Does a 65 percent LTV actually protect me if the appraisal is two years old

The assumption doing the most work in your comparison is that appraisal recency is a reliable proxy for accuracy. It usually is, but the Carolinas bridge book still needs the same stress test you applied to the Austin office loan. Eight-month-old appraisals on short-duration bridge are genuinely better than two-year-old peak comps on office, but ask what the bridge exits assume: if the borrower's refinance out depends on cap rates that have moved 75 to 100 bps since origination, the exit LTV on a current-value basis may have drifted more than the appraisal date suggests.

The risk you haven't mentioned is loan-level extension concentration. If a material share of the Carolinas book has already extended once, the appraisal date is increasingly fictional regardless of its vintage. Bridge loans that don't exit on schedule accrue duration risk that the original appraisal was never built to cover. Ask for the percentage of the portfolio that is on first or second extension, and what the fund's appraisal refresh policy is at extension.

On the 9.2 percent target: that's the number you're being shown, and the spread above a fund with tighter collateral quality has to compensate for something. The question is whether the spread is priced for the actual adjusted-LTV distribution or for the tape LTV. Those are different things, as your own recut on the Austin loan shows.

One structural point worth confirming with your attorney or financial advisor: how loss allocation works at the fund level if a subset of loans goes non-performing. That's a legal and structural question, not a math one.

What does the fund disclose about its appraisal refresh triggers, specifically whether extension events require a new or updated appraisal before the loan is modified?

The Austin office piece is the one I'd lose sleep over specifically. Office in that submarket saw cap rate expansion of 150-plus basis points between mid-2022 and early 2024, and that alone compresses values pretty hard before you even touch occupancy trends. A 78-80 real LTV on an office bridge loan with no stabilized rent roll behind it is basically mezzanine risk dressed up as senior debt pricing. The 9.2 net sounds decent until you think about what recovery looks like if that loan goes sideways and the fund has to work through a Texas foreclosure timeline, which is actually faster than most states but still costs real money and time. The Carolinas book being eight months current matters a lot more than the 6-point LTV difference on paper.