HCI Group, Inc. (HCI) stock intrinsic value fair value estimate & owner earnings analysis

HCI Group is a Florida-centred property insurer. Its two insurance companies, Homeowners Choice and TypTap, write homeowners, condominium and related property policies, a large share of them taken out of the state-run Citizens Property Insurance pool. HCI also manages two policyholder-owned reciprocal exchanges, Condo Owners Reciprocal Exchange (CORE, started 2024) and Tailrow (started 2025), and owns about 82.5% of Exzeo, its insurance software and policy-administration business, which listed separately on the NYSE in November 2025. Nearly all revenue is earned in Florida. The two things that drive results are the rate charged per policy relative to claims costs, and how much hurricane loss the company keeps after reinsurance. How we measured earnings. We start from operating cash flow and strip out the large timing swings in insurance balances: claim reserves, unearned premiums and amounts due from reinsurers move by hundreds of millions of dollars in hurricane years and reverse later. We then deduct stock-based pay as a real cost, deduct equipment spending, and deduct the share of profits that belongs to others: all of CORE's and Tailrow's earnings go to their policyholders, and since the Exzeo listing about 17.5% of Exzeo's earnings go to its outside shareholders. On that basis, earnings ranged between about -$1 and $4 per share from 2016 to 2021, with a small loss in 2017 (Hurricane Irma), and fell to -$8.36 per share in 2022 (Hurricane Ian). That era was dominated by Florida's claims-litigation problem. Florida's reforms of December 2022 and March 2023 removed one-way attorney fees and assignment-of-benefits claims, and results since then have been of a different order: $5.98 per share in 2023, $8.34 in 2024 and $23.50 in 2025. We treat 2023 onward as the business as it exists now and do not use earlier years to set the level. Those three years do not average out neatly. 2024 carried three hurricanes (Milton, Helene and Debby, about $128 million of net pre-tax losses) and still earned $8.34. 2025 had no catastrophe losses at all, which made it an unusually clean year. The first half of 2026, also free of storms, earned $11.05 per share against $10.57 a year earlier, and against $8.05 in the first half of 2024. Two consecutive quarters confirm that the book of policies in force is about a third larger per share than in 2024. We therefore set the typical level at about $15 per share. That is roughly what the current book earns before catastrophes ($22 to $23), less an average Florida hurricane load of about $5 per share (most years bring at least one storm), and less about $3 for the premium-rate declines now spreading across Florida. The state regulator is approving rate cuts, with average requests running about 5% lower, as reinsurance costs fall and competitors return. This is well below the 2025 figure, which we do not treat as normal. Protection against the big storm. For June 2026 to May 2027, HCI bought $4.06 billion of catastrophe reinsurance at an estimated cost of $381 million, 10% cheaper than the year before. Its maximum first-event retention is $162.6 million, and a second event costs about $75 million more. On today's earnings, a single maximum-retention hurricane costs roughly $10 per share after tax, and two in one season cost about $14. That is painful but not ruinous. A storm large enough to exhaust the reinsurance tower is a rarer event than any of our cases assume. These reinsurance terms come from the company's June 2026 announcement rather than an audited filing. Growth. We credit none. The Citizens pool that fed HCI's growth has shrunk from about 1.4 million policies at its 2023 peak to about 266,000 in August 2026, so large takeouts cannot repeat on the old scale. Growth at CORE and Tailrow mostly belongs to their policyholders, and wider sales of Exzeo's software to outside insurers have not yet happened. All values below are discounted at 6%. Base case ($180 per share): owner earnings of about $15 per share for ten years. Then, over the following ten years, excess returns fade until the company earns about 6% on its roughly $87 per share of shareholders' equity, the point at which that equity is worth its book value. Nothing we can point to guarantees today's post-reform margins beyond that, so we let them compete away. Bear case ($95): rate cuts and some return of claims litigation push typical earnings down to about $8 per share, roughly the 2024 level. On that lower earnings base, an early season with two maximum-retention hurricanes and some adverse claims development produces a loss of about $5 per share. Returns then fade to the cost of capital by year 15 on a somewhat smaller equity base of about $80 per share. Bull case ($218): the lower reinsurance cost locked in for 2026-27 offsets the rate cuts, pre-catastrophe margins stay near the 2025 and 2026 levels, and typical earnings run about $19 per share for ten years before fading the same way as the base case. What would make this wrong: Florida's 2023 litigation reforms are the main reason for the change in results. If they were rolled back or eroded by the courts, earnings could fall back toward the pre-2023 range, below the bear case. Heavy concentration in one state means a single severe season, or a reinsurer failing to pay, can take a large bite out of equity. On the other side, a run of quiet hurricane seasons would earn well above our base level. Notes: the weighted diluted share count for 2017 was not reported separately and was derived from that year's loss and loss per share. In the 2023 to 2025 cash-flow statements, two single-year items had to be assigned to their correct years by reconciling against the other reported figures.