Enact Holdings, Inc. (ACT) stock intrinsic value fair value estimate & owner earnings analysis

Enact Holdings insures the credit risk lenders take when a homebuyer puts down less than 20%, standing between the mortgage and the government-sponsored enterprises (Fannie Mae and Freddie Mac) that require this coverage. It is a mono-line business: one product, one risk (U.S. mortgage credit), regulated for capital adequacy directly by those enterprises (the "PMIERs" framework). Business is written nationwide across the U.S. with no material concentration in any single state. Because this is a financial company, its own borrowing is not treated as debt to be subtracted from a business value the way it would be for an industrial company -- it is priced into the underwriting itself, and the earnings used here are already after interest expense. The starting point for value is GAAP net income per diluted share, which for an insurer is a cleaner measure of true earning power than cash flow (cash flow swings with the timing of reserve additions and releases in a way net income does not). Looking at the last seven years -- which span a pre-pandemic normal year, the pandemic-era stress year in which reserves were built up for a wave of forbearance-related delinquencies, and the subsequent recovery years -- the median figure is about $4.16 per share. That is the anchor used here, rather than the higher figures of the last two years. The key fact shaping every case is this: 2025's loss experience (11% of premiums, versus 4% in 2024) rose only because a large one-time-in-nature tailwind is fading, not because new delinquencies are increasing. Losses tied to newly delinquent loans have been flat at roughly $275 million a year for three straight years; what changed is that the release of reserves set aside for older, now-cured delinquencies shrank from $258 million in 2024 to $189 million in 2025, continuing a decline from $248 million in 2023. That release is running out, because the pool of loans it came from -- mostly 2020-vintage, pandemic-era delinquencies -- has largely finished curing. Once it is fully gone, losses run at roughly 28% of premiums based on the current pace of new delinquencies, versus 11% today. That gap is the single biggest swing factor in this valuation, and it is a real, evidenced, and quantifiable one -- not a market mood. The base case assumes that fade plays out over the next two to three years, taking earnings per share down modestly before capital return -- an active share buyback program funded from a large capital cushion (available capital was $1,919 million above the regulatory requirement at the end of 2025, a 162% sufficiency ratio) -- brings per-share earnings back to, and modestly above, where they are today. From there, growth is limited to what a mature, non-expanding market and steady buybacks can plausibly deliver: roughly 1-2% a year. That produces a value of about $75 per share. The bear case assumes the reserve cushion disappears faster and is compounded by a real, if moderate, rise in new delinquencies -- the kind that would accompany a soft patch in employment or home prices, not a repeat of the 2008 crisis (underwriting since then has been materially more conservative, and the capital cushion described above exists precisely to absorb a downturn without threatening the company's survival). Losses run near 45% of premiums for a couple of years before easing back toward the 28% normalized rate, and no buybacks are assumed during the stress. That produces a value of about $57 per share. The bull case credits continued strong loan-cure performance (a demonstrated multi-year pattern) so the loss rate normalizes only gradually and gently over the coming decade, modest ongoing growth in insured loan volume, and a continuation of the buyback pace the company has actually run for the last three years (authorizations have grown from $100 million in 2023 to $350 million in 2025 and $500 million announced in February 2026, funded entirely from the surplus capital described above). That produces a value of about $92 per share. A note on persistency: an unusually high share of borrowers are staying in their existing (lower-rate) mortgages rather than refinancing, which keeps premium-paying policies on the books longer and has been a genuine tailwind. That reverses if mortgage rates fall meaningfully, which would shrink future premium -- a real risk to keep in mind, though not one assumed to happen in the base case since nothing currently points to it. The company pays a small quarterly dividend (recently $0.21/share) alongside the buyback program; both are funded from the capital surplus described above, and the fixed-rate borrowing outstanding ($750 million at 6.25%, due 2029) is small relative to that capital base and not a near-term refinancing concern. Given the real, if moderate, cyclicality in mortgage credit risk, a meaningful discount to the base-case value is warranted, and a margin-of-safety price of about $43 per share is used here -- below even the bear-case outcome, to allow for a downturn.