
Initiate at Buy with a $309 target as margins normalise more slowly than priced
We initiate on The Allstate Corporation (ALL) at Buy with a $309.41 target (+17.8% total return), though conviction is Low because +1pp on cost of equity moves the rating to Hold. Allstate is a top-four US personal-lines property and casualty (P&C) insurer earning ~$60bn in premiums. Its combined ratio (claims and expenses as a share of premiums; below 100% is profitable) is normalising more slowly than the share price implies.
- Underwriting normalises at 94%, better than the 96% priced in. A 3.3 percentage-point (pp) structural expense cut and auto rate restarts across 39 locations cushion auto give-back.
- Higher yields and about $11bn of FY27E-FY28E payouts support earnings and share reduction. Reinvesting the $87.4bn portfolio toward 5.0% adds about $0.3bn of annual net investment income (NII), while buybacks cut diluted shares 22% by FY30E.
- Reserve releases and 3% policy growth lift FY26E EPS about 7% above consensus. Favourable auto releases (8.2% of earned premiums in 9M26A) and light October catastrophes ($232m) cushion near-term earnings.
What the market is missing. At $266.59, the stock implies a 96.0% normalised Property-Liability (P-L) combined ratio and 15.7% return on equity (ROE), versus our 94.0% and 18.7%. Our adjusted earnings per share (EPS) sits 7.2% above consensus in FY26E ($40.63 vs $37.90) and 6.8% above in FY27E ($29.46 vs $27.58).
Catalysts. Q4 2026 results (Feb 2027) should confirm reserve releases and a new buyback from ~$9bn holdco capital; Q1 2027 (Apr 2027) tests 39-location auto rate earn-in; New York auto approval (within 12 months; no date given) and artificial-intelligence (AI) tools add upside.
Valuation and target. Residual income on adjusted book gives a $305.91 base and $287.58 weighted value (30% Bear, 50% Base, 20% Bull), rolling to $309.41. A peer trailing P/E gives $244: it misses FY27-30E investment income growth, and peer multiples sit on above-normal earnings.
Risks. Auto loss-cost inflation repeating the 2022 trough ($205 Bear case), severe catastrophes, faster reserve fade, or +1pp higher cost of equity (which drops the rating to Hold).
| Market capitalisation | $66.3bn |
| 52-week range | $191.13 – $281.11 |
| 3M avg. daily volume | 1.89m |
| Free float | 99.4% |
| Institutional ownership (SEC 13F) | 81.3% |
| Current annual dividend | $4.32 |
| Current dividend yield | 1.6% |
| Debt / total capital (3Q26) | 17.9% |
| GAAP book value/sh (BVPS) | $120.23 |
| Adjusted BVPS (3Q26) | $128.26 |
Source: Yahoo Finance closes to 18 Nov 2026
| Return to 18 Nov 26 | 1M | 3M | 12M |
|---|---|---|---|
| Allstate (ALL) | +8.0% | +2.1% | +27.0% |
| Rel. vs S&P 500 | +12.8pp | +6.9pp | +16.3pp |
| US$m (except per share) | FY24A | FY25A | FY26E | FY27E | FY28E |
|---|---|---|---|---|---|
| P-L premiums earned | 53,866 | 57,682 | 60,097 | 63,795 | 67,971 |
| Growth y/y | +11.2% | +7.1% | +4.2% | +6.2% | +6.5% |
| P-L underwriting inc. | 3,286 | 8,723 | 9,393 | 4,830 | 4,523 |
| P-L adj. combined ratio | 93.9% | 84.9% | 84.4% | 92.4% | 93.3% |
| Net inv. inc. (excl. PS) | n.a. | 3,350 | 4,109 | 4,449 | 4,726 |
| Adjusted net income | 4,906 | 9,304 | 10,437 | 7,067 | 7,059 |
| Adjusted EPS ($) | 18.32 | 34.83 | 40.63 | 29.46 | 31.32 |
| Adj. EPS growth y/y | n.m. | +90.1% | +16.7% | −27.5% | +6.3% |
| Multiples at $266.59 | FY24A | FY25A | FY26E | FY27E | FY28E |
|---|---|---|---|---|---|
| P/E (adjusted) | 14.6x | 7.7x | 6.6x | 9.0x | 8.5x |
| Price/book (P/B, adj.) | 3.54x | 2.48x | 1.95x | 1.76x | 1.56x |
| P/B (GAAP BVPS) | 3.68x | 2.46x | 2.08x | 1.87x | 1.65x |
| Dividend yield | 1.4% | 1.5% | 1.6% | 1.7% | 1.9% |
| Adjusted ROE | 26.8% | 38.3% | 33.6% | 20.5% | 19.5% |
| Debt / total capital | 26.7% | 19.8% | 17.3% | 16.8% | 16.0% |
| Adjusted BVPS ($) | 75.25 | 107.30 | 136.43 | 151.62 | 170.44 |
| P/E at target ($309.41) | 16.9x | 8.9x | 7.6x | 10.5x | 9.9x |
Source: Company filings (FY24A–FY25A); Yahoo Finance; Penmark Research model (FY26E–FY28E). Shaded columns are estimates. Net investment income (NII) shown excl. Protection Services (FY24A not separately reported); enterprise-value multiples omitted for insurers.
Base case of $306 (+15%); upside to bull is 1.6x the downside to bear
Source: Yahoo Finance; Penmark Research model
Auto underlying ratio held near 92%, normalised P-L combined ratio 92.5%, ongoing reserve releases, benign weather, 5.1% yield, NY/direct share gains.
Auto combined ratio drifts to mid-90s by FY28E, normalised P-L ratio 94.0%, modest releases fade by FY29E, normal catastrophes, yields reach 5.0%.
Auto loss trend outruns rate (FY28E repeats 2022 trough at 103.6% underlying), adverse reserve additions (−4.2%), FY23-level catastrophes, normalised ratio 96.5%.
| US$m | FY24A | FY25A | FY26E | FY27E | FY28E |
|---|---|---|---|---|---|
| P-L premiums earned | 53,866 | 57,682 | 60,097 | 63,795 | 67,971 |
| Auto underwriting inc. | 1,933 | 5,828 | 5,892 | 2,591 | 2,320 |
| Homeowners uw. inc. | 1,363 | 2,443 | 3,002 | 1,872 | 1,823 |
| Other P-L uw. inc. | (10) | 452 | 499 | 367 | 380 |
| P-L underwriting inc. | 3,286 | 8,723 | 9,393 | 4,830 | 4,523 |
| Net inv. inc. (excl. PS) | n.a. | 3,350 | 4,109 | 4,449 | 4,726 |
| Corp., PS, tax & pref. | 1,620* | (2,769) | (3,065) | (2,212) | (2,190) |
| Adjusted net income | 4,906 | 9,304 | 10,437 | 7,067 | 7,059 |
Source: Company filings; Penmark model. NII excl. PS; *FY24A net incl. total NII ($3,092m) and PS.
| % of earned premiums | FY24A | FY25A | FY26E | FY27E | FY28E |
|---|---|---|---|---|---|
| P-L adj. combined ratio | 93.9% | 84.9% | 84.4% | 92.4% | 93.3% |
| Auto adjusted CR | 94.7% | 84.7% | 84.7% | 93.6% | 94.6% |
| Auto underlying CR | 93.4% | 88.1% | 89.6% | 93.0% | 93.5% |
| Auto reserve releases (+) | +0.9% | +4.8% | +6.5% | +1.0% | +0.5% |
| Homeowners adjusted CR | 89.8% | 84.1% | 82.5% | 90.0% | 91.0% |
| Homeowners underlying | 62.5% | 57.9% | 59.5% | 61.0% | 62.0% |
| Homeowners cat load | 27.8% | 26.6% | 24.3% | 29.0% | 29.0% |
| Catastrophe losses ($m)* | 4,964 | 4,959 | 5,088 | 6,078 | 6,560 |
Source: Company filings; Penmark model. *FY24A-25A total P-L; FY26E-28E auto & homeowners.
| US$m | FY24A | FY25A | FY26E | FY27E | FY28E |
|---|---|---|---|---|---|
| Total investments | 72,610 | 83,237 | 89,317 | 93,331 | 98,800 |
| Common equity (GAAP) | 19,441 | 28,609 | 31,861 | 33,064 | 35,360 |
| Unrealised bond AOCI | (771) | 297 | (2,024) | (2,024) | (2,024) |
| Adj. common equity | 20,212 | 28,312 | 33,885 | 35,088 | 37,384 |
| Preferred equity | 2,001 | 2,001 | 2,001 | 2,001 | 2,001 |
| Total debt | 8,085 | 7,490 | 7,494 | 7,494 | 7,494 |
| Total adjusted capital | 30,298 | 37,803 | 43,380 | 44,583 | 46,879 |
| Adj. equity / P-L EP | 37.5% | 49.1% | 56.4% | 55.0% | 55.0% |
Source: Company filings; Penmark Research model.
| US$ or m shares | FY24A | FY25A | FY26E | FY27E | FY28E |
|---|---|---|---|---|---|
| Adjusted EPS ($) | 18.32 | 34.83 | 40.63 | 29.46 | 31.32 |
| Dividends per share ($) | 3.68 | 4.00 | 4.32 | 4.64 | 4.99 |
| Adjusted BVPS ($) | 75.25 | 107.30 | 136.43 | 151.62 | 170.44 |
| GAAP BVPS ($) | 72.35 | 108.45 | 128.28 | 142.87 | 161.21 |
| Diluted shares (m) | 267.8 | 267.1 | 256.9 | 239.9 | 225.4 |
| Common dividends ($m) | 962 | 1,036 | 1,095 | 1,138 | 1,139 |
| Share repurchases ($m)* | 2 | 1,233 | 3,770 | 4,726 | 3,624 |
| Total cash returned ($m) | 964 | 2,269 | 4,865 | 5,864 | 4,762 |
Source: Signals cash flows (FY24A-25A); Penmark model. *FY26E incl. other equity movements.
