$178.00
$280.00
bear $178.00
base $240.00
bull $280.00
margin of safety $110.00
market price* $168.77
Globe Life sells small life insurance and supplemental health policies, mostly to middle-income and union families, through its own exclusive agent networks: American Income Life (about 53% of life premium), Liberty National, Family Heritage, a direct-to-consumer arm and the Medicare supplement business United American. Premiums renew year after year, and total premium grew 5% in 2025 to about $5 billion. Roughly 2% of the $20.5 billion investment portfolio is rated below investment grade. Cash from operations has stayed between $1.28 billion and $1.48 billion in each of the last ten years, including the 2020 shutdown. How we measure earnings. For an insurer, growth in policy reserves and the up-front commissions that are capitalised and amortised are the normal working capital of the business, so we strip those timing items out, then deduct stock compensation and the capital spending needed to stay in business, while adding back non-cash investment gains and losses. On that basis owner earnings were about $1.0 billion in 2025 ($12.05 per share). Two judgements matter. First, a 2023 accounting change (applied back to 2021) lowered the amortisation of acquisition costs and raised reported profit by roughly a third, so earlier years are not comparable and we measure only from 2021. Second, 2025 capital spending of $142 million was double 2024; first-half 2026 spending ran at about half that pace, so we do not treat 2025 as the new normal. We anchor on the middle of 2021-2025 owner earnings, about $969 million a year, which is about $12.29 per share on the roughly 78.9 million diluted shares outstanding in mid-2026. We deliberately state the anchor in total dollars and divide by today's share count: buybacks already done are permanent, but the per-share growth of 11% in 2025 (from $10.81 to $12.05) came mostly from fewer shares, while total owner earnings grew only 2.6%. We therefore start the base case at 2.5% growth in the first year, not at the per-share rate. Borrowings are modest for an insurer: about $2.66 billion face value, mostly senior notes plus $306 million of commercial paper (a term loan and notes mature in 2027 and 2028). Interest of about $145 million a year is already deducted in the earnings above and the borrowings are the company's funding, so we do not subtract them a second time. In each case below, the business is worth the figure stated and the debt is paid out of its earnings rather than deducted again. Base case: about $240 per share (roughly $19 billion). Owner earnings grow 2.5% a year for five years, then 2%, 1.5%, 1% and 0.5%, reaching about $1.18 billion by year ten, then stay flat. What we credit: the agent-driven sales engine that has grown revenue about 4.8% a year since 2016, through the 2020 shutdown, funded from the company's own cash. We credit less than that, and taper it, because the American Income Life agent count fell 7% year on year in the second quarter of 2026. Management expects mid-single-digit agent growth in the second half; we leave that out because it has not happened yet. We also leave out management's expectation of $110-130 million of one-time gains from third-quarter actuarial updates, the better health claims experience (the health underwriting margin actually slipped to 23% from 26% a year earlier), and any gain from share repurchases beyond today's share count. Bear case: about $178 per share (roughly $14 billion). Owner earnings stay flat at about $969 million for two years, with no growth at all because recruiting does not recover. Then two named problems arrive. First, the Equal Employment Opportunity Commission found reasonable cause to believe that a group of American Income agents should be treated as employees; the conciliation ended without agreement in October 2024, and the agency has not communicated since but could still sue. If agents had to be treated as employees, payroll taxes and benefits on roughly $1 billion of agent compensation would cut owner earnings about 12% permanently (to about $853 million). Second, one full credit cycle on the $20.5 billion portfolio, sized from Torchmark (the company's predecessor name) in 2008: in the third quarter of 2008 alone it wrote off about $93 million on Lehman, Washington Mutual and Fannie Mae preferred holdings, about 1.2% of its portfolio in one quarter. We use total after-tax losses of about 3% of the portfolio, about $485 million spread over two years, which cuts owner earnings to about $610 million for those two years before returning to $853 million. Capital stays well above what is needed, so no stock issuance is assumed. Bull case: about $280 per share (roughly $22 billion). The agent networks execute on the levers visible today: Liberty National agents up 8% and Family Heritage agents up 7% in the second quarter, health premiums up 16%, and American Income agent count recovers. Owner earnings grow 3% in year one, then 5% a year (about the company's demonstrated revenue growth) for four years, 4%, 4%, 3%, 3% and 2%, reaching about $1.4 billion in year ten, then flat. What could make this wrong: the biggest unmodelled risks are the employee-classification issue turning into a lawsuit that also forces changes to how agents are paid and recruited (damaging the sales engine itself, not just costing 12%), and the shareholder class action and derivative suits that followed the April 2024 short-seller report (the company is indemnifying the individuals; no loss has been recorded and the potential size cannot be determined from the filings). Annual actuarial assumption updates can also move reported profit either way, and a prolonged fall in agent count would erode the premium growth we credit. How we reached the margin-of-safety price. We are very sure of the earnings level: owner earnings have sat between $933 million and $1,012 million for five years, and operating cash flow has been steady for ten. We are less sure about growth, which depends on agent recruiting that is currently weak in the largest division. The worst plausible event, outside our cases, is the loss of the exclusive-agent model through reclassification or regulation; the business would survive (the policies in force keep renewing and the borrowings are modest) but earnings could be about a third lower and growth would stop. If we are wrong about agents and credit together, the shares would be worth about $120-130. The price assumes a holding period of five years or more. During that period the business pays a small dividend (about $1 a share) and uses most of its owner earnings to retire shares, which shrinks the share count by roughly 4-5% a year. At $110, the anchor owner earnings of $12.29 a share are an 11% annual yield on the price, plus about 2.5% growth, and a rise of the shares to only about 14 times those earnings (around $172, well below our base value) over five years would imply a total return of about 20% a year. The margin-of-safety price is $110 and no higher because the margin for error disappears quickly once the agent-count and classification risks are considered. A previous analysis of this company was prepared on a different basis (profit per share instead of owner earnings after stock compensation and capital spending) and its figures are not comparable. What has changed in the business since then: American Income agent count fell 7% in the second quarter of 2026, health margins slipped, and first-half 2026 net income rose 10% to $558 million. Sources: the 2025 annual report and the second-quarter 2026 filing and call for facts and dates only.