LNCLincoln National Corporation

LNC — dossier, 2026-09-23

Slot names-insurance-2b, a one-name pass. Model-drafted unattended by the cloud research routine; nothing here is human-verified. Every figure carries the filing it came from. Dated file — never edited after today.

Price context: $41.01 close 2026-09-22 (ROIC NYSE:LNC). ADV over the 21 sessions to 2026-09-22: $85.6m/day, 1,987,257 shares.

---

1. What this pass found

Three facts that are not in the cluster files and change how the name reads.

1.1 The capital structure was rebuilt in three months, and the equity story is a swap

• June 2026. $500,000,000 of 6.800% Fixed-to-Fixed Reset Rate Subordinated Notes due 2056, ~$493m net. The prospectus says the money is for "which may include the repurchase and/or redemption of shares of our outstanding 9.250% Fixed Rate Reset Non-Cumulative Preferred Stock, Series C" (424B2 filed 2026-06-26, accession 0001193125-26-285466, Use of Proceeds, page S-12). The notes let LNC defer interest for up to five years.
• 10 August 2026. Issuer tender for up to $500m of liquidation preference across the 9.250% Series C and 9.000% Series D depositary shares (Schedule TO, accession 0001193125-26-341416). Wachtell on the cover.
• 9 September 2026. Results: $265,945,000 of Series C at $1,055.00 per $1,000, and $171,355,350 of Series D at $26.30 per $25, for ~$461.8m of cash, settling 2026-09-10. "The shares accepted for purchase represent approximately 53.2% of the Company's issued and outstanding Series C Depositary Shares and approximately 34.3% of the Company's issued and outstanding Series D Depositary Shares as of the date hereof." (exhibit (a)(5)(ii) to the Schedule TO/A, accession 0001193125-26-385942).

Read as arithmetic: $437.3m of 9%-ish perpetual preferred equity replaced with $500m of 6.8% deferrable subordinated debt, at a 5.2-5.5% premium to par. Preferred dividends fall by about $40m a year (Series C 9.25% x $265.9m = $24.6m; Series D 9.00% x $171.4m = $15.4m); the notes cost $34m a year pre-tax, roughly $27m after tax at 21%. Net benefit to the common is on the order of $13m a year, about $0.07 a share, for ~$25m of premium paid and a permanent step down in the quality of the capital. The 2026-07-30 release calls it "Completed $500 million subordinated debt issuance with proceeds supporting capital flexibility to repurchase and/or redeem outstanding preferred stock."

The leverage ratio was 25.1% at 2026-06-30 against a 25% target (Q2 earnings supplement, page 9) before the tender settled. That ratio is the first thing to check in the Q3 release.

1.2 Bain is the largest holder, the buyback is its exit door, and the door has a floor above the market

• 5 June 2025. LNC sold Bain Capital Prairie, LLC 18,759,497 new common shares at $44.00, $825,417,868, "based on a 25% premium to the 30-day volume-weighted average price as of April 8, 2025" (Q2 2026 10-Q, Issuance of Common Stock note). Bain part-funded it with $550,000,000 borrowed from SMBC, secured on the equity interest that holds the LNC shares (13D, accession 0000950170-25-085594, Item 4). $800m of the $825m went straight into LNL as a capital contribution (10-Q, Subsidiaries' Capital).
• Bain got a board observer, convertible from 2026-06-05 into a voting director (the "Prairie Designee"), approval rights over charter changes, liquidation, deregistration and delisting, and a standstill below 10%.
• 10 August 2026. Bain elected under Section 6.2 of the purchase agreement to sell shares back to the company as it repurchases, "to the extent necessary to prevent the Reporting Person from beneficially owning 10% or more of the voting securities of the Company, subject to a floor price equal to the Reporting Person's initial acquisition price of the Common Stock." (13D/A No. 1, accession 0001193125-26-347363, Item 4.)

The floor is $44.00. LNC closed at $41.01 on 2026-09-22 — 6.8% below it, and it has been below it every session since 2026-09-14.

The mechanical consequence, which no filing states and which the sell-side model is unlikely to carry: Bain holds 18,759,497 of 191,425,371 shares, 9.80%. Shares outstanding have to fall to about 187.6m before Bain crosses 10%. That is roughly 3.8m shares, ~$156m at $41.01, of buyback that costs nothing extra. Beyond that point, every further share LNC retires forces it to buy Bain's stock at $44.00 or above while its own stock trades lower. The authorisation is $714,000,000 remaining at 2026-06-30 (Equibles, from the 10-Q) — so about a fifth of it is unencumbered and four-fifths of it runs into the floor at today's price. And LNC has not repurchased a common share since Q3 2022: $0 in 2023, $0 in 2024.