Adjusted ROE settles near 20% while EPS sits about 7% above consensus
| % | FY24A | FY25A | FY26E | FY27E | FY28E |
|---|---|---|---|---|---|
| Growth (y/y) | |||||
| P-L premiums earned | +11.2% | +7.1% | +4.2% | +6.2% | +6.5% |
| Auto premiums earned | +10.7% | +4.4% | +1.2% | +5.1% | +6.1% |
| Homeowners premiums | +13.8% | +15.0% | +11.6% | +9.2% | +8.2% |
| Net inv. inc. (excl. PS) | n.a. | n.a. | +22.7% | +8.3% | +6.2% |
| Adjusted EPS | n.m. | +90.1% | +16.7% | −27.5% | +6.3% |
| Dividends per share | +3.4% | +8.7% | +8.0% | +7.5% | +7.5% |
| Underwriting & investment margins | |||||
| P-L underwriting margin | 6.1% | 15.1% | 15.6% | 7.6% | 6.7% |
| Auto adjusted margin | 5.3% | 15.3% | 15.3% | 6.4% | 5.4% |
| Homeowners adj. margin | 10.2% | 15.9% | 17.5% | 10.0% | 9.0% |
| Net investment yield (excl. PS) | n.a. | 4.30% | 4.78% | 4.85% | 4.92% |
Source: Company filings; Penmark Research model.
| % or x | FY24A | FY25A | FY26E | FY27E | FY28E |
|---|---|---|---|---|---|
| Returns & capital strength | |||||
| Adjusted ROE | 26.8% | 38.3% | 33.6% | 20.5% | 19.5% |
| Adj. equity / P-L EP | 37.5% | 49.1% | 56.4% | 55.0% | 55.0% |
| Debt / total capital | 26.7% | 19.8% | 17.3% | 16.8% | 16.0% |
| Debt / adjusted equity | 40.0% | 26.5% | 22.1% | 21.4% | 20.0% |
| Cash payout of adj. NI | 19.6% | 24.4% | 46.6% | 83.0% | 67.5% |
| Valuation multiples at $266.59 | |||||
| P/E (adjusted EPS) | 14.6x | 7.7x | 6.6x | 9.0x | 8.5x |
| Price/book (P/B, adj.) | 3.54x | 2.48x | 1.95x | 1.76x | 1.56x |
| P/B (GAAP BVPS) | 3.68x | 2.46x | 2.08x | 1.87x | 1.65x |
| Dividend yield | 1.4% | 1.5% | 1.6% | 1.7% | 1.9% |
| P/E at target ($309.41) | 16.9x | 8.9x | 7.6x | 10.5x | 9.9x |
Source: Company filings; Penmark Research model. No EV multiples shown for insurers.
| Measure (US$ or US$m) | Q4 26E | FY26E | FY27E | FY28E | Source of our gap |
|---|---|---|---|---|---|
| Adjusted diluted EPS (US$) | |||||
| Penmark estimate | $11.26 | $40.63 | $29.46 | $31.32 | Q4 light cats ($232m Oct) & releases |
| Consensus (MarketBeat / Zacks) | $8.49* | $37.90 | $27.58 | n.a. | Street embeds steeper margin fade |
| Prior-year period | $14.31 | $34.83 | $40.63 | $29.46 | FY25A peak 72.9% Q4 combined ratio |
| Penmark vs consensus (%) | +32.6% | +7.2% | +6.8% | n.a. | Higher NII (+$0.3bn/yr) & buybacks |
| Penmark y/y growth | −21.3% | +16.7% | −27.5% | +6.3% | Trough EPS in FY27E, growth resumes FY28E |
| Property-Liability premiums earned (US$m) | |||||
| Penmark estimate | 15,291 | 60,097 | 63,795 | 67,971 | Auto policies in force (PIF) +3.5% & Home average premium +6.0% in FY27E |
| Prior-year period | 14,776 | 57,682 | 60,097 | 63,795 | 39-location auto rate restarts earn in FY27E |
| Penmark y/y growth | +3.5% | +4.2% | +6.2% | +6.5% | Accelerating top line through FY28E |
Source: Company filings; MarketBeat (FY26E) and Zacks Research (FY27E) consensus as of 17–18 Nov 2026; Penmark Research model. *Q4 26E consensus implied ($37.90 FY26E less $29.41 9M26A).
Source: Company filings; MarketBeat (FY26E) and Zacks Research (FY27E) consensus; Penmark Research model
Allstate trades near its five-year book multiple despite a lower expense base
| Measure | Allstate | Peer med. | Diff. |
|---|---|---|---|
| Valuation multiples (19 Nov 2026) | |||
| Price / book (reported equity) | 2.04x | 1.94x | +5.2% |
| Price / trailing 12m (TTM) EPS | 5.67x | 10.90x | −48.0% |
| Profitability (trailing 12m) | |||
| Implied TTM ROE (P/B ÷ P/E) | 36.0% | 22.9% | +13.1pp |
| Market performance & scale | |||
| 6-month price change | +24.4% | +12.1% | +12.3pp |
| Market capitalisation (US$bn) | 66.3 | 36.4 | +82.1% |
Allstate model memo: 2.08x adjusted P/B (3Q26), 9.05x FY27E P/E, 18.7% normalised ROE, 1.62% dividend yield.
Source: Signals peer screen (19 Nov 2026: PGR, TRV, CB, HIG, CINF, WRB, MCY); Penmark Research model.
- Feb 2027Q4 2026 results & new buyback. Tests FY26E EPS beat ($40.63 vs $37.90 consensus) and fresh repurchase mandate from ~$9bn holdco capital.
- Apr 2027Q1 2027 auto rate earn-in. Tests whether restarted rate filings in 39 locations stabilise auto underlying combined ratio near 93%.
- Within 12mNew York approval & AI launch. Regulatory clearance (no date given) unlocks NY auto growth alongside early-2027 AI rollout.
- Aug 2027Q2 2027 mid-cycle check. Reveals whether P-L combined ratio holds near 92%–94% or slips toward the 96% priced in.
Source: Company disclosures; Penmark Research.
- Cost of equity sensitivity (Low conviction): Higher discount rates compress residual income and cap conviction at Low (see valuation sensitivities).
- Deeper auto loss-cost cycle: Severity outrunning rate repeats the FY22 trough (103.6% auto underlying CR), driving the $205 Bear case.
- Catastrophe tail events: Homeowners catastrophe load above 38% (FY23 level) erodes underwriting earnings and pauses buybacks.
- Faster reserve-release fade: Management warns not to model prior-year releases; earlier exhaustion trims near-term EPS support.
Source: Company filings; Penmark Research.
Source: Company filings; Yahoo Finance monthly closes (to 18 Nov 2026)
Source: Signals peer screen (19 Nov 2026); implied TTM ROE = P/B divided by TTM P/E for all eight carriers
The thesis in four charts: BVPS growth, shallower cycle, rising NII and policy gains
Four charts summarise why Allstate compounds book value faster than its current valuation implies. GAAP book value per share has tracked Allstate's share price over the past five years (correlation r = 0.82). Even as underwriting profits normalise from FY25-26E peaks, a lower expense base, a 24% larger investment portfolio at higher yields, and policy growth keep mid-cycle returns near 19%.
Source: Company filings (FY21A-3Q26A); Yahoo Finance closes (to 18 Nov 2026)
Source: Company filings (FY17A-FY25A); Penmark Research model (FY26E-FY30E)
Source: Company filings (FY25A excl. Protection Services); Penmark Research model (FY26E-FY30E)
Source: Allstate quarterly investor supplements (4Q24-3Q26)
Four debates separate our $309 target from today's $267 price
The market agrees personal-lines underwriting margins have peaked, but disagrees on how far they fall. The key drivers of Allstate's share price map to the four named debates below. Cycle depth carries the largest valuation gap, while reserve releases and auto margin pacing determine near-term earnings revisions.
| Debate | Market view | Penmark view | Key evidence | What changes view | Resolves |
|---|---|---|---|---|---|
| Cycle depth | P-L combined ratio mean-reverts to 96.0% (price-implied) or above the 94.4% FY17–25 mean. | Normalises at 94.0% (0.4pp better than history, excl. intangibles), sustaining 18.7% mid-cycle ROE. | Expense ratio structurally fell 3.3pp (24.6% to 21.4%); homeowners underlying CR rose from 57.9% in FY25 to 60.7% in 9M26 but stays profitable. | P-L combined ratio exceeds 95.5% in a normal cat year. | Q2 27 / Feb 28 |
| Auto margin pace | Competition forces steep price cuts; auto underlying CR jumps back toward 96%–100%. | Auto margin give-back is deliberate and gradual (93.0% FY27E, 93.5% FY28E). | Q3 26 adjusted auto CR of 91.4% sits below mid-90s target; rate restarted in 39 locations ($175m). | Auto underlying loss per policy rises >5% while rate stalls. | Q1 27 (Apr 27) |
| Reserve release fade | Favourable prior-year auto releases end immediately in Q4 26 as severity rebounds. | Releases taper gradually (+6.5% FY26E, +1.0% FY27E, +0.5% FY28E, 0% FY29E+). | 9M26A auto releases reached 8.2% of EP (+9.1% in 3Q26) on lower bodily-injury claim severity. | Prior-year reestimates turn adverse (<0%) before FY28E. | Q4 26 (Feb 27) |
| Capital return & yield | Higher rates compress P/B multiple; capital sits idle or is consumed by catastrophes. | $87.4bn book reprices to 5.0% (+$0.3bn/yr NII); $14.1bn of FY27E-28E adj. NI plus ~$9bn holdco liquidity fund $10.6bn payouts. | FY26E model yield rises to 4.78% (excl. PS) vs 4.30% in FY25A; $1.3bn left on $4bn buyback requires early-2027 renewal. | No new buyback authorised in Feb 2027 or yield stalls <4.7%. | Q4 26 (Feb 27) |
Source: Company filings; Q3 2026 earnings call; Penmark Research model.
Source: Penmark Research model sensitivities
Source: Company reporting calendar; Penmark Research
How a 95-year-old Sears side project turned a 107% underwriting trough into a 38% ROE engine
Allstate insures 39.3 million property-liability policies (auto, homeowners, renters and specialty) and 178 million protection plans on $60 billion of annual premiums. After repair inflation pushed losses past premiums in 2022, cost cuts stripped 3.3 points of expenses out of a business the market still prices like the old one.
Every six months, an American driver pays roughly $737 up front to renew an Allstate auto policy, long before a fender-bender ever happens. Founded in 1931 inside Sears, Roebuck & Co. and spun off in 1995, Allstate now protects one in ten US personal vehicles and nearly one in ten US homes with about 53,000 employees.
In everyday words, Allstate pools small payments from millions of households to cover rare, costly shocks: car crashes, hailstorms, house fires and broken phones. Customers buy through three doors: exclusive Allstate agents, 58,700 independent agencies added largely via the $4.0bn National General deal in 2021, and 2,200 direct reps. Separately, Protection Services sells device warranties in over 15,000 retail stores (~2% of FY26E adjusted earnings).
How the money comes in
Allstate earns money on two engines: underwriting profit and investment income on its $87.4bn portfolio (~$45bn of policyholder float). Personal auto brings 64% of FY26E Property-Liability earned premiums ($38.5bn), homeowners 29% ($17.2bn), and other lines 7% ($4.4bn). At our 94.0% mid-cycle combined ratio, 73 cents of each dollar pay claims and 21 cents cover expenses, leaving 6 cents of profit, plus $4.1bn of FY26E investment income earned on an $87.4bn bond-heavy portfolio.
Where the story turns
Five years ago, Allstate looked like a slow legacy carrier. In H2 2021, management sold life and annuity units just before repair inflation pushed recorded combined ratios to 106.6% in FY22 and 104.5% in FY23 (106.1% and 104.0% excl. intangibles). Leadership raised rates through 2022-24 alongside its 2019 Transformative Growth plan (trimming 3.3pp of expenses by 2025), rolled out Affordable, Simple, Connected (ASC) auto across 46 states, lifted new business 125% since 2019 (12.4m policies), and sold Health & Benefits in 2025 ($3.25bn announced; $3.1bn proceeds).
