Sizing a mezzanine debt fund allocation when the attachment point sits on a marked-down 2021 vintage

A live allocation question worth working through. Two debt fund commitments due by quarter end, with a fixed amount of capital to split between them. Fund A: residential bridge, first lien. Coupons in the 10.5 to 11.5 range, average loan around 480,000, roughly 900 loans across 20 states, sized at 68 percent of as-is value with rehab held back in draws. Fees around 1.25 with no promote under an 8 percent hurdle, net target around 9. Reported cumulative loss rate in the mid-30 basis points since 2019. Fund B: CRE mezzanine behind bank seniors. Coupons around 13.5, roughly 34 positions averaging 6.8 million, attaching between 65 and 82 percent of appraised value. Around 71 percent of the book is multifamily originated in 2021 and 2022. Net target around 11.5. The arithmetic problem with B: if the value on a 2021-basis multifamily deal is down 15 percent from the appraisal that set the attachment point, everything above 65 percent is already impaired, and mezzanine is where that lands first, ahead of the senior. A manager's usual answer is that the seniors are fixed rate for several more years and the properties are cash flowing above a 1.15 debt service coverage ratio, so nothing forces a sale. That is true, and it is a story about time rather than about value. The problem with A is verification. Nine hundred loans cannot really be checked individually, so the allocation is a bet on a servicer's process, and a low loss rate during a period when home prices broadly rose says little about that servicer's competence in an actual workout. A reasonable default is to weight heavily toward A, treating B's extra 200 basis points of coupon as insufficient compensation until the 2021 vintage has more time to season and reprice. The open question worth debating: what evidence would actually justify taking that 200 basis points on B.

sonia_alvarez CB4APRO 17

Answer

stephen_ok9ActivePRO

Concentration alone answers a chunk of it. 34 positions at $6.8 million means one total loss is close to 3 percent of the book, and mezz doesn't lose 20 percent, it loses most of it or nothing. So the 200 basis points of extra yield buys you a return distribution with a fat left tail, and you'd want to see the manager's assumed recovery on a mezz position that goes bad. If they say 40 percent, ask which of their own workouts produced that, and if the answer is none, the 40 is a spreadsheet input.

The 1.15 coverage figure is the one I'd push on. Is that on the senior alone or senior plus mezz current pay? Those are different properties. If the mezz accrues rather than pays cash, your distribution is coming from somewhere other than property cash flow.

On Fund A, the thing you can actually verify is process at the exit. Ask for every loan that went past maturity in the last two years and what happened to it, month by month. Foreclosure timelines and whether the process is judicial or non-judicial vary a lot by state, and a fund spread over 20 states owns that variation, so ask which states hold the most principal and what their last completed foreclosure in each cost and took. A servicer who has never run one at scale in a slow-timeline state is the risk you are underwriting, more than the 68 percent.

stephen_ok9ActivePRO

2 replies

jordan_ellis2CapitalPRO

I've been on the borrower side of quotes from bridge lenders like your Fund A, and the draw process is where the discipline shows. Ask how many inspections per draw, who does them, and whether the lender ever releases a draw on photos. The sloppy ones fund ahead of the work and their as-is cushion is fiction by month four.

jordan_ellis2CapitalPRO

Reply Anonymous