EAGLE BANCORP INC (EGBN) stock intrinsic value fair value estimate & owner earnings analysis

Eagle Bancorp owns EagleBank, a commercial bank lending and taking deposits almost entirely in the Washington, D.C. metropolitan area: the District, Northern Virginia and the Maryland suburbs. Its business is commercial real estate, construction and business lending to local developers and companies, funded by business operating deposits and, to an unusual degree, brokered deposits ($3.3 billion, 36% of deposits at the end of 2025, down from 44% a year earlier). What happened. The office-property downturn hit the loan book hard. The bank wrote off $248 million of loans in 2025 (3.2% of average loans), mostly office buildings and loans with office collateral, after a $66 million provision and a $104 million goodwill write-off in 2024. It lost $138 million in 2025 and cut the quarterly dividend to $0.01. Management has shrunk the riskiest exposure: income-producing office loans fell from $862 million to $576 million during 2025, nonperforming assets fell from $211 million to $109 million, and $91 million of troubled loans were moved to held-for-sale at fair value. The first half of 2026 returned to a profit of $21.6 million, with a $34.8 million provision against $164.4 million a year earlier. Write-offs are still running high, though: $75 million in the first half of 2026, more than the new provision, so the reserve fell from $160 million to $121 million. The workout is not finished. The bank has also shrunk. Loans are down from $7.9 billion at the end of 2024 to $6.6 billion at mid-2026 and deposits from $9.1 billion to $8.2 billion. Net interest income is running near $250 million a year, against $289 million in 2024 and $333 million in 2022. How we measured earnings. Operating cash flow at a bank adds back the loan-loss provision as a non-cash charge. But a loan that is never repaid is a real loss to the owners, so we charged a normal level of credit losses against earnings. This is the main reason our figures are much lower than our previous analysis, which did not make this deduction. Those earlier figures are not comparable. We built normal earning power from the bank as it is today, not from its larger past balance sheet. Pre-tax, pre-provision earnings were $98.5 million in 2025, which was depressed by workout costs, and are running at about $113 to $116 million a year in 2026 (four straight quarters near that level). Our base case uses $115 million. From that we subtract normal credit losses of about $25 million a year (roughly 0.38% of loans; Eagle's own losses averaged about 0.15% a year from 2017 to 2023, so this is deliberately above that record because of the remaining office and multifamily exposure) and a 24% tax rate. That leaves about $68 million, or $2.24 a share. That is roughly a 6% return on the $1.15 billion of shareholders' equity, all of it tangible after the goodwill write-off ($37.7 a share). Because this is a bank, deposits and borrowings are its raw material. We valued earnings after interest and did not subtract debt; the $77 million of long-term debt is already paid for inside those earnings. No growth is credited in any case. The loan book is shrinking, and Eagle has no long, steady record of growth through a downturn that would put growth in the bag. Beyond ten years each case assumes the deposit franchise keeps earning at its case level, with no growth. That rests on 28 years of relationship banking in one of the country's largest metro economies. Base case, about $36.30 a share. The office workout runs its course through 2026. Earnings are about $1.45 a share in 2026 (first-half results annualized) and $1.90 in 2027 as write-offs normalize, then hold at $2.24 a year. The 2025 loss year makes the trailing growth rate meaningless. The 2026 recovery assumption rests on a dated fact: the provision fell 79% in the first half of 2026 against the first half of 2025, and nonperforming assets halved during 2025. Bear case, about $24.90 a share. One full commercial-real-estate credit cycle of the kind CRE-concentrated banks went through in 2008 to 2011 lands in 2027 to 2029, on top of the unfinished office workout. A named local risk adds to it: the 10-K flags federal workforce and spending cuts as a direct threat to the D.C. economy. In this case cumulative losses reach 5% of loans (about $330 million over three years). Pre-provision earnings fall to $100 million as deposit costs stay high, and the bank posts small losses of about $0.25 a share for three years. Equity falls by about $250 million, to roughly $900 million. Starting from a 13.1% common equity tier 1 ratio, the bank stays above well-capitalized minimums and does not need to issue shares at trough prices. That is why the bear value does not go lower. Afterwards earnings settle at about $1.84 a share, with credit losses at 0.40% of loans. The bear value is below tangible book value, which is consistent with a bank earning less than its capital should. Bull case, about $46.80 a share. The bank's own levers work. Brokered deposits keep running off and are replaced by cheaper funding, continuing the 42-basis-point drop in funding cost seen in 2025. Pre-provision earnings recover to about $130 million, and credit losses return to near Eagle's pre-2023 record at 0.20% of loans. Normal earnings reach about $2.91 a share. Margin-of-safety price: $21. How sure are we of the earnings level? Only moderately. The pre-provision earnings are well established across four quarters. Credit costs are not: write-offs still exceed new provisions, and the remaining $576 million office book and $925 million multifamily book have not finished repricing. The worst plausible outcome no case fully models is a second wave of office and multifamily losses running alongside a sharp D.C. downturn from federal cutbacks, while a third of the funding is brokered, rate-sensitive money that regulators restrict if capital falls below well-capitalized. The bank has about $1.15 billion of equity and no goodwill, so it can absorb losses of roughly the 2025 scale once more without new shares. A third such year would likely mean dilution. The things we are most likely to be wrong about are how much of the office book is still overvalued, whether the shrinking loan book keeps shrinking pre-provision earnings below $115 million, and how fast brokered funding can be replaced. If all three go against us, the shares are worth something close to the bear case, in the mid-$20s. While waiting, the $0.01 dividend pays almost nothing, so the whole return has to come from the gap between price and value closing; the price assumes a holding period of roughly three years. At $21 the price is about nine times normal earnings for a bank with about $37.70 of tangible equity per share. That price sits comfortably below what the bank is worth even if the next credit cycle arrives on top of the current one, which is what a 20% annual return needs when the bank pays almost nothing while waiting. One item could not be fully reconciled from the filings: the reserve-for-loan-losses roll-forward for the first half of 2026 (provision $34.8 million, gross write-offs $75.2 million, reserve down $38.5 million) implies recoveries of only a few million dollars. We took the reported figures as given.