Why it works and why it might not
The loop works because a 3.3pp lower expense ratio versus its FY17-19 average lets Allstate price competitively across three sales channels while sustaining an estimated 19% normalised ROE. The one force that can break the loop sits at the heart of our Cycle depth and Auto margin pace debates: if rivals slash prices or bodily-injury inflation outruns state rate approvals, underwriting margins could snap back toward break-even.
P-L expense cut
Portfolio simplification left a focused personal-lines insurer with an $87bn portfolio
Source: Company filings; Penmark Research
1. Policyholders
2. Channels
3. Earned Premiums
4. Claims & Cost
5. Float & Yield
6. What Is Left
Source: Allstate 3Q26 filings, DEF 14A and Penmark Research model (FY26E base case). *FY26E buybacks include other equity movements.
Fixed vs variable cost & cash generation
Claims are variable with miles driven and weather, while technology and overhead scale with policy count. Cutting 3.3pp from the expense ratio since FY17–19 came from digitising claims, retiring legacy IT systems and restructuring agency commissions. Because premiums arrive upfront while bodily-injury claims settle over months or years, Allstate generated $10.1bn of operating cash flow in FY25A against just $0.2bn of capital spending.
Customer, supplier & geographic concentration
Granular household risk eliminates single-customer concentration, leaving state regulation as the main exposure. With 39.3m P-L policies across all 50 US states and Canada, no single policyholder is material. Geographic concentration centres on large personal-lines states, especially California, New York, New Jersey, Texas and Florida, where rate approval timelines and catastrophe reinsurance pricing govern local profitability.
Auto and homeowners generate 93% of premiums; policy growth and rate drive top line
Source: Penmark Research model (FY26E; NII $4.1bn excl. Protection Services, Prot. Services adj. NI $0.2bn shown separately)
Source: Penmark Research model (FY27E base case)
Pricing, mix & the ASC product transition
Auto average earned premium dipped 2.7% y/y in 3Q26 to $1,474 as Affordable, Simple, Connected (ASC) product mix shifts toward lower-coverage options rather than broad rate cuts. ASC unbundles coverages to win price-sensitive shoppers across 46 states while holding target margins. With underlying auto margins running ahead of target (91.4% adjusted in 3Q26 vs mid-90s goal), management selectively restarted modest rate increases (+0.4% overall, $175m across 39 locations in 3Q26) that earn in over 12 to 18 months.
Homeowners pricing power & Protection Services
Homeowners remains Allstate's quiet compounding engine. Policies in force grew 3.3% y/y in 3Q26 with new business up 10.2%, while earned premiums climbed 11.5% y/y as inflation-indexed dwelling limits and rate increases took hold. Protection Services, led by SquareTrade and Arity telematics (bolstered by the October 2026 INRIX acquisition and pending Replica close), contributes $224m of FY26E adjusted net income (~2% of FY26E adjusted earnings) and adds valuable driving-behaviour data for auto pricing.
A $550bn+ US personal-lines market exits a historic hard market into mid-cycle normalisation
US personal auto and homeowners insurance is a mandatory, non-discretionary market anchored by vehicle and housing stock. Industry direct premiums written (DPW) reached $188.9bn in homeowners and over $211bn in private auto liability alone in 2025 (estimated total private passenger auto exceeds $360bn). Following a brutal 112.2% industry personal-auto combined ratio in 2022, aggressive rate hikes restored industry auto profitability to 91.8% in 2025 (92.9% total P&C), putting the sector just past peak underwriting margins.
Source: Allstate 10-K filings (Auto recorded CR); AM Best US P&C & Homeowners Outlooks (2021-2025); US BLS (Oct 2026); LexisNexis (2025)
Source: Allstate 10-K, DEF 14A and Q3 2026 disclosures; Penmark Research
Loss-cost inflation & catastrophe trends
Repair costs have re-accelerated recently and bodily-injury social inflation stays elevated. Motor-vehicle maintenance and repair consumer price index (CPI) rose 2.4% over the three months to October 2026 (~10% annualised), posing a near-term loss-cost risk. Meanwhile, attorney-represented claims lifted bodily-injury payments from under 20% of claim dollars in 2022 to above 26% in 2025, and convective storms added 7.6 points to the industry combined ratio in 2025.
State regulation & rate-filing friction
Insurance is regulated state-by-state, creating lags between loss spikes and rate earn-in. File-and-use states allow swift pricing adjustments, whereas prior-approval states, most notably California, New York and New Jersey, can delay approvals by 6 to 18 months. Allstate's pending New York auto product approval illustrates both the regulatory drag and the latent growth catalyst once approved.
Allstate holds #2 US homeowners and #4 auto share with three-channel reach
US personal-lines scale is concentrating in the top four national carriers, which together control nearly 60% of private passenger auto. Allstate ranks fourth in US private passenger auto at 10.2% share, behind State Farm and Progressive (18.6% each) and GEICO (11.6%), and second nationally in homeowners at 9.4% share (behind State Farm at 18.7%, ahead of USAA at 7.0% and Travelers at 4.6%).
Source: AM Best (2026 rankings) / NAIC via S&P Global Market Intelligence (2025 US DPW; State Farm 18.6% auto & 18.7% home)
Source: Penmark Research assessment based on company filings and NAIC data
Durable advantages: homeowners & three-channel scale
Few peers match Allstate's combined personal-lines breadth across all three distribution doors. Progressive dominates direct auto and efficiency but carries a smaller, less historically profitable homeowners footprint; State Farm lacks independent-agent distribution; GEICO relies on third-party partners for home coverage. Allstate's bundled exclusive agencies, 58,700 independent-agent locations and $2.2bn direct marketing engine capture shoppers regardless of how they buy, lifting new business 125% since 2019.
Data and pricing infrastructure deepen the moat. Arity's telematics database, expanded through the October 2026 INRIX acquisition and pending Replica close, feeds real-time driving risk into ASC pricing across 46 states, while in-house catastrophe modelling keeps homeowners underlying combined ratios near 60%.
Vulnerabilities & counter-evidence on moat
Allstate is not the lowest-cost direct carrier, and claims satisfaction trails leaders. Progressive's lean direct-acquisition engine still operates at a structural expense edge in pure-play auto, forcing Allstate to spend $2.2bn on advertising and launch a new exclusive-agent operating model in January 2028 to protect acquisition economics.
Customer claims scores leave room to improve. Allstate scored 693 out of 1,000 in the 2025 J.D. Power auto claims study (below industry average), underscoring the retention risk if competitors undercut renewal pricing. Sustaining 3%+ policy growth without sacrificing margins requires disciplined execution on both AI-enabled service and claims speed.
92% of CEO pay is at risk, backed by a stake worth 129 times salary (proxy basis)
| Who | Tenure & background | How they are paid (FY25 DEF 14A) | What they own (Mar 2026) |
|---|---|---|---|
| Thomas J. WilsonChair, President & CEO | 19 yrs as CEOJoined 1995 (31 yrs); CEO Jan 2007. Prior: Sears / Dean Witter. Age 68. | $22.92m total Salary/oth 8% · Cash 24% · Equity 68%. Performance Stock Awards (PSAs): 60% Perf. Net Income (PNI) ROE, 40% Rel. TSR. |
$737.2m (1.07%)2.77m sh & ex. opts · 128.9x salary (proxy basis; vs 8x req.). Form 4: Nov 26 sales (~174k sh). |
| Christian M. LownEVP & CFO NEW | <1 yr in roleJoined 3 Aug 2026. Prior: CFO CoStar Group (2024–26), CFO Freddie Mac (2020–24). | $5.91m target (est. from 8-K) $875k salary · 200% cash · 375% equity (60% PSA/20% Restricted Stock Units (RSU)/20% opt) + $2m sign-on. |
n.d.Joined after Mar 2026 proxy record date; 4x salary guideline applies. |
| Jesse MertenEVP; President, P-L | 1 yr in P-L roleFormer CFO (2022–25); runs core Property-Liability underwriting engine. | $6.86m total Salary/oth 16% · Cash 33% · Equity 51%. Tied to PNI, PNI ROE & Rel. TSR. |
$18.0m (<0.1%)67.5k sh & ex. opts · 8.3x salary (proxy basis; exceeds 4x req.). |
| Mario RizzoEVP & COO | 1 yr as COOLong-tenured executive; former CFO and President, Property-Liability. | $7.30m total Salary/oth 19% · Cash 30% · Equity 51%. Tied to PNI, PNI ROE & Rel. TSR. |
$97.2m (0.14%)364.5k sh & ex. opts · 17.1x salary (proxy basis; exceeds 4x req.). |
| John DugenskePres., Inv. & Corp. Strategy | 9+ yrs at AllstateOversees $87.4bn investment book; served as interim CFO to Aug 2026. | $7.78m total Salary/oth 14% · Cash 35% · Equity 51%. Tied to PNI, PNI ROE & Rel. TSR. |
$78.4m (0.12%)294.1k sh & ex. opts · 10.2x salary (proxy basis; exceeds 4x req.). |
Source: 2026 DEF 14A (10 Apr 2026), Form 8-K (14 Jul 2026), Form 4 filings (Nov 2026). Ownership valued at $266.59; includes exercisable options per proxy; unvested PSAs excluded. All 21 insiders own 1.55%. Multiples follow DEF 14A guidelines (31 Dec 2025 basis vs FY25 salaries); dollar values at $266.59.
Source: Signals SEC cash flow statements (FY21A-FY25A sum)
Pay alignment & governance safeguards
Incentives reward per-share value and capital discipline. Long-term Performance Stock Awards (PSAs) vest 60% on three-year average Performance Net Income ROE and 40% on Relative Total Shareholder Return (TSR), directly aligned with book-value compounding. Shareholders backed say-on-pay with over 95% approval in 2025. Both Dodd-Frank restatement and discretionary misconduct clawbacks cover cash and equity; no executive pledges shares or receives excise-tax gross-ups.
Board independence & succession watch
The board maintains a strong independent majority with an independent Lead Director overseeing the combined Chair/CEO structure. Combined Chair/CEO Thomas Wilson (age 68) brings 19 years of cycle experience, while internal rotations of Mario Rizzo to COO and Jesse Merten to Property-Liability President plus external CFO hire Christian Lown strengthen succession bench depth. Nov 2026 Form 4 filings showed CEO sales of ~174k shares ($45.2m, not marked 10b5-1); without purpose context, we note these neutrally beside his 128.9x salary stake (proxy basis).