1.3 The cluster's kill criterion for LNC is measured on a number the 10-K itself qualifies

_clusters/cluster-insurance-2026-09-22.md §5 sets LNC's kill criterion as "recoverables/equity rising above 3.0x", and the re-score scored capital 0 on $28,012m of reinsurance recoverables against $10,906m of equity, 2.57x. Both are gross.

The same 10-K says, in the MD&A's Reinsurance section:

"As of December 31, 2025, 84%, or $23.5 billion, of our total reinsurance recoverable was secured by collateral for our benefit."

with $22.0bn in reserve credit trusts held by non-affiliated reinsurers, $1.4bn in funds withheld and $142m of LOCs. Uncollateralised recoverable is therefore about $4.5bn against $10,906m of equity — 41%, not 257%. At 2026-06-30 gross recoverables are $27,225m against $11,349m of equity, 2.40x, i.e. falling.

That does not make the counterparty exposure trivial: LNC "ceded $1.2 trillion of life insurance in force to reinsurers for reinsurance protection", the collateral share is management's own number and is not audited as a covenant, and trust assets are marked. But a kill criterion set at 3.0x of a gross number that is already falling, against a collateralised net number a quarter of the size, is not a test that can fire for the reason the cluster intended. It is replaced in the scorecard.

---

2. S2 — kill test

All XBRL from data.sec.gov/api/xbrl/companyfacts/CIK0000059558.json, 10-K and 10-Q values only.

Survives, and the solvency question is not close. RBC is ">420%" in every quarter of the Q2 2026 release's capital table, and the proxy discloses the real number: "During 2025, the RBC ratio increased from 433% to 460%, inclusive of the remaining proceeds from the Bain Capital transaction earmarked for future deployment." Holding-company available liquidity is $1,803m gross, $903m net of the $400m debt prefunding and the $500m preferred prefunding. Total debt $6,465m at 2026-06-30, leverage 25.1%.

Three things the kill test does flag.

1. Operating cash flow has been negative in three of the last four years (−$2,074m, −$2,007m, −$167m, and −$420m in H1 2026 against +$741m in H1 2025). For a life and annuity writer this is less damning than it looks — deposit-type contract flows sit in financing, not operating — but the direction is the wrong one while the company is buying back preferred with borrowed money.
2. The share count is the story. 169.7m outstanding at end-2023 to 191,425,371 at 2026-06-30, +12.8%, of which 18.76m is the single Bain issue. Diluted weighted average 180.6m (Q2 2025) to 196.4m (Q2 2026), +8.8%.
3. Adjusted operating EPS is going backwards while GAAP EPS looks spectacular. Q2 2026 GAAP diluted EPS $6.72 against adjusted operating EPS $2.24 — and $2.24 is down from $2.36 a year earlier. H1: $3.89 against $3.97. Adjusted operating income available to common actually rose, $427m to $439m; the per-share line fell because of the extra shares. The gap between the two EPS numbers is $1,497m pre-tax of "net annuity product features" in the quarter — market risk benefit marks, which the company itself calls "the non-economic impact of changes in market risk benefits".

Analyst coverage count: not obtained (no source in this environment).

---

3. S3 — filings, at gate-minimum

Read: FY2025 10-K Item 1A in management's own ordering, the MD&A Executive Summary, Significant Operational Matters and Reinsurance sections, and the EY report; the Q2 2026 10-Q MD&A, segment tables, equity note and liquidity section; the 2026-04-16 DEF 14A ownership and compensation sections; the Q2 2026 earnings release and the Q2 2026 earnings supplement; the June 424B2; the 2026-08-10 8-K; the Schedule TO and its results amendment; the two Bain 13Ds; four Form 4s; and a full efts.sec.gov red-flag sweep.

3.1 The three biggest risks management itself discloses

Item 1A opens with "Market Conditions" and these are its first three headings, in the company's order:

1. > "Weak conditions in the global capital markets and the economy generally may materially > adversely affect our business and results of operations."
2. > "Changes in interest rates and sustained low interest rates may cause interest rate > spreads to decrease, impacting our profitability, and make it more challenging to meet > certain statutory requirements."
3. > "Increases in interest rates and sustained higher interest rates may negatively affect our > profitability, capital position and the value of our investment portfolio and may also > result in increased contract withdrawals and surrenders."

