GLRE — Greenlight Capital Re, Ltd. · 2026-09-11 · Verdict: WATCH · Conviction 3
Price $15.02 (screen row, v2 screen run 2026-09-11; no live price) · Mkt cap $494M (32,881,538 shares, 10-Q 2026-06-30) · Financial lane, so no EV/EBIT: P/fully diluted book 0.73x · LTM P/E 9.7x · Debt $8.8M · ADV $1.8M Sources read: 10-K 2026-03-09 (Items 1, 1A, 7), 10-Q 2026-08-04, DEF 14A 2026-04-24, 8-Ks 2026-04-06, 2026-06-01, 2026-08-03, 2026-08-04, 2026-09-01, Form 4s (12m), Q2 2026 call transcript.
Desk stats - Revenue trend. FY2025 gross premiums written +10.7%, net premiums earned +6.6% to $661.1M (10-K, Item 1). Q2 2026 GWP +2% to $183.1M, net written -10.8%, net earned flat at $161.8M (8-K 2026-08-04, EX-99.1). Mix, not price: Innovations GWP +42.6% year to date on Syndicate 3456 and new financial and specialty treaties, against deliberate Open Market shrinkage and the 2025 casualty non-renewal (10-Q; Q2 call, CEO). - Normalised after-tax profit. H1 2026 net income was $6.2M ($0.18 diluted) against $30.0M, Q2 a $29.6M loss. Adjustments: add back $25.0M of Middle East conflict reserves, $6.5M for the QatarEnergy fire (10-Q, CAT losses) and a $1.5M Innovations private-equity impairment (10-Q); remove the $5.5M H1 FX loss and the $5.8M half-year Solasglas result, replaced with an 8% return on the $493.4M Solasglas balance ($39M) against a five-year record of 7.5/25.3/9.4/9.8/7.5% (10-K, Item 1). A 96.5% normalised combined ratio on ~$632M of annualised net earned premium gives ~$22M underwriting income; add ~$21M other investment income, less ~$21M corporate expense and ~$0.5M interest. Pretax ≈ $60M; at the actual ~4% effective rate (tax $273K on $6.4M pretax, H1 2026) ~$58M, or ~$45M at the desk's 25%. GAAP LTM net income $51.0M. - EV / normalised after-tax profit. EV is meaningless where liabilities are float. Equivalent: $494M / $58M = 8.5x (10.9x taxed at 25%), versus 9.7x GAAP LTM and the screen's 6.6x. - Financial-lane leverage. P/E 9.7x LTM; tangible book 0.71x basic, 0.73x fully diluted (no goodwill or intangibles); ROTE 11.1% CY2025, 7.5% LTM, 1.8% annualised H1 2026; fully diluted book per share $18.97 to $20.61 in twelve months, +8.6%. Debt $8.8M against $697.7M equity, 31.2% equity/assets. - Is the growth sustainable? Half of it. Innovations is compounding, with approval in principle from the Council of Lloyd's to become a full syndicate from 1 January 2027 (Q2 call, CEO); Open Market is shrinking on purpose. Book growth is cash-backed ($68.7M H1 operating cash flow) but it is a mid-90s combined ratio plus a hedge fund return, and the second part is not forecastable. - What the screen got wrong. The 6.6x P/E and 10.7% ROTE are CY2025; LTM they are 9.7x and 7.5%. The 0.70x tangible book uses basic book; management's own denominator is fully diluted book of $20.61, so 0.73x (10-Q non-GAAP reconciliation). The -14.1% "latest quarter revenue" is a Solasglas mark, not premium: gross premiums written rose 2%.
1. What the business actually does
Greenlight Re writes multiline property and casualty reinsurance from Cayman and Ireland plus Lloyd's Syndicate 3456, in two segments: Open Market (broker-placed treaty, $332.5M of H1 GWP) and Innovations (direct placements with insurtech MGAs it also invests in, $78.5M) (10-K, Item 1). The float is invested unlike any peer: substantially all investable assets sit in Solasglas Investments LP, a levered long/short equity fund run by David Einhorn's DME Advisors, capped at 70% of surplus since 1 August 2024 and standing at $493.4M, or 71% of equity, at 30 June, then 98.0% long and 64.6% short delta-adjusted with 9.0% in gold (10-Q).
2. Why it is mispriced: the edge case
There is none, and that is the finding. No spin, no index event, no forced seller; the stock is 20.9% off its 52-week high with +22.5% twelve-month momentum on $1.8M of daily volume. The discount to book is chronic and explained: a related party takes 1.5% a year plus a performance allocation on 70% of surplus, PFIC and controlled-foreign-corporation rules make the shares awkward for US taxable holders (10-K, Item 1A, Risks Relating to Taxation), and results swing on one fund's quarterly marks. No edge case caps the verdict at WATCH.