Earnings and ~$9bn of holdco liquidity fund $11bn of FY27E-FY28E payouts
Source: Signals SEC cash flows (FY21A-FY25A = $11.9bn); Penmark Research model (FY26E-FY28E; FY26E incl. other equity movements)
| Commitment / Target | Stated guidance | Delivery / Model view | Assessment & implication for value |
|---|---|---|---|
| Auto combined ratio | Mid-90s (94%–96%) through cycle | 85.0% FY25A; 93.6% FY27E; 95.1% FY29E | Outperformed target during 2025–26 recovery; deliberate reinvestment in growth via rate & marketing brings ratio smoothly inside target by FY29E. |
| P-L policy growth | Positive PIF growth in FY25 (Feb 2025) | Met: +2.0% FY25A; +3.3% y/y in 3Q26 | Management delivered on its growth inflection ahead of schedule; new business up 125% since 2019 across exclusive, independent and direct channels. |
| Share repurchases | $4.0bn mandate (Feb 26–28); $1.3bn left | $3.8bn* FY26E; $4.7bn FY27E; $3.6bn FY28E | Paused buybacks in FY24 ($2m) to rebuild capital after FY22–23 losses, then resumed aggressively. Requires fresh authorisation by Q4 26 results (Feb 2027). |
| Portfolio simplification | Exit non-core Life (2021) & Health (2025) | Met: $3.25bn announced ($3.1bn proceeds) | Removed capital-heavy, lower-multiple units; redeployed proceeds into the $87.4bn investment portfolio and ~$9bn deployable holding-company liquidity. |
| Long-term adj. ROE | 14%–17% historical goal (FY20 release) | 38.3% FY25A; 18.7% model normalised | Higher bond yields (~5.0%) and 3.3pp lower expense ratio lift sustainable ROE above the legacy 14%–17% band without extra financial leverage. |
Source: Q4 2024, Q4 2025 and Q3 2026 releases and earnings calls; Penmark Research model.
Return on cash uses vs cost of capital
Repurchases below intrinsic value offer the highest risk-adjusted return on excess capital. With adjusted equity rising to 56.4% of earned premiums by end-FY26E (vs a 55.0% target) and debt-to-capital falling to 17.3%, Allstate holds roughly $9bn of deployable holdco liquidity with no need to inject capital into operating subsidiaries. Retiring shares at 9.0x FY27E earnings (~11% earnings yield) beats our 9.21% cost of equity, while small Arity bolt-ons (INRIX, Replica) avoid large M&A integration risk.
What management & capital allocation mean for value
Disciplined share retirement bridges the gap between moderating profits and rising per-share value. Even though group adjusted net income steps down from $10.4bn in FY26E to $7.1bn in FY27E–FY28E as margins normalise, returning $10.6bn over two years cuts diluted shares from 256.9m to 225.4m (and 200.4m by FY30E). That buyback engine lifts adjusted EPS back to $38.91 by FY30E and underpins our $309.41 target.
Book-value growth from auto margins, not multiple expansion, drove three-year gains
We tested six candidate stock drivers across four empirical tests: event-day reactions over eight earnings releases, 1- to 5-year return decomposition, three-year weekly factor regressions and 164 earnings-call analyst questions. Four company-specific drivers passed at least two independent tests. Notably, book-value-per-share (BVPS) growth added $205/sh against a $133/sh three-year price gain, more than offsetting a 22% compression in price-to-book (from 2.83x to 2.22x).
| Confirmed driver | Tests passed | Market reaction vs model value | Linked debate | Next number to watch |
|---|---|---|---|---|
| Auto combined ratio & cycle depth | 4 of 4 tests (event, return, factor, Q&A) | Market prices 96.0% norm. CR ($266.59) vs our 94.0% ($305.91): +$39.32/sh (+14.7%) gap. | Cycle depth; Auto margin pace | Q4 26 auto underlying CR (early Feb 27; Penmark 91.5%) & Q1 27 rate earn-in (Apr 27). |
| P-L policy growth & distribution | 3 of 4 tests (event, return, Q&A) | 39.6% of call Q&A; +3.3% PIF growth on Q3 26 call (5 Nov: ALL −0.9% vs S&P −4.4% to 7,353.73; 6 Nov: −4.5%). Model: +$1.5/sh Bull 1 / −$2.0/sh Bear 4. | Auto margin pace; Capital return | Q4 26 P-L PIF growth (early Feb 27; watch >3.0% y/y) & NY auto approval (within 12m; no date given). |
| Prior-year reserve releases | 3 of 4 tests (event, return, Q&A) | Q4 25 (+7.5% EP) & Q3 26 (+9.1% EP) drove beats; multi-period reserve value (Q4 26E-FY28E) is $2.21/sh vs zero. | Reserve release fade | Q4 26 auto prior-year reestimates (early Feb 27; Penmark +1.5% of EP vs 0% Street). |
Source: Yahoo Finance daily/weekly returns (2021–2026); Signals transcripts (8 calls); Penmark Research model.
Source: Yahoo Finance daily closes (ALL less S&P 500 on first trading day after release, Feb 2025 to 5 Nov 2026)
Source: Signals filings (30 Sep 2023 common BVPS $47.62 to 30 Sep 2026 $120.23); Yahoo Finance closes (17 Nov 2023 to 18 Nov 2026)
Event-day evidence: post-release reactions
Allstate beat the S&P 500 by +2.8pp to +6.1pp on five of eight post-release call days, including +3.5pp on 5 Nov when shares fell 0.9% ($266.72) as the S&P 500 dropped 4.4% (7,353.73) in a tech-led selloff, before falling 4.5% (−4.8pp rel.) on 6 Nov on auto margin drift. Q2 25 (+6.1pp) saw auto underlying CR fall 3.4pp to 87.8%; Q4 25 (+5.1pp, 5 Feb 2026) printed 72.9% P-L CR with 7.5% releases, whereas $2.2bn of Q1 25 catastrophes moved shares just −1.2pp.
Return decomposition: pure book-value engine
Over one year (+29.0% total return), the P/B multiple barely moved (2.19x to 2.22x); over three years (+107.4%), P/B fell 22% while book value drove the gain. Three-year GAAP BVPS rose from $47.62 (3Q23) to $120.23 (3Q26), adding +$205/sh at 2.83x while P/B compressed to 2.22x (−$72/sh) and dividends added $11.76/sh. Over 5Y (+152.4%), BVPS added +61.5pp and post-2021 re-rating +74.3pp.
Sector and company factors explain 97% of weekly variance; S&P 500 explains just 3%
Source: Signals transcripts across 8 quarterly calls (Q4 2024 to Q3 2026); questions can span multiple topics, so shares sum to 143%
Auto combined ratio & cycle depth (links to Cycle depth & Auto margin pace)
Claim: Sustained underwriting profitability, not temporary catastrophe noise, sets Allstate's valuation. Market reaction: Auto margin and pricing drew 48.8% of 164 call questions; management answered confidently on rate restarts in 39 locations while hedging exact loss-trend timing. Model vs market: Every 1pp of normalised P-L combined ratio moves value by ~$19.20/sh; pricing 96.0% vs our 94.0% undervalues shares by $39.32 (+14.7%). Next watch: Q4 26 auto underlying CR (early Feb 27; Penmark 91.5%).
P-L policy growth & distribution (links to Auto margin pace)
Claim: Policy growth proves Allstate can expand without destroying margin. Market reaction: Policy count and marketing drove 39.6% of analyst questions; following Q3 26 (+3.3% P-L PIF growth), shares fell only 0.9% on 5 Nov ($269.06 to $266.72) as the S&P 500 dropped 4.4% (7,688.62 to 7,353.73, +3.5pp excess) in a tech-led selloff, before a 4.5% drop (−4.8pp rel.) on 6 Nov as investors weighed the higher auto underlying combined ratio. Model vs market: Bull (+4.5% FY27E auto PIF) vs Bear (+1.0%) shifts growth trajectory meaningfully alongside margin. Next watch: Q4 26 P-L PIF (Feb 27) and pending NY auto approval (within 12m; no date given).
Prior-year reserve releases (links to Reserve release fade)
Claim: Auto bodily-injury reserve redundancies create near-term earnings beats even if excluded from terminal value. Market reaction: Reserve questions jumped to 11 of 18 total turns over the last three calls as auto releases hit 7.5%–9.1% of EP; CEO Wilson hedged ("would not model releases"). Model vs market: Our tapered Q4 26E-FY28E releases add $2.21/sh vs zero; near-term Q4 releases and light October catastrophes ($232m) drive our +7.2% FY26E EPS gap vs consensus. Next watch: Q4 26 reserve reestimates (early Feb 27; Penmark +1.5% auto EP).
| Dropped candidate | Why dropped (failed ≥2 tests) |
|---|---|
| Single-quarter catastrophes | Q1 25 ($2.2bn cats) moved stock just −1.2pp rel.; market looks through weather noise to underlying margins. |
| Protection Services | Only 3.0% of call Q&A (5 of 164 turns) and 2% of FY26E adjusted net income; negligible share-price impact. |
| Broad S&P 500 beta | Explains only 3.3% of 3Y weekly variance (R² = 0.033; 5Y beta 0.12–0.45); ALL trades on P&C fundamentals. |
Source: Yahoo Finance weekly returns (n=156 weekly returns, 157 closes, 3Y to 18 Nov 2026); Signals transcripts. Note: sample covers a hard-market upswing; factor betas can rise during market stress.
Personal-lines margins should normalise near 94%, two points better than priced
Source: Allstate 10-K filings (FY17-FY25); Q3 2026 investor supplement
What the market believes
Investors treat FY25–26E's 84%–85% combined ratios as a fleeting windfall that will mean-revert past history. At $266.59, our reverse residual income model shows the market prices a 96.0% normalised Property-Liability combined ratio and 15.7% normalised ROE, 1.6pp worse than Allstate's FY17–25 full-cycle average of 94.4% (which already included the 106.1% FY22 inflation trough) and 2.8pp worse than its 93.2% nine-year median.
Independent evidence (three sources)
- Reported 10-K expense step-down: Allstate's P-L expense ratio dropped structurally by 3.3pp, from a 24.6% average in FY17–19 to 21.4% across FY23–25 (and 21.7% TTM 3Q26), driven by technology consolidation, digital claims and revised agency payouts.
- Q3 2026 operating KPIs & filings: Adjusted underlying auto combined ratio stood at 91.4% in Q3 26 (still comfortably better than management's mid-90s target) while management proactively restarted rate filings across 39 locations ($175m written impact) that earn in across FY27E.
- Homeowners underlying improvement & loss-cost watch: Allstate's homeowners underlying combined ratio improved from 69.6%-71.1% in FY21-22 to 60.7% in 9M26A, cushioning auto give-back even while repair inflation (+2.4% over three months to Oct 2026) remains a live risk.
What it is worth per share
Each 1.0 percentage point on the normalised P-L combined ratio is worth ~$19.20 per share in our residual income model. Holding our base normalised ratio at 94.0% rather than the 96.0% embedded in the share price explains virtually the entire $39.32 per share gap between the $266.59 stock price and our $305.91 base-case value. If the ratio only returns to its FY17-25 average of 94.4% (excluding intangibles), giving back the whole expense gain, base value is still $298 (+12%).
