UNRESOLVED; RILA $79.6B; 2026 >$450B forecast) — trade-body data via press, not SEC. FNF 10-K states FIA $130B / RILA $62B for 2024 (closer to the higher aggregator series).
2. ALTA full-year 2025 single-underwriter shares (FAF 23.1%, FNTI 14.5%, ORI 14.0%, Chicago 13.1%, Stewart 10.9%) — via the companion First American analysis; FNF 10-K corroborates only the 32% family share and Demotech top-4 = 80%.
3. MBA/iEmergent/NAR 2026 forecasts — via the companion First American analysis’s citations.
4. AZI percentiles, FactorsToday loadings, ROIC multiples — vendor computations.
5. Zacks consensus $1.10 (Q1-26 “miss”) — third-party; management disputes normalization basis.
6. FG short interest (+21% Feb 2026 to 1.2%) — news feed only, unverified.
7. Regulatory pipeline details (NAIC proposals, Treasury/Bermuda, ALIRT, Oliver Wyman, McKinsey $700B) — trade press; regulator originals not pulled.
8. Ayers v. Foley post-hearing status — not in any source as of today.
9. FNTA purchase price, LoanCare settlement amount — undisclosed anywhere.
10. SOTP ~$50/sh, normalized EPS scenarios, F&G MOIC ~1.4-1.5x, Foley grant ~$50M value — analyst constructions/derivations on verified inputs (labeled ASSUMPTION/derived throughout), not sourced facts.