WhiteHorse Finance, Inc. (WHF) stock intrinsic value fair value estimate & owner earnings analysis

WhiteHorse Finance is an externally managed business development company. It lends mainly first-lien, floating-rate loans to smaller U.S. companies and holds a stake in a joint venture with the Ohio state teachers' retirement system. Its shareholders are paid out of net investment income, meaning interest and fees collected less the company's own costs. At the end of 2025 the loan book was roughly $557 million at fair value, funded by about $260 million of shareholder equity and about $329 million of borrowings, including 7.875% notes due 2028. Asset coverage was 179%. How we measured earnings. A lender's reported operating cash flow mostly reflects loans made and repaid, so it says little about earning power. We used net investment income per share instead. It has no stock compensation and no capital spending to deduct. Loan losses are not in that figure, so we charge them separately in each case. Net investment income per share was $1.36 in 2021, $1.60 in 2022, $1.84 in 2023, $1.60 in 2024 and $1.13 in 2025. In the first half of 2026 it was $10.3 million, about $0.47 a share, or roughly $0.94 annualized. What has been going wrong. Net asset value per share fell from $15.23 at the end of 2020 to $11.68 at the end of 2025 even though the company paid out roughly $1.44 to $1.79 a share each year. Net realized and unrealized investment losses ran about $0.93 (2022), $0.96 (2023), $1.13 (2024) and $0.51 (2025) per share. Earnings were also pressed by lower yields, loans placed on non-accrual and a smaller portfolio. The dividend was cut to $0.26 a quarter for early 2026 from $0.385 through the third quarter of 2025. Loans on non-accrual fell from $69.6 million at cost at the end of 2024 to $19.7 million at the end of 2025 (fair value $10.6 million), so much of the cleanup has been recognized. The filing also warns of further markdowns after year-end: the first-lien loans to Outward Hound were expected to be marked from 85.8% to roughly 55-65% of face value, and those to Camarillo Fitness from 40% to roughly 15-25%. The company raised its share repurchase authorization to $22.5 million in February 2026 and replaced its auditor, Crowe LLP, with Deloitte & Touche LLP. Growth path. Net investment income per share fell about 30% in 2025 (from $1.60 to $1.13). For 2026 the base case starts at $0.94, a smaller decline of about 17%, because the first half of 2026 is already reported at that pace. We do not assume any recovery or growth. We credit no new lending growth, because it would rest on new loans not yet made and on a lower base management fee (cut from 2.00% to 1.75% of gross assets in 2024) offsetting a shrinking book. Each dollar of net asset value lost to credit losses removes roughly 10 cents a year of income, so income steps down about $0.06 a share a year in the base case. Bear case ($2.70 per share). Credit losses continue at about $0.85 a share a year, the 2023-2025 average, and income starts near $0.80 and falls about $0.10 a share a year. Leverage of roughly $329 million against $260 million of equity makes each loss cut income more sharply. Base case ($5.70 per share). Losses continue at about $0.55 a share a year, consistent with the five-year average of about $0.69 but with less non-accrual exposure than in 2024. Income starts at $0.94 and declines about $0.06 a year, with earnings running out over the long term as losses erode the portfolio. Bull case ($8.10 per share). Losses fall to about $0.35 a share a year, as they would if the non-accrual level of $19.7 million did not rise, and income starts near $1.00, close to the first-quarter 2026 pace, declining about $0.04 a year. The cases are valued at a 6% discount rate with no add-back of cash and no deduction of debt; borrowing costs are already in net investment income and leverage is reflected in the bear case. Because reported net asset value is $11.68 a share, these values sit well below book value. That gap reflects our choice to credit only income that is already visible and to charge the historical pace of losses. It would be wrong if losses stop abruptly, if the loans marked below par recover their value, or if repurchases below net asset value lift per-share results more than we allow; we did not credit repurchases, since they have not yet shown a lasting effect on income per share. One item we could not reconcile: the latest full annual report in our records is the 2025 10-K, and the second-quarter 2026 figures come from structured filing data only, so the pending markdowns above are not confirmed in a full document.