All three located in FY2025 10-K, Part I Item 1A, "Market Conditions". The third is the one with a live number attached in the same paragraph: "during 2024 and 2025, our Annuities business experienced an increased outflow rate primarily due to an increase in full surrenders as a result of the elevated interest rate environment and strong equity markets."

The fourth, and the one this repo's cluster thesis rests on, is under "Operational Matters": "We face risks of non-collectability of reinsurance and increased reinsurance rates, which could materially affect our results of operations." — §1.3 above is the reading of it.

Gate: met. Three risks, management's words, locators given.

3.2 Auditor and red-flag sweep

• Ernst & Young LLP, PCAOB ID 42, unqualified on the FY2025 statements and on internal control, report dated 2026-02-19. Two critical audit matters: the $42.1bn future contract benefits liability (universal life with secondary guarantees; investment margin, mortality and lapse assumptions) and market risk benefits ($4.8bn asset, $1.1bn liability).
• efts.sec.gov full text for CIK 0000059558: zero hits for "changes in accountants"; no NT 10-K in the index; 30 hits for "material weakness", all in 2023 filings or older; 8 for "non-reliance".
• The one that matters is history with a live rhyme. The 2023-03-27 8-K carries Items 2.02, 4.02 and 7.01. It says of the 2021 Resolution Life coinsurance transaction: "The Company has determined that the realized gain should have been recognized at the time of the transfer of the assets" — restating ~$498m of one-time gain into 4Q21 and adding ~$467m to 2022 year-end equity. LNC's only non-reliance event on the public record is about reinsurance transaction accounting, and reinsurance transaction accounting is exactly the lever management says it is pulling next: LPINE affiliate reinsurance, the fourth-quarter 2025 captive restructuring, and "looking at an external risk transfer deal" in 2026-2027.

3.3 Compensation — what the bonus is paid on

2025 AIP, corporate financial measures 75% for the CEO:

CEO payout 147.6% of a $3,630,000 target, $5,357,880. Two things are worth stating plainly.

• The 45% metric is a per-share metric, calculated "in accordance with the terms of the 2025 AIP" with "certain defined exclusions" listed in an exhibit. It still certified at $8.72 in the year the company issued 18.76m new shares.
• The 15% DE Actions metric is certified through the CFO. "each management action to increase distributable earnings was reviewed by the CFO as meeting the criteria established by the Committee for inclusion in the DE Actions goal based on its impact to increase the Company's distributable earnings and RBC ratio." It paid 175%, and the committee's own explanation of why is the RBC sentence quoted in §2 — which credits "the remaining proceeds from the Bain Capital transaction". An equity raise that diluted the 45% metric helped the 15% one. The CFO who signed off on that metric gave notice five months later.

3.4 The transcript — Q1 FY2026, read Q&A-first

Only one transcript was read, not two. Equibles has no transcript for the Q2 FY2026 call of 2026-07-30 (ListInvestorEvents reports the event with transcript "not available"), and fool.com has no LNC Q2 2026 page. The Q2 earnings supplement was read from EDGAR instead, which is primary but is not a call.

Speaker resolution is unreliable in this source and the repo has a standing warning about it: in this transcript only "Suneet Kamath (Analyst — Jefferies)" is resolved; the CEO and CFO appear as "Speaker 2" and "Speaker 3". Every quote below is given as a located span with the speaker label as the source printed it, not as an attribution to a named person.

Guidance and capital changes (the CFO section): holding company cash "increased to over $800 million, net of the prefunding for the quarter. It's the highest it's been in a long time"; alternative investments "about $4.2 billion in size", "over 400 funds", "about 10% returns with relatively minimal volatility over the past couple of years", explicitly on a lag; earnings growth "came through at 16%".

What analysts asked about, by topic: free cash flow conversion and how much of holdco cash is real (Wells Fargo); alternatives mark lag into Q2 (Wells Fargo); the disability loss ratio and PFML (KBW); the annuities ROA guide of 66-70bp (KBW); the other-operations loss (Dowling); fixed annuity sales mix, FIA versus MYGA (Dowling); whether the industry's first merger of equals in twenty years forces consolidation (Jefferies); how much of the Bain money has been spent (Jefferies); Apollo's "irrational competition" comment (Evercore); the large-case group lapse (Evercore). Seven of ten questions are about whether the earnings are durable; two are about scale; one is about the balance sheet.

