FEDERAL AGRICULTURAL MORTGAGE CORP (AGM) stock intrinsic value fair value estimate & owner earnings analysis

Farmer Mac (Federal Agricultural Mortgage Corporation) is the federally chartered secondary market for rural America. It buys and guarantees farm and ranch mortgages, makes AgVantage loans secured by the obligations of large agricultural lenders, holds USDA-guaranteed loans, and since 2022 lends to rural electric cooperatives, broadband networks, renewable energy projects and corporate agribusiness. Its charter lets it borrow at spreads close to those of government agencies, and it relends that money at a wider margin. That funding advantage is a legal privilege granted by Congress rather than something the company built, but it has operated without interruption for more than three decades. Because borrowing is the raw material of this business, we treat interest as an operating cost and measure what the common owners earn: net income after the dividends on its preferred stock. No debt is subtracted, because the company's notes are its funding, not a claim that ranks ahead of an operating business. On that basis earnings per share rose from $5.97 in 2016 to $6.60, $8.83, $8.69 and $8.80 in 2017 to 2020, then to $10.27, $13.87, $15.80, $16.44 and $16.62 in 2021 to 2025, with the share count almost unchanged at about 11 million. The 2020 and 2021 figures are the restated amounts from later filings. Return on common equity has run around 15 to 17 percent, and book value per common share was $111.54 at the end of 2025. The level. Earnings have stepped up year after year, so a ten-year median would describe a much smaller company. We start from the business as it is now: 2025 earnings of $16.62 on $111.54 of book value. 2025 earnings grew only about 1 percent, because credit costs rose to $32.9 million from $11.5 million. That included $14.6 million charged off in the Corporate AgFinance portfolio, a $4.4 million Farm & Ranch charge-off and a $5.1 million provision on a broadband loan. The first half of 2026 then grew about 20 percent, with earnings of $10.15 per share against $8.49. Both quarters were up roughly 18 to 21 percent, and the provision fell back to $11.3 million. Those two quarters are why our first-year growth of about 7 percent is above 2025's 1 percent. We do not treat the first half's 20 percent as the new run rate. Base case: return on common equity of about 16 percent in year one, easing to 15 percent and then 14 percent. Book value grows 8 percent in year one, 7 percent a year through year five and 4 percent a year after that. Retained earnings fund this growth while the $6.00 annual dividend continues. We credit growth because of a demonstrated, self-funded record: total assets grew from $15.6 billion in 2016 to $35.4 billion in 2025, about 9.5 percent a year, with almost no new shares issued. The base rate sits below that record and tapers. We give no credit to any new lending line beyond the portfolios that already exist. After year ten the case assumes a 13 percent return on a book that no longer grows. What sustains that return is the statutory funding advantage, which is already in place. This is worth about $329 per share. There is no debt to subtract. Bear case: the ten years on record contain no real agricultural downturn, and this company is leveraged about 20 to 1, so the bear case has to supply one. We model a farm-credit cycle of the 1980s type, when farmland values fell by a third or more across the Midwest, together with the kind of losses Farmer Mac itself took in 2008. That year, losses on securities issued by Fannie Mae and Lehman Brothers forced an emergency capital raise. In the case, cumulative after-tax credit losses reach about $33 per share, roughly $360 million or 3 to 4 percent of the credit-exposed farm, corporate agribusiness and infrastructure book, spread over years three to five. Returns fall to 10 percent, growth stops for four years and the dividend is cut. Afterwards growth resumes at only 3 percent, at 10 percent returns. Beyond year ten, returns hold at 9 percent with no growth. At the end of 2025 the company held $583 million of capital above its regulatory minimum, which is enough to absorb those losses without issuing common shares, so the case includes no dilution. This is worth about $156 per share, with no debt subtracted. That is above book value. The reason is that even after the losses, the company would still borrow at agency-like spreads, and a 9 to 10 percent return on capital is worth more than the capital itself. Bull case: returns hold near the 17 percent the company has already earned, and volume keeps growing at the 10 percent pace of the past decade for five years, then 7 percent. Infrastructure lending continues at its recent scale. This is worth about $450 per share. What could make this wrong. Congress can amend the charter, and five directors are appointed by the President, so the funding advantage depends on Washington. Interest-rate and liquidity mismatches are amplified at 20 times leverage. The newer lending lines in broadband, renewable energy and corporate agribusiness produced most of 2025's charge-offs and have no history through a downturn. Margin-of-safety price. We are fairly sure of the earnings level. It rests on ten years of steady gains, a 2025 that already absorbed higher credit costs, and two strong 2026 quarters, though none of it has been tested by a farm recession. The worst plausible outcome is a credit loss combined with a funding disruption, as in 2008. The company's capital cushion and its access to agency-like funding should let it survive that, but the common shares could fall far and stay down. If we are wrong about the return on equity, so that it settles at 11 to 12 percent instead of 14 to 15, and wrong about growth, the shares would be worth roughly $170 to $230. In the meantime, the business earns about $18 to $20 a share, pays a $6.00 dividend and adds about 9 percent a year to book value. At $110, roughly what the company's capital is worth today, the dividend alone is about 5.5 percent of the price and the business compounds book value near 9 percent. If the shares only reached our bear value of about $156 over five years, that price implies a return in the low double digits, and any outcome near the base case implies well over 20 percent a year. That is why the margin-of-safety price is $110 and not above. Our previous figures were calculated on a different basis and are not comparable with these. The values apply equally to the Class A voting and Class C non-voting common shares, which share earnings and dividends equally. The listed preferred series are separate securities.