Franklin BSP Realty Trust has spent the past four quarters reshaping itself around a quieter, more capital-disciplined business. The June 30, 2026 10-Q reports $6.4B of total assets, a $4.3B core mortgage portfolio of 172 loans, 80% multifamily, and a $59.8B Agency servicing book acquired through NewPoint that now drives recurring fee income. The company closed a new $880.4M CRE CLO in the second quarter at an 88.4% advance rate, repurchased 1.84M shares at an $8.70 average price, and lifted fully converted book value by $0.06 to $14.24 per share. Net interest income of $29.9M and distributable earnings of $28.3M ($0.25 per fully converted share) show the core engine still generates cash above the $0.20 dividend, but the headline return on equity of 3.3% and 2.7% for the quarter and year to date underline how much of the earnings story is being absorbed by credit costs and acquisition overhead.
The investment debate turns on whether the credit cycle is cresting. Non-performing loans reached $344.2M of amortized cost across nine positions at quarter end, up from $214.0M at year end 2025, and the specific allowance on one North Carolina multifamily loan alone consumed $13.2M of provision in the first half. With core portfolio risk rating now at 2.4 (down from 2.5), twelve loans on watch list, and a $54.5M CECL reserve against $4.3B of loans, the portfolio is not broken, but it is no longer pristine either. The stock trades at $8.43, a 41% discount to fully converted book value, with a 13.1% indicated dividend yield and a 0.57x price to book ratio that reflect market skepticism about the durability of distributable earnings once realized credit losses flow through.
The strongest argument for the stock is that the market is paying roughly 7x forward distributable earnings for a self-amortizing mortgage book that is now 80% multifamily, where the new origination platform is generating 238 basis point spreads on committed loans and where each accretive share repurchase at a 40% discount to book value compounds book value per share. The strongest argument against it is that the credit cycle has not bottomed. If non-accrual loans continue to grow and the $344.2M of non-performing exposure starts to resolve with principal impairment rather than cure, the dividend coverage and the book value path both deteriorate quickly. The next two quarters of credit migration and the pace of REO liquidation are the variables that determine whether the discount closes or widens.