Dodges, marked:

1. PFML sizing. Analyst asks to "help us size the PFML impact on either in dollars or on the disability loss ratio". Answer: "we're not going to size the specific impact", then a discussion of claim-resolution normalisation. A stated refusal, not an evasion, but the number was not given.
2. Bain deployment — the real one. Analyst asks "can you give us a sense of how much of that has been deployed at this point?". The answer opens "I don't know that I would add a lot to what we said at the fourth quarter" and then restates the three uses without a figure, closing "we like the trajectory of the deployment". Fifteen months after the money arrived, the amount deployed is not on the public record.
3. Competitive conduct. Analyst quotes Apollo on "irrational competition in the annuity market coming from incumbent insurers". Answer: "I obviously can't speak to the comment of another peer", then a pivot to product differentiation. Partial.

Nothing in the dodge log before today — this is the first pass on the name.

---

4. S5 — ownership

13F by manager at 2026-06-30, from 3spread. The default resolution was wrong and the run had to override it — see §6.

⚠️ 413 filers this quarter against 609 last — the quarter is not fully ingested. Both Vanguard entities, Dimensional, Morgan Stanley, Geode, First Trust, Northern Trust, UBS and Jupiter have no 2026-06-30 filing in the stream; no exit was read from an absence. The Vanguard disappearance has a documented cause and this proxy confirms it for a third issuer after APH and ALL: "The Vanguard Group subsequently reported that due to an internal realignment it no longer has, or is deemed to have, beneficial ownership over Company securities beneficially owned by various Vanguard subsidiaries and/or business divisions."

⚠️⚠️ The Sixth Street / Charles Schwab Investment Management 13F defect appears for the eighth issuer: both report exactly 2,193,476 shares / $77.5m, Sixth Street flagged NEW. Discarded per the standing rule, not carried as a holding.

13D/G. Bain's is the only 13D on the common. The 2026-04-20 and 2026-05-18 Schedule 13Ds in the feed are LNC itself reporting stakes in the Lincoln Bain Capital Total Credit Fund (5,202,624.72 shares, 5,237,272.66 at the amendment) and the Lincoln Partners Group Royalty Fund (9,787,323.12) — i.e. LNC seeding Bain-managed and affiliated credit vehicles. The same counterparty sits on both sides of the relationship: Bain is the anchor shareholder and a manager of Lincoln money.

Insider ownership: 0.64%. All 20 directors and executive officers together held 1,233,827 common shares at 2026-03-20, 1,974,460 including phantom units (1.03%), with 666,064 of the common being options exercisable within 60 days. The CEO holds 672,832 including 362,342 options. No individual is at 1%.

Insider transactions, 12 months. 171 Form 4 filings. Four read in detail:

• 2026-08-12 — Gary C. Kelly (director). Sold the family trust's entire 15,006.550 Series D depositary shares at a weighted $26.50, "prices ranging from $26.50 to $26.54, inclusive", shares owned after: 0. The proxy says he was the only director or officer holding any LNC preferred. He sold two days after the company launched a tender for that exact security at $26.30 and did not tender into it. Entirely proper; also a precise reading of the company's own offer by the one insider who could take it.
• 2026-08-17 — Sean Woodroffe (EVP). Sold 11,000 common at $45.28, 119,947 after.
• 2026-08-24 — Adam M. Cohen (interim CFO). Two awards, 7,866 and 23,597 shares, holding 71,375 after — the retention and LTIP grants the 2026-08-10 8-K described.
• 2026-08-17 — Curtis Chesney (EVP, President of Annuities). Gift of 637 shares.

Five Form 144s since 2026-03-27 (2026-06-01, 06-05, 08-17, 08-18, 09-03). Transaction codes were not itemised across all 171 filings — the 3spread endpoint returns them only per filing.

Short interest (FINRA, via Equibles):

+33% in one settlement, to 7.2% of shares outstanding and 7.6 days to cover, across exactly the window that contains the CFO's notice (08-06), the tender launch (08-10) and Bain's sell-back election (08-12). Equibles' model puts the unpublished 2026-09-15 settlement at ~14.5m shares — that is an estimate, not a FINRA figure, and FINRA publishes the real one on 2026-09-24, the day after this pass.

Shelf, ATM, buyback. No ATM. The June notes were drawn off registration 333-292076, so a live shelf hangs over the name and has just been used. Buyback: the 2021-11-10 $1.5bn authorisation with $714,000,000 remaining at 2026-06-30 — untouched since Q3 2022 ($0 in 2023, $0 in 2024), and now partly encumbered by the Bain floor (§1.2).

---

5. S8 — valuation

Modelled on adjusted operating income available to common, because GAAP is unusable here: Q2 2026 GAAP EPS was $6.72 against $2.24 adjusted, and the difference is market risk benefit marks the company calls non-economic.

