Cherry Hill Mortgage Investment Corp (CHMI) stock intrinsic value fair value estimate & owner earnings analysis

Cherry Hill Mortgage Investment Corporation is a small, internally managed real-estate finance company. It earns money two ways: (1) it owns roughly $1.2 billion of agency-backed residential mortgage bonds (RMBS), financed largely with about $1.14 billion of short-term repurchase borrowings and hedged with interest-rate swaps, and (2) it owns about $215 million of mortgage servicing rights, which pay a fee on the loans being serviced. Servicing fee income was $53.4 million in 2023, $48.5 million in 2024 and $43.3 million in 2025. In November 2024 the company ended its outside management agreement and hired its own management team, without paying a termination fee. Ahead of the common shares sit $109.6 million (liquidation value) of 8.20% and 8.25% preferred stock, which costs about $9.8 million a year in dividends. How we measured earnings. We start from operating cash flow, remove swings in working capital and the non-cash value of stock awards, and then deduct the preferred dividends, because they must be paid before common holders receive anything. The company has no capital spending in the usual sense. We add back the periodic fair-value write-downs on servicing rights, as the accounting requires, but treat the run-off of those rights as a real decline, which we carry in the growth path rather than by lowering the starting level. On that basis, earnings per common share were about $3.37 in 2016, peaked near $4.51 in 2019, and then fell to $2.19 in 2021, $1.34 in 2022, $0.56 in 2023, $0.04 in 2024 and $0.47 in 2025. We use the median of 2023 to 2025, $0.47, as the starting level. The business today is heavier on servicing rights and much more leveraged to funding costs and hedging results than in the years before 2021, and it has been run by its own staff only since late 2024, so the earlier, larger figures describe a different setup. 2024 is included as a genuine weak year: operating cash flow was negative $4.7 million, driven by a $18.6 million drop in accrued liabilities around the internalization. Per-share earnings also carry ongoing dilution: common shares outstanding rose from 31.6 million to 36.7 million during 2025 through at-the-market sales. Growth check. Earnings per share rose sharply in 2025 versus 2024, but only because 2024 was a trough; on the median level, we assume the first year is about 10% lower, not higher. The servicing fee stream has shrunk about 9% in 2024 and about 11% in 2025, and the company has essentially stopped buying new servicing rights (purchases were a few thousand dollars in each of 2024 and 2025). We found no contracted or already-funded source of growth, so none is credited. Growth in the bond portfolio depends on selling new shares, which has not been shown to add value per share, and is left out. Base case, $2.63 per share. Earnings per share start at about $0.47 and shrink 10% a year, matching the observed decline in servicing income and the steady dilution, and we continue that decline beyond year 10 because we can name nothing that reverses it. Discounted at 6%, this is worth about $2.63 per share. Common equity on the balance sheet is roughly $130 million, or about $3.5 per share at year-end 2025, so this case values the shares below stated book value. Bear case, $0.48 per share. A repeat of the 2024 experience (earnings near zero) in year 1, about $0.10 per share, followed by 15% annual declines as servicing income keeps running off and the fixed preferred dividends and corporate costs weigh on a shrinking earnings base. The balance sheet is highly leveraged, with over $1.1 billion of repurchase borrowings against roughly $238 million of total equity including preferred, so a bad hedging or funding year can erase common earnings quickly. Every year from 2016 to 2025 was positive on our measure, so we do not assume a wipe-out. Bull case, $4.25 per share. Earnings hold at about $0.47 in year 1 and then decline only 5% a year. The evidence for this is that in 2025 net interest income on the bond book improved to $11.3 million from roughly zero in 2024 and a $1.7 million loss in 2023, offsetting a $5.2 million drop in net servicing income, and expenses fell after the internalization (no management fee in 2025). Such offsets would need to continue. What we tested. The main risk to the base case is that servicing decay is faster than 10%. The filing attributes the 2025 write-down of $18.8 million on servicing rights to changes in valuation inputs and paydown of underlying loans, so part of it is rate-driven rather than permanent, which is why 10% rather than something steeper is used. The common dividend has fallen from $1.08 per share in 2022 to $0.50 in 2025, and the first quarter of 2026 was declared at $0.10, consistent with the shrinking earnings base. First-half 2026 operating cash flow was $17.3 million versus $13.1 million a year earlier, which does not contradict the base case; we did not read the narrative of the latest quarterly report, only its reported figures. What would make this wrong. Stable or falling mortgage rates that slow loan payoffs, continued cost savings, or deploying new equity at attractive spreads could push results toward the bull case. Widening funding or hedge costs, faster prepayments on serviced loans, or another year like 2024 would push toward the bear case.