Dated signposts to confirm or disprove
- Early Feb 2027 (Q4 26 results): Confirm auto underlying CR near 91.5% and P-L expense ratio ≤22.0%.
- Late Apr 2027 (Q1 27 results): Verify 39-location auto rate filings begin lifting average earned premium sequentially (our FY27E forecast: +1.5% y/y) while auto underlying CR holds ≤93.5%.
- Aug 2027 (Q2 27) & Feb 2028 (FY27 results): Confirm full-year FY27E P-L adjusted combined ratio lands near 92.4%; thesis breaks if normal-cat combined ratio breaches 95.5%.
Higher yields and retained earnings double adjusted BVPS to $215 as payouts reach $11bn
Source: Company filings (FY24A-FY25A); Penmark Research model (FY26E-FY30E)
What the market believes
Consensus focuses on falling FY27E headline EPS (−27.5% y/y) and under-credits per-share compounding from investment income and buybacks. Because personal-lines earnings step down from FY26E's peak ($40.63 EPS) to FY27E ($29.46), investors apply a trough 9.0x forward P/E, ignoring that Allstate's earnings mix is shifting toward steadier, higher-multiple investment income while retiring roughly 6% of its share count annually.
Independent evidence (three sources)
- Investment portfolio repricing (10-Q & model): With 10-year US Treasury yields at 5.07%, Allstate's $87.4bn investment portfolio (rising to $109.1bn by FY30E) reprices from a 4.30% FY25A yield (excl. PS) to 4.78% in FY26E and 5.00% by FY30E, adding ~$0.3bn per year of recurring net investment income ($4.11bn FY26E to $5.33bn FY30E).
- Deployable holdco capital (Q3 26 call): Management confirmed roughly $9bn of broadly deployable holding-company liquidity (bolstered by $3.1bn of 2025 Health & Benefits sale proceeds) with zero need to inject capital into operating insurance subsidiaries.
- Capital ratio cushion & repurchase cadence: Adjusted equity reaches 56.4% of earned premiums at end-FY26E versus our 55.0% operating target, freeing $4.73bn for FY27E buybacks and $3.62bn in FY28E alongside $1.14bn of annual dividends, shrinking diluted shares from 256.9m in FY26E to 200.4m by FY30E (−22%).
What it is worth per share
Higher investment income contributes roughly $0.50/sh immediately to normalised earnings power, while accretive buybacks lift FY30E EPS back to $38.91. In our indicative value bridge, higher yields offset part of underwriting normalisation so normalised adjusted ROE settles at 18.7%, more than double our 9.21% cost of equity. Even at a conservative 2.03x justified terminal P/B, compounding adjusted BVPS to $214.95 by FY30E supports our $305.91 base value today.
Dated signposts to confirm or disprove
- Early Feb 2027 (Q4 26 release): Board authorises a fresh multi-year share repurchase programme (only $1.3bn remained on the $4.0bn Feb 2026 mandate at 3Q26; Q4 26E consumes ~$1.0bn).
- Quarterly FY27E prints: Confirm net investment yield (excluding Protection Services) tracks 4.85%+ in FY27E and quarterly common buybacks average ≥$1.0bn.
- Disprove trigger: Holdco capital diverted to dilutive M&A, 10-year yields falling below 4.0%, or buybacks paused due to a catastrophe shock.
Reserve releases and 3% policy growth buffer FY26E-27E, though conviction stays Low
Source: Company filings (FY21A-FY25A); Penmark Research model (FY26E-FY28E)
What the market believes
The Street takes management's conservative reserve comment literally, modelling zero prior-year releases immediately, while doubting auto policy growth durability. On the Q3 2026 call, CEO Wilson stated he "would not model" future favourable reserve releases, anchoring consensus at $37.90 for FY26E and $27.58 for FY27E despite nine-month 2026 actual adjusted EPS already reaching $29.41.
Independent evidence (three sources)
- Reserve redundancy momentum (3Q26 supplement): Auto non-cat prior-year releases surged from +0.9% of earned premiums in FY24A to +4.8% in FY25A and +8.2% in 9M26A (+9.1% in 3Q26 alone) as 2022–24 injury reserves settled below initial inflation assumptions.
- October catastrophes ($232m) require a quiet Nov-Dec: October alone used 49% of our $476m Q4 estimate (vs ~$160m/month even pace), so achieving our $11.26 Q4 EPS ($40.63 FY26E, +7.2% vs consensus) needs light late-quarter weather.
- Broad-based policy-growth inflection (3Q26 deck): P-L policies in force rose 3.3% y/y to 39.3m (auto +3.4%, homeowners +3.3%), above underlying US car and home growth, with pending NY auto approval and early-2027 AI sales tools providing fresh upside.
What it is worth & strongest counter-evidence
Near-term releases add $2.21/sh directly versus zero, while our +6.8% FY27E EPS gap ($29.46 vs $27.58) creates positive revision momentum. Crucially, our model fades auto releases cleanly to +1.0% in FY27E, +0.5% in FY28E and 0% from FY29E onward, so terminal value does not rely on reserve gains.
Strongest counter-evidence against our Buy rating: First, cost-of-equity sensitivity makes our Buy fragile. Raising our equity discount rate by just 1pp (from 9.21% to 10.21%) lowers probability-weighted value from $287.58 to $256.24 ($278.24 12m target, +4.4% price upside / +6.1% total return), shifting the rating to Hold and keeping our conviction explicitly Low. Second, Q3 26 auto underlying combined ratio rose 380bp y/y to 89.8% (91.4%, or +540bp, excluding 1H26 claim development) as average earned premium per policy fell 2.7% (total auto earned premiums rose 0.6%); if loss severity outruns 39-location rate restarts, FY28E could repeat the 2022 trough ($205 Bear case).
Dated signposts to confirm or disprove
- Early Feb 2027 (Q4 26 results): Confirm FY26E EPS ≥$39.50 and positive Q4 reserve releases (≥+1.0% auto EP).
- Within 12 months (no date given): NY auto product approval plus sustained ≥3.0% P-L PIF growth without auto underlying CR breaching 94.0%.
Underwriting recovery restored GAAP-to-adjusted alignment and cut leverage to 18%
Over the past five years, Allstate moved through a complete personal-lines cycle, from a 106.1% combined-ratio trough in FY22A to an 84.9% peak in FY25A (excluding intangibles; 106.6% and 85.2% recorded) and 39.7% trailing adjusted ROE in 3Q26A. Earnings quality has strengthened sharply: adjusted net income ($7.62bn in 9M26A) now tracks GAAP net income ($7.72bn) within 1.3%, operating cash flow exceeded $10.1bn in FY25A, and debt-to-total-capital fell from 26.7% in FY24A to 17.9% at 3Q26A.
Source: Allstate 10-K filings (FY21A-FY25A)
| US$m or % | FY21A | FY22A | FY23A | FY24A | FY25A | 3Q26A / 9M26A |
|---|---|---|---|---|---|---|
| P-L premiums earned ($m) | 40,454 | 43,909 | 48,427 | 53,866 | 57,682 | 44,806 (9M) |
| P-L premiums growth y/y | +13.7% | +8.5% | +10.3% | +11.2% | +7.1% | +4.4% (9M) |
| P-L combined ratio (excl. intang.) | 95.3% | 106.1% | 104.0% | 93.9% | 84.9% | 84.8% (9M) |
| Catastrophe losses (% of EP) | 8.3% | 7.1% | 11.6% | 9.2% | 8.6% | 10.3% (9M) |
| Prior-year reserve dev., adv. (+) | +0.3% | +3.9% | +1.2% | −0.5% | −3.1% | −5.5% (TTM) |
| Adjusted ROE (trailing 12m) | 16.9% | −1.3% | 1.5% | 26.8% | 38.3% | 39.7% (3Q) |
| GAAP less adjusted NI ($m) | (2,548) | (1,154) | (567) | (356) | +861 | +99 (9M) |
| Unrealised AOCI, period end ($m) | +598 | (2,253) | (604) | (771) | +297 | (2,024) (3Q) |
| Adjusted equity / P-L EP | 55.9% | 40.4% | 33.8% | 37.5% | 49.1% | 54.3% (TTM) |
| Debt / total adjusted capital | 24.5% | 28.8% | 30.2% | 26.7% | 19.8% | 17.9% (3Q) |
Source: Allstate 10-K filings (FY21A–FY25A); Q3 2026 10-Q and investor supplement; Penmark Research model.
Accounting items that flatter or distort results
Three insurer-specific items require scrutiny: reserve releases, AOCI bond marks and purchased intangibles. First, auto prior-year reserve releases boosted 9M26A underwriting results by 8.2% of auto EP; because reserve releases are cyclical rather than permanent, our model tapers them to zero by FY29E. Second, rising 10-year yields pushed unrealised fixed-income losses (AOCI) to −$2,024m at 3Q26; while adjusted equity excludes AOCI, our residual income valuation deducts the full $2,024m (−$8.03/sh) from equity value. Third, purchased intangibles amortisation adds ~0.3pp to recorded combined ratio ($183m in FY25A) and is cleanly excluded from adjusted earnings.
Cash conversion, reserves & balance-sheet health
Cash conversion and balance-sheet capitalisation remain strong. Operating cash flow reached $8.93bn in FY24A and $10.11bn in FY25A (109% of adjusted net income), while capital expenditures stayed modest at $210m-$228m in FY24A-FY25A ($1.4bn across FY21-25 including platform spend). Allstate holds $87.4bn of investments against $30.3bn GAAP common equity ($32.3bn adjusted) and $7.49bn of long-term debt, leaving adjusted equity at ~54% of trailing premiums and ~$9bn of unencumbered holding-company liquidity.
Premiums compound 6% annually while adjusted EPS troughs near $29 in FY27E
Our explicit forecast runs quarterly through Q4 26E and annually through FY30E using the revenue driver tree established in Exhibit 27. Personal-lines earned premiums grow from $60.1bn in FY26E to $75.8bn by FY30E (+6.0% compound annual growth rate). Underwriting margins step down smoothly from FY26E's 84.4% combined ratio to 92.4% in FY27E and 93.6% in FY30E, pulling adjusted EPS down 27.5% in FY27E before investment income and buybacks restore +6% to +13% annual EPS growth through FY30E.