3. Insurer fundamentals: combined ratio and reserve development
Combined ratio 101.4% (2024), 94.6% (2025), 98.1% (H1 2026), 100.1% in Q2 (10-K, Item 7; 8-K 2026-08-04). The Q2 move is event-driven: 17.1 points of cat and large losses against 4.0 a year earlier, being $20M of further Middle East war reserves on top of $5M in Q1 plus $6.5M for the QatarEnergy fire (Q2 call, CFO). Underneath, the attritional loss ratio improved 4.3 points to 51.7% and acquisition costs fell 1.8 points; prior-year development was +0.4% in Q2 and favourable at -0.6% year to date. The caveat is the CFO's own: of the $25M conflict reserve, $7.6M is a known full-limit loss, $9.9M other specific events and $7.5M pure IBNR, with "a high degree of uncertainty... due to limited access to affected areas" (10-Q). Innovations printed 89.7% against 107.0%.
4. Balance sheet and capital allocation
Effectively unlevered: $8.8M of debt and $300M of fully cash-collateralised CIBC letter-of-credit facilities maturing December 2027 (8-K 2026-04-06). No dividend by policy. The capital story is the buyback: $9.8M in 2025 at $13.76, then $23.1M year to date 2026 at about $17.42, retiring 4% of shares, with $36M left under the 28 April 2026 authorization (Q2 call, CFO). Because buybacks lift Einhorn's percentage and the company judges that adverse for tax, it has twice agreed to buy from the David M. Einhorn 2021-07 Family Trust an extra 33% of whatever it repurchases in the market at the same average price, closing 3 August and again around 30 October (8-Ks 2026-06-01, 2026-08-04). That is not Einhorn selling a view: it retires a third more stock than the headline and holds his stake flat. Director Leonard Goldberg bought 12,000 shares at $15.62 and $15.42 on 7 and 10 August (Form 4, 2026-08-11), at today's price. Two experienced directors joined on 1 September, one a Casualty Actuarial Society fellow and former Aspen group CUO (8-K 2026-09-01). Executive PRSUs vest 65% on fully diluted book growth, 35% on combined ratio (DEF 14A).
5. Management: said versus did
It said it would buy stock at a discount and did, at three descending prices, and said it would hold underwriting discipline and did, cutting net written premium 11% in a quarter when gross rose. Pressed to lift the Solasglas ceiling, the CEO went as far as "I would not be surprised if that happens" while the CFO framed the choice as buybacks versus Solasglas (Q2 call); neither is a commitment. Einhorn said Solasglas returned 4.9% in July, taking 2026 to +6.1%.
6. Valuation
Fully diluted book is $20.61. Base: normalised ROE 8.5%, book compounds ~8% with ~2.5 points of buyback accretion at 0.75x book, reaching ~$23.7 in two years at 0.85x: $20.1. Bear: the war IBNR proves light and costs another $30M, Solasglas returns 0% for two years, combined ratio 101%, book stalls near $20.60 at 0.65x: $13.4. Bull: the cap is lifted, the fund compounds at its five-year 11.7% with the performance allocation halved to 10% until an additional 66.6% is earned (10-K, Item 1), ROE reaches 13% and the stock re-rates to book: $27.0. Weighted 55/25/20 gives about $19.2, 28% above price. Reverse DCF: at 0.73x fully diluted book with 2% terminal growth and an 11% cost of equity, the price implies a permanent ROE of about 8.6%, against 11.1% earned in 2025 and ~8.5% normalised here.
7. Catalysts and timeline
Q3 results in early November: whether the Middle East reserve holds. The Einhorn repurchase closing around 30 October and whether the board re-ups beyond $36M. A decision on the 70% Solasglas cap. Syndicate 3456 converting on 1 January 2027 and that renewal season.
8. Kill criteria (pre-registered)
- Adverse development on the Middle East reserve above $15M in any quarter, or the total conflict reserve above $50M.
- Combined ratio above 100% for two consecutive quarters excluding cat and large losses, or the attritional loss ratio back above 56%.
- The repurchase authorization lapses or goes unused for two consecutive quarters while the stock is below 0.85x fully diluted book.
- An A.M. Best downgrade below A-, which lets cedents cancel contracts or demand collateral (10-K, Item 1A).
9. Verdict and summary
WATCH, conviction 3. Greenlight Re trades at 0.73x its own fully diluted book while earning a normalised return on equity near 8.5%, with almost no debt, an A.M. Best rating upgraded to A in November 2025, and 4% of its shares retired in six months under an arrangement that quietly retires a third more than the open-market figure by buying the same proportion back from Einhorn's family trust. The screen's 6.6x P/E and 10.7% return on tangible equity are stale CY2025 arithmetic: on the last twelve months they are 9.7x and 7.5%, and H1 2026 annualises to 1.8%, because Q2 carried a $25M Middle East war reserve management concedes is largely an estimate made without access to the affected areas, alongside a 5.4% quarterly loss in Solasglas. What stops it being an idea is that nothing is forcing the discount: the stock is only 21% off its high on $1.8M of daily volume, and the gap to book is a decade-old feature of a structure where a related party takes 1.5% plus a performance allocation on 70% of surplus and where the tax paperwork deters US taxable holders. The reverse DCF says $15.02 already implies a permanent 8.6% return on equity, roughly what this business earns, so you are paid for buyback accretion and get any Solasglas or Innovations upside free rather than at a bargain.
Research for discussion, not investment advice. Positions and sizing are the reader's decision.