MidCap Financial Investment Corp (MFIC) stock intrinsic value fair value estimate & owner earnings analysis

What the business is. MidCap Financial Investment Corporation is a business development company: a publicly traded lending fund, managed externally by an Apollo affiliate, that borrows money and lends it to mid-sized, private-equity-backed US companies. At June 30, 2026 it held $2.77 billion of loans and investments at fair value (cost $3.06 billion), about 94% first-lien senior secured and nearly all floating-rate. It was funded by $1.10 billion of shareholder equity (net asset value of $13.37 per share on 82.4 million shares) and $1.74 billion of borrowings, with $43 million of cash. The company was formerly Apollo Investment Corporation, whose older book of subordinated loans, oil and gas, aircraft leasing and shipping destroyed a lot of capital: net asset value per share fell from $24.54 in March 2015 to $15.10 in December 2022. It was renamed in 2022, repositioned to lend alongside Apollo's MidCap Financial platform, had its fees reset from January 2023 (1.75% a year on net assets plus a 17.5% income fee limited by a three-year total-return lookback), and merged in two Apollo closed-end funds in July 2024. How we measured earnings. A lender's operating cash flow swings each year because new loans and repayments run through it, so we set loan activity aside the way we would set aside working capital. What remains is net investment income, from which we subtract realized credit losses as the cost of keeping the loan book whole. There is no stock compensation and no capital spending. Income includes interest that borrowers add to their loans instead of paying in cash (about $8 million in the first half of 2026); we count it, because any of it never collected comes back later as a loss. On this basis the company earned $1.78 per share in 2023, $0.72 in 2024 and $0.99 in 2025. We anchor on the three years since the strategy and fee reset, median about $0.99, because the years before 2023 swung from a loss of $1.71 to a gain of $2.07 per share and describe a different portfolio. A stricter test, the change in net asset value plus dividends, fell from $1.83 per share in 2023 to $1.29 in 2024 and $0.72 in 2025, and was a loss of $0.19 per share in the first half of 2026, or $0.51 before the boost from buying back shares below net asset value. How we treated the debt. For a lender, borrowing is the raw material of the business: the $1.74 billion funds the loan book, and the interest on it (about $105 million a year, roughly $1.28 per share) is charged as an operating cost before we count any earnings. We therefore value the earnings left after interest and do not subtract the borrowings a second time; the risk of the leverage is carried in the cases, above all the bear case. The borrowings are laddered: the $125 million of 4.50% notes came due on July 16, 2026; the senior secured credit facility runs to October 1, 2030; the three loan securitizations are floating-rate and match-funded to 2035–2037; and $80 million of 8.00% notes mature December 15, 2028. The binding constraint is not a maturity but the law: a business development company must keep assets of at least 150% of its borrowings. At June 30, 2026 the company stood at about 163%, a cushion of roughly $225 million. A fall of only about 8% in the value of the loan book would use that cushion up. What is happening now. Credit has worsened for 18 months. Loans on non-accrual rose from 1.4% of cost in March 2025 to 3.8% in December 2025. In the first half of 2026 the portfolio lost $111 million in value ($13 million realized, $99 million in write-downs), and net asset value per share fell from $14.18 to $13.37 even though buybacks below net asset value added to it. ChyronHego (where $60 million of debt was converted to equity at a $21.5 million loss), Midwest Vision Partners, New Era Technology, American Restoration and Thomas Scientific drove most of the damage, alongside long-standing problem holdings such as Carbonfree Chemicals and LendingPoint, now marked at well under half of cost. The gap between cost and fair value is $292 million, about $3.54 per share. The quarterly dividend was cut from $0.38 to $0.31 in February 2026. The company is shrinking on purpose: in the first half it bought $150 million of new loans against $456 million of repayments and sales, repaid $253 million of borrowings net, and bought back 9.8 million shares for $108 million. June-quarter net investment income of $32.8 million, about $0.40 per remaining share, carried no income fee at all, because the lookback switched it off after losses; fee and prepayment income were small ($0.6 million), so the quarter was not flattered by one-off items. The cases. The loans are senior, floating-rate and mostly repay within about five years, so we value ten years of earnings plus the net asset value still standing at year ten, and nothing beyond that, because the company's long record of losing capital does not support treating today's earnings as permanent. No growth is credited in any case: the book is shrinking, new lending is paused by choice, and further buybacks depend on selling loans. Owner earnings rose 37% in 2025, from $0.72 to $0.99 per share; the base case assumes $0.90 next year, below both that pace and that level. Base case, about $13.92 per share. Before the income fee, net investment income runs about $1.50 per share: the June-quarter pace on today's share count, less about $0.06 for cutting leverage toward 1.4 times and $0.02 for replacing the 4.50% notes with borrowing costing about 6%. The income fee stays off in year one, returns partly in year two and then runs about $0.18 a year, what the lookback allows at this level of returns. New credit losses, beyond the write-downs already taken, are $0.60 per share in year one, $0.50 in year two and $0.45 a year after that, about 1.35% of the portfolio a year, which is above the long-run loss rate for first-lien lending and so already allows for an ordinary recession somewhere in the decade. That is set against the company's own record since the reset: almost no losses in 2023, then roughly $0.42, $0.80 and, in the first half of 2026 alone, about $1.28 per share of value lost to credit. Owner earnings