Source: Penmark Research model (full-model scenario runs as of 18 Nov 2026)
| Key operating driver | FY25A | FY26E | FY27E Base | FY28E Base | FY28E Bear | FY28E Bull | Rationale / vs Street |
|---|---|---|---|---|---|---|---|
| Auto PIF growth (y/y) | +2.3% | +3.4%* | +3.5% | +3.0% | +1.0% (27E) | +4.5% (27E) | 3Q26 +3.4%; NY upside |
| Auto avg. earned premium | n.a. | −2.7%* | +1.5% | +3.0% | n.a. | n.a. | 39-location rate earns in |
| Homeowners PIF growth | +2.5% | +3.3%* | +3.0% | +3.0% | n.a. | n.a. | New bus. +10.2% 3Q26 |
| Homeowners EP growth | +15.0% | +11.6% | +9.2% | +8.2% | n.a. | n.a. | Eases to 6.6% by FY30E |
| Auto underlying CR | 88.1% | 89.6% | 93.0% | 93.5% | 103.6% | 92.0% | Drifts to mid-90s goal |
| Auto reserve releases (+) | +4.8% | +6.5% | +1.0% | +0.5% | −4.2% | +2.0% | 0% from FY29E onward |
| Homeowners cat load | 26.6% | 24.3% | 29.0% | 29.0% | 38.6% | 26.0% | FY21-25 avg. 30%; model 29% |
| Net investment yield (excl. PS) | 4.30% | 4.78% | 4.85% | 4.92% | 4.78% | 5.10% | Reprices toward 5.07% |
| Normalised P-L CR (term.) | n.a. | n.a. | n.a. | 94.0% | 96.5% | 92.5% | Price implies 96.0% |
| Adjusted EPS (Penmark) | $34.83 | $40.63 | $29.46 | $31.32 | n.a. | n.a. | Street $37.90 / $27.58 |
Source: Company filings (FY25A; *3Q26 actual y/y pace); MarketBeat (FY26E) and Zacks Research (FY27E) consensus; Penmark Research model.
Why our FY26E–FY27E EPS sit ~7% above consensus
Three measurable drivers explain our +$2.73/sh FY26E gap and +$1.88/sh FY27E gap versus consensus. First, 9M26A already delivered $29.41 of adjusted EPS and October catastrophes were just $232m, making consensus's implied ~$8.49 Q4 26E estimate overly conservative against our $11.26. Second, rising portfolio yields lift FY27E net investment income by +$340m y/y to $4.45bn. Third, $4.73bn of FY27E share repurchases cut average diluted shares by 6.6% y/y to 239.9m.
Costs, capital intensity & balance-sheet path
Capital generation remains abundant even at normalised margins. Holding adjusted equity at 55.0% of earned premiums from FY27E onward, above the 38%–49% levels of FY24–25, still releases $4.73bn in FY27E and $3.62bn in FY28E for repurchases after paying $1.14bn in common dividends ($4.64 and $4.99 dividends per share, DPS). Debt stays flat at $7.49bn, lowering debt-to-total-capital to 14.6% by FY30E.
Combined ratio (±$38/sh) and cost of equity (+$41/−$33/sh) drive valuation
Valuation sensitivity concentrates in two variables: where the Property-Liability combined ratio settles after FY30E and the discount rate applied to excess returns. Across full-model runs, every 1.0pp shift in the normalised combined ratio changes base equity value by ~$19.20/sh, while a ±1.0pp move in cost of equity swings base value from $272.97 to $346.83, and probability-weighted value from $256.24 (Hold) to $326.53 (Buy).
Source: Penmark Research model full-run sensitivity grid
Source: Penmark Research model sensitivities (base value $305.91)
| Cost of equity / Norm. ROE | 14.7% | 16.7% | 18.7% (Base) | 20.7% | 22.7% |
|---|---|---|---|---|---|
| 7.21% (−2.0pp) | $327.73 | $363.40 | $399.07 | $434.74 | $470.41 |
| 8.21% (−1.0pp) | $286.25 | $316.54 | $346.83 | $377.12 | $407.41 |
| 9.21% (Base COE) | $253.67 | $279.79 | $305.91 | $332.03 | $358.14 |
| 10.21% (+1.0pp) | $227.38 | $250.18 | $272.97 | $295.77 | $318.56 |
| 11.21% (+2.0pp) | $205.69 | $225.79 | $245.88 | $265.98 | $286.07 |
Source: Penmark Research model (Scenarios sensitivity matrix). Green cells exceed $266.59 price; red cells sit below $266.59; blue cell = base case ($305.91).
Why near-term auto shifts matter less than terminal margin
Temporary FY27E–FY30E auto ratio noise moves equity value by only ±$4.57/sh (Exhibit 48). Because terminal residual income accounts for 67.1% of value above opening book, one bad weather season or a two-quarter lag in state rate filings barely alters intrinsic value, provided mid-cycle profitability still settles at 94.0% or better. Sharp market reactions to a single quarter's auto loss tick tend to overshoot when normalised margins hold near 94%.
Why our conviction is explicitly Low
Discount-rate robustness is the binding constraint on our Buy call. While base value stays above $266.59 at a 10.21% cost of equity ($272.97), adding the 30% Bear case pulls probability-weighted value down to $256.24 ($278.24 12-month target: +4.4% price upside / +6.1% total return, a Hold). Because our conservative 1.0x beta override already sits well above Allstate's 0.12 five-year regression beta and 0.48 industry beta, we treat our 9.21% COE as disciplined rather than aggressive, yet still flag Low conviction transparently.
Residual income values Allstate at $306 base; the stock prices a 96% combined ratio
We value Allstate primarily via a Residual Income model on adjusted common book value, a standard method for insurers, and cross-check with peer trailing P/E. Residual income measures earnings generated above shareholders' required cost of equity (COE) and adds their present value to opening book value; enterprise-value multiples (EV/EBITDA) and free-cash-flow discounted cash flow (DCF) are rejected because policyholder float and debt sit inside operating liabilities.
| US$m (except per share / ratios) | Q4 26E | FY27E | FY28E | FY29E | FY30E | Terminal |
|---|---|---|---|---|---|---|
| Opening adjusted common equity | 32,334 | 33,885 | 35,088 | 37,384 | 39,541 | 41,696 |
| Adjusted net income to common | 2,817 | 7,067 | 7,059 | 7,351 | 7,798 | 7,788* |
| Less: equity charge at 9.21% COE | (720) | (3,121) | (3,232) | (3,443) | (3,642) | (3,840) |
| Residual income (excess profit) | 2,097 | 3,946 | 3,827 | 3,908 | 4,157 | 3,948 |
| Annualised return on opening equity | 39.7% | 20.9% | 20.1% | 19.7% | 19.7% | 18.7% norm. |
| Terminal value (0% growth, $m) | n.a. | n.a. | n.a. | n.a. | n.a. | 42,859 |
| Discount factor (mid-period, 9.21%) | 0.989x | 0.936x | 0.857x | 0.785x | 0.719x | 0.719x |
| Present value of residual income | 2,074 | 3,694 | 3,280 | 3,067 | 2,987 | 30,800 |
Source: Penmark Research model. *Terminal net income = 18.68% normalised ROE on $41,696m FY30E closing book ($7,788m); terminal value = $3,948m / 9.21% = $42.9bn ($42,859m), discounted at 0.719x (3.75 yrs mid-year) = $30,800m PV.
Source: Penmark Research residual income model (252.1m diluted shares as of 30 Sep 2026)
Discount rate & terminal-growth definitions
Cost of equity (COE) is the minimum annual return shareholders demand for owning Allstate's risk; we set it at 9.21% using a 5.07% 10-year Treasury yield, 4.14% equity risk premium and a conservative 1.0x beta override (rejecting Allstate's low 0.12 statistical beta). Terminal growth is the perpetual growth rate of excess profit after FY30E, held conservatively at 0%, implying a 2.03x terminal P/B on FY30E adjusted book.
Reverse residual income: what $266.59 implies
A reverse DCF / residual income check solves for the assumption needed to make today's $266.59 price fair value. Holding our explicit forecasts fixed, $266.59 prices either: (1) a 96.0% normalised P-L combined ratio (+2.0pp worse than our 94.0% base and +1.6pp worse than FY17–25 history), (2) a 15.7% normalised ROE (−3.0pp below our 18.7%), or (3) a 10.43% cost of equity (+1.22pp above our 9.21% discount rate).
Peer trailing P/E yields $244 (-20% vs base) by omitting future investment and book growth
| Company (Ticker) | Business mix | Price | Mcap ($bn) | P/B | P/E TTM | Implied TTM ROE | 6M ret. |
|---|---|---|---|---|---|---|---|
| Progressive (PGR) | Personal auto & home | $230.13 | 132.4 | 3.84x | 11.4x | 33.7% | +15.7% |
| Travelers (TRV) | Commercial & personal | $390.62 | 80.3 | 2.77x | 9.6x | 28.8% | +27.1% |
| Chubb (CB) | Global P&C & A&H | $370.20 | 141.7 | 1.94x | 12.6x | 15.4% | +12.1% |
| Hartford (HIG) | Commercial & personal | $135.64 | 36.4 | 1.93x | 7.9x | 24.5% | −0.3% |
| Cincinnati Fin. (CINF) | IA comm. & personal | $188.09 | 29.0 | 1.79x | 10.9x | 16.4% | +11.7% |
| W.R. Berkley (WRB) | Specialty commercial | $77.39 | 28.6 | 2.97x | 15.3x | 19.4% | +14.7% |
| Mercury Gen. (MCY) | CA personal lines | $82.66 | 4.6 | 1.65x | 7.2x | 22.9% | −19.2% |
| US Peer Median (7) | Model peer set | n.a. | 36.4 | 1.94x | 10.9x | 22.9% | +12.1% |
| Allstate (ALL) at price | Personal auto & home | $266.59 | 66.3 | 2.08x* | 5.7x | 36.0% | +24.4% |
| Allstate at $309.41 target | Implied at 12m target | $309.41 | 77.0 | 2.41x* | n.a. | n.a. | n.a. |
Source: Signals peer screen (19 Nov 2026) and Penmark Research model (Peers tab). Implied TTM ROE = P/B ÷ TTM P/E. *On 3Q26 adjusted BVPS ($128.26); reported-equity screen P/B = 2.04x. At our $309.41 target, implied FY27E P/E is 10.5x and model normalised ROE is 18.7%.
Source: Yahoo Finance; Signals peer screen; Penmark Research model
Explaining the 20.3% gap vs Peer TTM P/E cross-check
Applying the 10.9x US peer median trailing P/E to Allstate's normalised TTM EPS ($22.37 to 30 Sep 2026) produces $243.87, 20.3% below our $305.91 base value. The gap exceeds 15% for a clean structural reason: a static multiple on trailing normalised earnings values Allstate's 30 Sep 2026 book with zero credit for four years of 19%–34% ROE compounding, $5.6bn of FY26E retained capital, rising bond yields and a 22% share-count reduction through FY30E. Multiples also understate value when peer TTM earnings still sit near cycle highs.
Target reconciliation to Page 1 & robustness checks
Probability-weighting our full-model runs (30% Bear $205.02, 50% Base $305.91, 20% Bull $365.60) yields $287.58 at 18 Nov 2026 (+7.9% return to value). Rolling forward 12 months at 9.21% cost of equity less the 1.62% dividend yield gives our $309.41 target (+16.1% price return; +17.8% total return including dividends = Buy). Robustness checks stay Buy on historical mean CR ($283.42, +14.4%), industry beta ($385.10) and COE −1pp ($326.53), but flip to Hold at COE +1pp ($256.24; $278.24 12m target, +4.4% price / +6.1% total return), setting Low conviction.
