Employers Holdings, Inc. (EIG) stock intrinsic value fair value estimate & owner earnings analysis

Employers Holdings sells workers' compensation insurance, mainly to small businesses, in most U.S. states. It writes through independent agents, specialty distribution partners and digital marketplaces. ADP, the payroll company, is the largest single channel at 18.6% of 2025 premiums. California is 46% of premiums, so this is a concentrated bet on one state's workers' compensation environment. The company has a long record of profitability, but owner earnings here fell sharply in 2025. How we measure earnings: we start from operating cash flow. We remove the swings caused by claims payments, reserve changes, premium collections and reinsurance recoveries, so results reflect the claims the company actually incurred rather than when cash moved. We then subtract stock compensation and the spending needed to keep operations running. On that basis, owner earnings were about $100 to 155 million a year from 2016 to 2024, with a low of $69 million in 2021. They fell to about $14 million in 2025. The cause was a jump in California cumulative-trauma claims. That pushed the 2025 accident-year loss ratio to 72.0%, from 64.1% in 2024. The company also added $39.6 million to reserves for accident years 2023 and 2024. The combined ratio was 110.9%, against 97.9% in 2024. Reserve strengthening means the 2023 and 2024 earnings were somewhat overstated, so we do not treat the 2022 to 2024 level as normal. Our working earnings level is about $73 million a year. That is roughly the five-year average, trimmed for the roughly $200 million of capital the company spent on buybacks since the end of 2024. Share count: the company has shrunk its share count heavily, from about 23.5 million (2025 average) to about 18.0 million in July 2026. We value the business in total dollars and divide by today's share count. Recent trend: first-half 2026 net income was $39.2 million, close to the $42.5 million of first-half 2025. Earned premiums are shrinking, though: the second quarter was down about 12% from a year earlier and the first half about 7%. We give no credit for a premium rebound. Growth check: owner earnings per share fell about 86% in 2025 ($0.60 versus $4.19). Our base case starts at $65 million, below the anchor, so it assumes no rebound beyond the anchor. It sits below the $73 million level because premiums are shrinking. Bear case ($16.15 per share): The California cumulative-trauma problem is not fixed and results stay near 2025 levels, with owner earnings of about $15 million a year for three years. They then recover only to $35 million a year through year 10, and fade to zero over the following ten years. This uses only disclosed trends: the filing says cumulative-trauma frequency in California has "accelerated beyond historical levels". Reserve estimates rest on long-tail claims. The company has almost no debt, so leverage does not add to the bear case. Base case ($35.53 per share): Owner earnings of $65 million a year for ten years, then fading to zero over ten more. That corresponds to a combined ratio of roughly 104%, with investment income of about $110 million a year offsetting an underwriting loss. It is supported by a decade of positive owner earnings, including through 2020's premium refunds and 2025's reserve charge, and by investment income that stayed steady at $107 to 117 million a year in 2023 to 2025. We credit no growth. Buybacks are not counted either, because at fair prices they add no value. We also give no credit for premium growth, rate increases or a market recovery, because none is contracted or already in hand. Bull case ($53.57 per share): The company's own levers work. Claims handling and pricing bring results back to the 2022 to 2024 level of about $105 million, less the smaller capital base, so about $98 million a year for ten years, then a ten-year fade. The company ran a combined ratio at or below 98.6% from 2023 to 2024, and 96.0% in 2023, so this is a level it has already reached. We do not assume any growth beyond it. What would make this wrong: the base case needs the loss ratio to stabilise. A further year of prior-year reserve increases, or accident-year loss ratios above 75%, would move results toward the bear case. Weaker premium rates or continued shrinkage would do the same. One item could not be reconciled: the filing's cash flow includes several smaller lines (income taxes, reinsurance gain amortisation, contingent commissions) that we left in operating cash flow rather than stripping out. They are small relative to the totals. Margin-of-safety price: given the reserve uncertainty and California concentration, the margin-of-safety price reflects about a 40% discount to our base case. That gives $21. This also sits just above the bear case, which is a reasonable floor. The margin-of-safety price also assumes signs that the California trend has stopped worsening.