are about $0.90 a year for two years, then $0.87. Net asset value at year ten stays near today's $13.37 because losses are paid out of income rather than out of capital. On this path the loans and other assets are worth about $2.85 billion, the borrowings net of cash are about $1.70 billion, and shareholders are left with about $1.15 billion. Bear case, about $9.30 per share. This case runs today's troubles straight into a full credit downturn of the size middle-market lenders actually suffered in 2008–2010. The arithmetic comes from that episode: the US leveraged loan default rate peaked at 10.8% in November 2009, roughly 15% of borrowers defaulted over the three years, and first-lien loans to middle-market companies recovered only about 55 cents on the dollar (Fitch's 2007–2013 study), so about 6.75% of a first-lien book was lost outright, roughly $187 million or $2.27 per share on today's portfolio. This company lived through that cycle once already: net asset value fell from $15.83 to $9.82 per share in the year to March 2009 (before a later 1-for-3 share consolidation) and total assets shrank by about a third as it sold loans to stay within its borrowing limits. Today's book is far more senior, which is why we size the loss from first-lien history rather than from that old portfolio, but leverage is higher. In the case: short-term rates fall about 1 percentage point, which the company estimates would cut income by $8.8 million a year (about $0.11 per share); losses stay near today's pace at $1.10 per share in year one; the downturn then arrives in years two to four, with losses of $1.00, $1.80 and $1.21 per share, including about $0.39 lost selling some $400 million of loans at around 92 cents to stay above the 150% asset-coverage floor once write-downs eat through today's $225 million cushion; non-accruals near 9% of the book and the smaller book cut pre-fee income to about $1.00 per share at the trough; then losses ease to $0.55 and $0.45 a year, income recovers only to about $1.15 because the book is permanently smaller, and the income fee returns at just $0.10. Owner earnings run about $0.24, $0.25, minus $0.80, minus $0.26, $0.50, then $0.60 a year. Over the decade about $7.90 per share, roughly 24% of today's portfolio, is lost to credit on top of today's write-downs; to keep paying lenders, that loss has to be retained out of income rather than distributed. Interest is covered by income in every year, so lenders are paid; the shareholders bear the whole loss. On this path the loans are worth about $2.47 billion, the borrowings net of cash are about $1.70 billion, and shareholders are left with about $0.77 billion. This is still well below today's net asset value, so no franchise value above book is assumed. Bull case, about $16.34 per share. Income holds near $1.52 per share at today's rates, new losses fall to $0.30 and then $0.25 a year as the named problem credits are worked out, and about $1.00 per share of today's write-downs comes back over three years as marked-down loans repay at par. That has happened here before: in 2025 Merx Aviation gained $32 million from Russian insurance settlements. The income fee returns in full at about $0.19. Owner earnings are about $1.45, $1.41 and $1.42 in the first three years and $1.08 after that. The loans are worth about $3.05 billion, the borrowings net of cash about $1.70 billion, leaving about $1.35 billion for shareholders. What we left out. In May 2026 news reports said Apollo was in talks to sell the company, possibly to another business development company for shares. No agreement or terms have been disclosed in the company's filings, so no case includes it. What would make this wrong. More than half of the value rests on net asset value being real. If write-downs continue at the first-half-2026 pace, about $1.28 per share every six months, value heads toward the bear case before any recession arrives. Larger realized losses on loans made after 2022 would confirm that the problem lies with the new strategy, not only leftovers from the old one. On the upside, falling non-accruals, the income fee returning and marked-down loans repaying at par would support the base case or better. What changed since our previous review. No new quarterly report has been filed since then, so the business facts are unchanged. The bear case and the margin of safety price are built on a different basis than before, so those earlier figures are not comparable. Margin of safety price, $8.00. How sure are we of the earnings level? Moderately at best: income before credit losses is visible to within about ten cents a share, but credit losses are the swing factor, they have tripled in two years, and the latest quarter's income benefited from the income fee being switched off. The worst plausible outcome is a 2008-style credit cycle hitting a book run at about 1.5 times debt to equity: losses compound through leverage (each 1% of loan losses costs shareholders about 2.5% of book), an 8% fall in loan values reaches the legal borrowing limit, and the company would be forced to sell loans or issue shares at the bottom. The structure would survive it, because the debt is long-dated, match-funded or not due until 2028–2030, and the loans are first-lien and floating-rate, but shareholders would absorb years of income. What could also happen that no case models: Apollo, which controls the manager, could sell the company on terms set mainly by the acquirer's share price, or keep it shrinking indefinitely. If we are wrong about the two things most likely to go wrong, loss rates staying near 2025–2026 levels and no return of the income fee, the shares are worth roughly $11–12; if a real recession arrives too, about $9. The price assumes a three-to-five-year holding period. At $8.00 the business earns about $0.87 a share in owner earnings, close to 11% a year on the price, and the current dividend of $1.24 is about 15%. In the base case, over three years with dividends, $8.00 grows to roughly $16–17, implying about 25–28% a year; the bear case, through the downturn, implies roughly 12–13% a year with its value still above that price. At $9.00 the base case implies only about 20–23% a year, too thin for a book this uncertain, so the margin-of-safety price is $8.00.