Five dated catalysts resolve our thesis; auto severity and rates are the binding risks
| Date | Catalyst / event | Penmark expectation vs consensus | Driver tested & stock move |
|---|---|---|---|
| Mid-Dec 2026 | Nov 2026 monthly catastrophe release | Oct catastrophes ($232m) used 49% of our $476m Q4 estimate (vs ~$160m/m even pace); needs quiet Nov-Dec. | Catastrophe load; ±1–2% if quiet vs late storm |
| Early Feb 2027 | Q4 26 results & 2027 capital plan | Adj. EPS $11.26 vs ~$8.49 implied consensus (+33%); $1.0bn Q4 buyback; 2027 repurchase pace ≥$4.5bn. | Reserve releases & buybacks; +3% to +5% on beat |
| Late Apr 2027 | Q1 27 results & NY/NJ volume update | Auto underlying CR ≤93.0%; auto PIF growth ≥3.5% y/y; earned pricing stabilises near flat to +1%. | Auto margin give-back; ±3–5% on CR vs PIF |
| Aug 2027 | Q2 27 results & 2027–28 reinsurance | 39-location rate increases earn through P&L; homeowners cat load near 29% normal; yield reaches ~4.85%. | Depth of cycle; largest 2027 inflection point |
| Feb 2028 | FY27 full-year results & FY28 outlook | FY27E adj. EPS $29.46 (+6.8% vs $27.58 consensus); P-L CR near 92%-93% supports our 94.0% normalised floor. | Depth of cycle resolution; re-rating to 2.4x P/B |
Source: Company investor calendar; MarketBeat (FY26E) and Zacks Research (FY27E) consensus; Penmark estimates.
| Indicator | Thesis confirmed if | Thesis broken if |
|---|---|---|
| Auto underlying CR | Stays ≤93.5% through FY27E while PIF grows ≥3.0% y/y | Breaks above 96.0% for two quarters without rate offset |
| Normalised P-L CR | FY27E P-L CR prints ≤93.0% on normal 29% home cat load | Full-year P-L CR trends toward ≥96.0% (price-implied) |
| Investment yield | Portfolio yield expands toward 4.85%–4.92% in FY27E–28E | 10-year US Treasury (UST) drops below 3.5%, stalling portfolio yield at ≤4.4% |
| Capital return | FY27E repurchases ≥$4.0bn; adj. equity stays ≥55% of EP | Buybacks halted by cat losses or credit impairments |
| Cost of equity / rates | 10y UST holds ≤5.2% (COE ≤9.3%); weighted upside >10% | 10y UST jumps ≥6.07% (COE +1pp → Hold at $256.24) |
Source: Penmark Research model and thesis framework.
| Risk factor | Likelihood | Value impact | Mechanism, regulatory / ESG / key-person context & thesis-breaker |
|---|---|---|---|
| Auto severity surge / deep cycle | Medium (30% Bear) | −$101/sh Bear (−$3.5/sh Bear 1) | Bodily injury litigation (social inflation) or parts tariffs outrun 39-location filings; FY28E auto CR jumps to 103.6% and normalised P-L CR hits 96.5% (Bear $205.02). |
| Higher discount rate (COE +1pp) | Medium | −$31/sh weighted | Our Low-conviction binding constraint: a +1.0pp jump in COE to 10.21% cuts weighted value to $256.24 ($278.24 12m target: +4.4% price upside / +6.1% total return) and flips Buy to Hold. |
| Severe weather / cat frequency (environmental, social & governance, ESG) | Medium | −$2.5/sh (Bear 3 cat) | Physical climate volatility (convective storms, wildfires) lifts homeowners cat load from 29% normal toward 38.6% FY23 peak load, eroding equity and buybacks. |
| Regulatory rate-filing caps | Low–Med | −$1.5/sh (Bear 5 rate) | State insurance commissioners in CA, NY or NJ delay or reject rate increases needed to offset loss inflation, trapping margins below target ROE. |
| Reserve reversal & CFO transition | Low–Med | −$1.3/sh (Bear 2 res.) | Auto prior-year releases flip adverse (−4.2% Bear) on older bodily injury (BI) claims; new CFO Christian Lown (Aug 2026) shifts capital-return cadence below $4bn/yr. |
Source: Penmark Research model scenarios and sensitivities; Allstate FY25 10-K risk factors.
Top three asset managers hold 28% of shares; short interest sits at four days to cover
Allstate has a liquid, institutional, widely distributed shareholder base with no controlling shareholder, no lock-up overhang and modest short interest. As of 30 Sep 2026, 13F institutional filings accounted for 201.1m shares (81.3% of basic shares outstanding, slightly down from 84.0% in 2Q26 as buybacks retired stock). The top three asset managers, Vanguard (13.2%), BlackRock (9.4%) and State Street (5.1%), collectively hold 27.7%.
Source: Signals Form 13F filings as of 30 Sep 2026
Source: Signals FINRA short-interest series (Aug-Oct 2026)
| Trading & ownership metric | Value | Trading & ownership metric | Value |
|---|---|---|---|
| Primary listing & ticker | NYSE: ALL | Share price (18 Nov / 19 Nov) | $266.59 / $266.76 |
| Basic shares (cover)* / float | 247.45m / 245.92m | Market cap (18 Nov; 248.8m sh)* | $66.33bn |
| 52-week trading range | $191.13 – $281.11 | 50-day / 200-day moving avg. | $247.26 / $233.05 |
| 3-month avg. daily volume | 1.89m sh (~$504m/d) | Major index membership | S&P 500, S&P 100, KIE |
| Form 13F institutional ownership (3Q26) | 81.3% (201.1m sh) | Insider group ownership (Mar 26) | 1.55% (4.01m eq. sh) |
| FINRA short interest (30 Oct 26) | 6.57m sh (2.66% sh) | Days to cover (DTC) / 2W change | 4.03 days / +7.5% |
| Forward dividend / ex-div date | $4.32 (1.62%) / 30 Nov 26 | Lock-ups / dual-class / overhang | None (1 share = 1 vote) |
Source: Signals 13F filings (30 Sep 2026), FINRA short interest (30 Oct 2026), DEF 14A (Apr 2026), Yahoo Finance (19 Nov 2026). *247.45m is the later Q3 10-Q cover-page count; 248.8m is the 30 Sep balance-sheet count used in our model market cap.
Short interest & market positioning
Short interest declined from 9.0m shares in late May to oscillate tightly between 6.0m and 7.0m shares (~2.4%–2.8% of shares) across August–October 2026, signalling little crowded bearish conviction. At 6.57m shares on 30 Oct 2026 (4.03 days to cover), bears are expressing cycle-peak caution primarily through lower valuation multiples (9.0x FY27E P/E) rather than aggressive short selling. Liquidity is deep at ~$504m traded daily, easily absorbing Allstate's ~$15m–$20m daily buyback flow.
Insider holdings & Form 4 activity context
Directors and officers collectively hold 1.55% of stock-based equity (4.01m shares including options/RSUs), led by CEO Thomas Wilson's 1.07% stake (2.77m total equity units). Following the Q3 2026 earnings window, Wilson sold ~174k shares ($45.2m) across 6–10 Nov 2026 at ~$260/sh and EVP Christine DeBiase sold ~28.7k shares ($7.5m). Without filing context on option expiries or personal diversification, these transactions should be noted neutrally rather than read as a bearish operating signal, given Wilson's 128.9x salary holding.
Full explicit model: underwriting normalises while BVPS compounds to $215 by FY30E
| US$m (except per share / %) | FY24A | FY25A | FY26E | FY27E | FY28E | FY29E | FY30E |
|---|---|---|---|---|---|---|---|
| Auto premiums earned | 36,475 | 38,090 | 38,544 | 40,491 | 42,957 | 45,352 | 47,648 |
| Homeowners premiums earned | 13,360 | 15,363 | 17,151 | 18,725 | 20,251 | 21,588 | 23,012 |
| Other P-L premiums earned | 4,031 | 4,229 | 4,403 | 4,579 | 4,762 | 4,953 | 5,151 |
| Total P-L premiums earned | 53,866 | 57,682 | 60,097 | 63,795 | 67,971 | 71,893 | 75,811 |
| P-L premiums growth y/y | +11.2% | +7.1% | +4.2% | +6.2% | +6.5% | +5.8% | +5.5% |
| Auto adjusted combined ratio | 94.7% | 84.7% | 84.7%* | 93.6% | 94.6% | 95.1% | 95.1% |
| Homeowners adjusted combined ratio | 89.8% | 84.1% | 82.5%* | 90.0% | 91.0% | 91.0% | 91.0% |
| Total P-L adjusted combined ratio | 93.9% | 84.9% | 84.4% | 92.4% | 93.3% | 93.7% | 93.6% |
| P-L underwriting income, adjusted | 3,286 | 8,723 | 9,393 | 4,830 | 4,523 | 4,561 | 4,818 |
| Net investment income (excl. PS) | n.a. | 3,350 | 4,109 | 4,449 | 4,726 | 5,038 | 5,325 |
| Corporate pre-tax costs (excl. NII) | n.a. | n.a. | (460) | (470) | (475) | (480) | (485) |
| Tax on operations (21.2% eff. rate) | n.a. | n.a. | (2,765) | (1,868) | (1,860) | (1,933) | (2,048) |
| Protection Services adj. net income | 217 | 218 | 224 | 242 | 261 | 282 | 305 |
| Preferred stock dividends | (117) | (117) | (117) | (117) | (117) | (117) | (117) |
| Adjusted net income to common | 4,906 | 9,304 | 10,437 | 7,067 | 7,059 | 7,351 | 7,798 |
| Weighted diluted shares (m) | 267.8 | 267.1 | 256.9 | 239.9 | 225.4 | 213.1 | 200.4 |
| Adjusted diluted EPS (US$) | $18.32 | $34.83 | $40.63 | $29.46 | $31.32 | $34.50 | $38.91 |
| Adjusted ROE (avg. adj. equity) | 26.8% | 38.3% | 33.6% | 20.5% | 19.5% | 19.1% | 19.2% |
| Closing adjusted common equity | 20,212 | 28,312 | 33,885 | 35,088 | 37,384 | 39,541 | 41,696 |
| Adjusted BVPS (excl. unrealised) | $75.25 | $107.30 | $136.43 | $151.62 | $170.44 | $191.16 | $214.95 |
| GAAP BVPS (AOCI held at 3Q26) | $72.35 | $108.45 | $128.28 | $142.87 | $161.21 | $181.38 | $204.52 |
| Common dividends per share (DPS) | $3.68 | $4.00 | $4.32 | $4.64 | $4.99 | $5.37 | $5.77 |
| Share repurchases ($m) | 2 | 1,233 | 3,770† | 4,726 | 3,624 | 4,035 | 4,469 |
| Total cash returned to common ($m) | 964 | 2,269 | 4,865 | 5,864 | 4,762 | 5,194 | 5,643 |
| Total investments, period end | 72,610 | 83,237 | 89,317 | 93,331 | 98,800 | 103,938 | 109,072 |
| Debt / total adjusted capital | 26.7% | 19.8% | 17.3% | 16.8% | 16.0% | 15.3% | 14.6% |
Source: Company filings (FY24A–FY25A); Penmark Research model (FY26E–FY30E). *Implied full-year from 9M26A + Q4 26E. †Incl. other equity movements. NII shown excl. Protection Services (FY24A not separately reported). FY26E operating lines sum to $10,384m ($53m below $10,437m adj. NI) because 9M26A reported NI exceeded the 9M build by $53m (health legacy and tax).