H1 2026 actual, annualised as the base year: Annuities $562m, Life Insurance $98m, Group Protection $259m, Retirement Plan Services $92m, Other Operations $(201)m → $810m of adjusted income from operations in the half, $1,620m annualised, less ~$92m of preferred dividends (falling to ~$52m from Q4 2026 after the tender).

Base ~$53 against $41.01 is about 29% of upside. On this repo's running count that makes LNC the third name in thirteen — after AX and AIG — whose correct operating forecast produces a materially positive return at the traded price.

Assumptions, stated so they can be disagreed with. Base 2028 adjusted operating income of $1,760m is +4.2% a year off the annualised H1 2026 run rate — spread earnings compounding on a $182bn annuity book growing 9% and a scaling funding-agreement programme, against a Group Protection loss ratio that management has already said is normalising (70% to 73% year on year) and a variable annuity block running off at ~10% a year. Bear assumes an assumption unlocking charge and continued spread compression; bull assumes the legacy life risk-transfer deal lands. The share count barely moves in base because the buyback is constrained (§1.2); the bull column is the only one that assumes LNC buys stock in size, which at $44 to Bain is a worse trade than it looks.

Cross-check on book, which is the denominator that matters for a risk carrier. At $41.01:

Correction to the cluster files. cluster-insurance-2026-09-22.md §6 puts LNC at book $59.29 and P/B 0.70, and the re-score at 0.71. $59.29 is total stockholders' equity ($11,349m) divided by common shares (191.4m) — it includes the $1bn of preferred. The company's own BVPS including AOCI is $53.68 and the honest common P/B is 0.76x. Both cluster files carry a caveat that the preferred is not tagged at a current date; the caveat is right and the number in the table is still the wrong one.

The AOCI hole is the whole of the gap: a $8.5bn pre-tax net unrealised loss on AFS securities at 2026-06-30, down from $9.1bn, about $23.71 a share of the difference between the two book values. That accretes back as bonds pull to par, which is why the base case does not need a re-rating to work.

Buy price and size: ⏳ OPEN (user).

---

6. Access, tooling and what was thin

• SEC EDGAR — reachable throughout, User-Agent sent on every call. XBRL companyfacts, submissions, Archives and efts.sec.gov full text all answered.
• 3spread — key present, HTTP 200. New failure mode, the fourth distinct one this repo has found. threespread.py holders LNC resolved the ticker to CUSIP 534187885, which is the Series D depositary share, and returned one filer holding 81,775 shares. The common is 534187109. Re-run with --cusip 534187109 it returns 413 filers and 110,445,310 shares. Any name with listed preferred is exposed to this; it does not error, it answers a different question. Recorded in the scorecard and in the README-level list of 3spread defects. owners also prints the issuer's name in the Filer column for every row, so a 13D/G row cannot be attributed without opening its index, and Bain's 2026-08-12 13D/A is absent from the owners list although EDGAR carries it under LNC.
• Equibles — reachable. 6 calls of the 25 cap: ListInvestorEvents, GetEarningsCallTranscript (Q1 FY2026), GetShortInterest, GetBuybackPrograms (four distinct endpoints; the transcript came back in one page). No Q2 FY2026 transcript exists in the source. Speaker resolution is again poor — the CEO and CFO are unnamed speaker numbers — which is the fourth name in a row where Equibles transcript attribution could not be trusted.
• ROIC.ai — prices fine. The earnings calendar again requires a paid plan, sixth consecutive slot.
• fool.com — 404 for the Q2 2026 call; a WebSearch restricted to fool.com found LNC Q1 2026 and nothing later.
• EDINET DB / eu_data — not applicable to a US domestic filer; not called.

Thin, named:

1. One transcript, not two — the Q2 FY2026 call is not available from any rung reachable here. Substituted the Q2 earnings supplement, which is primary but has no Q&A.
2. No scuttlebutt of this pass's own. S6 is carried forward and the carried row is [search-summary] only: Morgan Stanley Equal Weight, PT $42, dated before the CFO departure, the tender and the Bain election. Nothing in this repo about LNC comes from a customer, an agent, an employee or a competitor.
3. Form 4 transaction codes not itemised across all 171 filings — four read in full.
4. No analyst-coverage count.
5. Next earnings date DERIVED from filing cadence (Q3 calls 2024-10-31 and 2025-10-30; Q1 2026-05-07, Q2 2026-07-30) → ~2026-10-29, not a scheduled date.
6. The annual assumption review is the single biggest unmodelled item and it lands in Q3. The re-score's evidence has it at −$50m (2025), +$216m (2024), −$167m (2023) on net income. Nothing in this pass improves on that.