| Quarterly bridge & segment split ($m) | 9M26A | Q4 26E | FY26E | Segment adj. NI mix | FY25A | FY27E | FY30E |
|---|---|---|---|---|---|---|---|
| Auto premiums earned | 28,845 | 9,699 | 38,544 | Property-Liability & Inv. | 9,086 | 6,825 | 7,494 |
| Homeowners premiums earned | 12,690 | 4,461 | 17,151 | Protection Services | 218 | 242 | 305 |
| Total P-L premiums earned | 44,806 | 15,291 | 60,097 | Total Adj. NI to common | 9,304 | 7,067 | 7,798 |
| P-L adjusted combined ratio | 84.8% | 83.1% | 84.4% | Protection Services share | 2.3% | 3.4% | 3.9% |
| Adjusted net income / EPS | 7,620 / $29.41 | 2,817 / $11.26 | 10,437 / $40.63 | Adjusted equity / P-L EP | 49.1% | 55.0% | 55.0% |
Source: Q3 2026 10-Q and investor supplement; Penmark Research model. Note: 9M26A ($7,620m) + Q4 26E ($2,817m) = $10,437m; quarterly EPS sums to $40.67 vs $40.63 full-year diluted EPS on weighted shares.
Model architecture & capital-return mechanics
Capital return is set to hold adjusted equity at 55.0% of earned premiums from FY27E onward. Holding equity at 55.0% of P-L earned premiums, above FY24A (37.5%) and FY25A (49.1%), preserves conservative statutory buffers while freeing $4.73bn in FY27E and $3.62bn–$4.47bn annually in FY28E–FY30E for share repurchases after common dividends grow 7.5% per year.
Share-count reduction & terminal-year bridge
Buybacks executed along a rising forward equity curve retire 58.1m ending shares between Q4 26E and FY30E. Retained earnings lift adjusted BVPS from $136.43 in FY26E to $214.95 in FY30E (+12.0% CAGR), while buybacks cut weighted diluted shares 22.0% (from 256.9m to 200.4m) even as underwriting normalises to a 93.6% combined ratio in FY30E and 94.0% in terminal perpetuity.
GAAP reconciliation, insurance definitions and Penmark methodology summary
| Earnings & book-value reconciliation (US$m or $/sh) | FY24A | FY25A | 9M26A / 3Q26 | Treatment in Penmark model |
|---|---|---|---|---|
| GAAP net income applicable to common | 4,550 | 10,165 | 7,719 | Starting point from 10-K / 10-Q |
| Net (gains) / losses on investments & derivatives | +225 | +168 | (105) | Excluded; market-timing noise |
| Pension & postretirement remeasurement | (37) | (35) | (155) | Excluded; non-operating actuarial mark |
| Amortisation of purchased intangibles | +280 | +231 | +139 | Excluded; non-cash acquisition item |
| (Gain) / loss on disposition of operations | (16) | (1,616) | (6) | Excluded; FY25 Health & Benefits sale gain |
| Income tax on non-operating adjustments | (96) | +391 | +28 | Tax effect of reconciling items |
| Adjusted net income applicable to common | 4,906 | 9,304 | 7,620 | Core operating earnings base |
| GAAP common shareholders' equity (period end) | 19,441 | 28,609 | 30,310 | Includes fixed-income mark-to-market |
| Less: unrealised net capital gains / (losses) (AOCI) | (771) | +297 | (2,024) | Excluded from ROE; deducted in full from value |
| Adjusted common shareholders' equity | 20,212 | 28,312 | 32,334 | Opening book for Residual Income model |
| GAAP BVPS / Adjusted BVPS (per diluted share) | $72.35 / $75.25 | $108.45 / $107.30 | $120.23 / $128.26 | AOCI gap = −$8.03/sh at 30 Sep 2026 |
Source: Allstate FY2025 10-K, Q3 2026 10-Q and Q3 2026 Investor Supplement; Penmark Research model.
Glossary & insurance ratio definitions
Combined ratio (CR): Claims incurred plus underwriting expenses divided by earned premiums. Below 100% indicates an underwriting profit; above 100% an underwriting loss.
Underlying combined ratio: Combined ratio excluding catastrophe losses, prior-year reserve reestimates and purchased-intangibles amortisation, the cleanest gauge of ongoing pricing power.
Policies in force (PIF): Active insurance contracts at period end; Allstate's volume unit (39.3m in Property-Liability, 177.8m in Protection Services at 3Q26).
Prior-year reserve development (PYD): Adjustments to estimated claims from prior accident years. Favourable releases lower reported CR; adverse development raises it.
Adjusted book value per share (Adj. BVPS): Common equity excluding unrealised fixed-income gains/losses (AOCI), divided by closing diluted shares ($128.26 at 3Q26).
Residual income: Adjusted net income minus an equity capital charge (opening adjusted equity × cost of equity). Positive residual income justifies a P/B multiple above 1.0x.
Catastrophe (cat) load: Weather and disaster losses expressed as a percentage of earned premiums (normalised at 29.0% for homeowners and 1.6% for auto).
Social inflation: Rising bodily-injury claim severity driven by attorney representation, litigation funding and nuclear jury verdicts rather than general CPI.
Methodology note, peer set & data cut-off
Business model: Personal-lines Property & Casualty insurer (Property-Liability = ~98% of FY26E adjusted earnings; Protection Services = ~2%). In line with Penmark insurance standards, enterprise-value multiples (EV/EBITDA, EV/FCF) and free-cash-flow DCF are excluded because policyholder reserves and float are operating liabilities.
Valuation & conviction framework: Primary valuation uses Residual Income on adjusted book (explicit Q4 26E–FY30E, 9.21% COE with 1.0x beta override, 0% terminal residual-income growth, full AOCI deduction). Cross-check applies the 10.9x US peer median TTM P/E to normalised TTM EPS ($22.37). Conviction is flagged Low because a +1.0pp COE stress shifts the rating from Buy to Hold. Ratings follow the Penmark total-return rule on page 28 (12-month target upside plus expected dividends: Buy >+10%, Hold −10% to +10%, Sell <−10%).
Peer set selection: Seven US P&C peers (PGR, TRV, CB, HIG, CINF, WRB, MCY) matched on personal and commercial P&C underwriting and capital structure (Exhibit 53).
Data sources & cut-off dates: Valuation model, forecasts and share price ($266.59) are as of 18 Nov 2026; Signals peer screen and Yahoo Finance trading snapshots as of 19 Nov 2026; SEC filings through Q3 2026 10-Q (4 Nov 2026) and October catastrophe 8-K (19 Nov 2026); FINRA short interest as of 30 Oct 2026; 13F ownership as of 30 Sep 2026; industry market shares from NAIC/AM Best 2025 compilations.
Analyst certification and important disclosures
Ratings
Ratings are based on the total return from the share price to our 12-month price target, including dividends. The price target is weighted across bull, base and bear scenarios.
| Buy | Total return to price target above +10% |
| Hold | Total return to price target from −10% to +10%, inclusive |
| Sell | Total return to price target below −10% |
We review ratings at each set of results and whenever the facts change materially. A share price can move outside its band before the rating changes, and targets may not be reached.
Analyst certification
The Penmark Team certifies that (1) the views in this report accurately reflect its views about the subject company and its securities, and (2) no part of its compensation was, is or will be directly or indirectly related to the specific ratings, price targets or views in this report.
Contact
Penmark Research
Questions about our research:
research@penmarkresearch.com
All our research and coverage:
penmarkresearch.com
Basis of preparation
This report uses information publicly available up to its publication date. Share prices are closing prices on the trading day before the publication date, unless stated otherwise. Forecasts, valuations and price targets are Penmark Research’s own, and forecast periods are marked E. Sources are shown beneath each exhibit.
Valuation and price targets
Price targets are based on the valuation methods set out in each report and are weighted across bull, base and bear scenarios. The main risks to each rating and price target are described in the report. Penmark Research updates its coverage after each set of results but is not obliged to update a report at any other time.
Independence and conflicts of interest
Penmark Research is independent. It does not provide investment banking, brokerage, market-making or asset management services, does not trade securities, and does not accept payment from any company for research coverage. Companies do not see or review Penmark Research reports before publication.
No member of the Penmark Team holds a long or short position, directly or through derivatives, in the securities of any company Penmark Research covers.
Regulatory status
Penmark Research publishes general, impersonal investment research on a regular basis. It is not registered as an investment adviser with the US Securities and Exchange Commission or any state securities authority, is not a registered broker-dealer and is not a member of FINRA. Penmark Research does not give advice tailored to any person’s circumstances.
Not investment advice or an offer
This report is for information only. It is not personalised investment advice and is not an offer or solicitation to buy or sell any security or other financial instrument. It does not take into account any reader’s objectives, financial situation or needs. Readers should make their own investment decisions and seek independent professional advice where appropriate.
Forward-looking statements and risk
Forecasts, valuations and price targets are estimates based on assumptions that may prove wrong, and they may not be achieved. Opinions are current as of the publication date and may change without notice. Investing in equities involves risk, including the loss of capital. Share prices can be volatile, and past performance is not a reliable indicator of future results.
Data and sources
Company financials, operating data and filings come from company filings, earnings releases and earnings calls. Market data and any consensus estimates come from third-party sources named beneath the relevant exhibit; consensus figures from aggregators are unofficial. Penmark Research does not use other brokers’ research. Third-party data belong to their providers and may not be extracted from this report.
Accuracy and liability
Information is taken from sources believed to be reliable, but Penmark Research makes no representation or warranty as to its accuracy, completeness or timeliness. To the fullest extent permitted by law, Penmark Research accepts no liability for any loss arising from use of this report. Descriptions of litigation and regulatory matters reflect allegations and public records, not findings, unless stated. If a published report is corrected, the corrected version carries an “Updated” date at penmarkresearch.com.
Distribution and access
Penmark Research reports are available to registered users of penmarkresearch.com. Downloads and programmatic access are governed by the Penmark Research subscription terms. This report is not directed at, or intended for, any person in Singapore or India, and access from those countries is blocked. It is not intended for distribution to, or use by, any person in any jurisdiction where that would be contrary to law or regulation or would subject Penmark Research to registration or licensing requirements there.
Copyright and permitted use
© Penmark Research. All rights reserved. Penmark Research is a trade name of Brightriver Labs. Readers may share a link to this report but may not copy, reproduce, redistribute, sell or use its content, including for text and data mining, without written consent. Contact: research@penmarkresearch.com.