The infill math stops working at $42k a home, so what am I missing

Your pad-level math is doing what it should. At $420 lot rent and a high-5s exit cap, a $92k new home doesn't pencil on lot rent alone, and that's the standard result. Infill at scale doesn't get rescued by the lot rent. It gets rescued by three other things and if none of them are present in those six parks you should stop.

First, the home sale itself has to return most of the capital. Operators running this well sell the home at $75k to $95k on a note at a rate that recovers the cost over the term, which means the home is a self-liquidating investment and the lot rent is the actual yield. That business requires retail installment sales licensing and consumer lending compliance that varies by state, and the disclosure and servicing rules are strict enough that you want counsel before the first contract, not after. Some states also treat the dealer function separately from the lender function.

Second, manufacturer programs. If you're at $88k for a single-section, you're buying like a one-off retail customer. Portfolio buyers negotiate at volume with a factory, standardize to two or three floorplans, and use a captive transport and set crew. The set and finish work is often a third of your number and it's the part that compresses most.

Third, the market has to actually be undersupplied at that price point. A $395 to $440 lot rent market with 470 vacant pads may be telling you the demand isn't there, in which case infill is expensive vacancy conversion into a market that doesn't want it.

The risk you haven't priced: repossession. When you finance homes to residents at scale, some percentage comes back. You then own a used home on your own pad, you resell it, and the recovery rate on those is the number that decides whether the whole program works. Model it at a default rate you'd be embarrassed by and see if it survives.

The home note book is the part lenders will look at hardest. Agency lenders on the pad debt generally don't want park-owned homes or a large chattel receivable sitting inside the borrowing entity, so you end up with the home program in a separate affiliate with its own line, priced like consumer paper rather than real estate. That separation is not free. Ask your lender in writing how they treat the affiliate for covenant purposes before you build the program around it.

Rerun it with the note. $92k cost, sell at $85k on a 15-year note at whatever rate is legal in that state, you get roughly $700 to $800 a month in principal and interest plus $420 lot rent. The lot rent capitalizes at your 5.75 and creates the $87k. The note is a separate return stream on capital you got back over time. Your original math treated the home as a permanent capital outlay, which is where the deal died. The question is whether you believe a 580-score borrower pays for 15 years.

@ledger that's the version I've seen in decks but I've never seen anyone show me actual seasoned performance on those notes. Everyone models it and nobody publishes it.

Your $5,000 of NOI per filled pad is too high. You add $420 a month gross, $5,040 a year, and then you carry incremental expense on that pad. Water and sewer if you're not submetered, trash, more wear on the roads, higher management load, plus property tax reassessment once the pad is occupied in some jurisdictions. Call it $1,100 to $1,600 of expense and you're at $3,400 to $3,900 of NOI, which is $59k to $68k of value at your cap. The gap versus $92k is wider than you wrote.

The operational side of infill is what nobody budgets. Six parks, 470 pads, you're running a construction schedule with transport permits, set crews, utility inspections and a county that will slow-walk the electrical sign-off. On the parks I've managed through this, the realistic pace was four to six homes a month per park with a dedicated person, and if you don't have that person you get two. At two a month you're looking at years, and your fund has a clock.

@sextant is right about reassessment, that one bit a portfolio I looked at in Florida. Also worth pulling: in those six parks, what happened to the previously occupied pads that went vacant? If homes were removed because they were unlivable, fine. If residents moved homes out, that's a market signal and infill is fighting it.

So the honest read of this thread is that infill is a home finance business wearing a real estate hat. That reframes what an LP is actually underwriting when they commit to Fund II, and I'd want it said out loud in the PPM rather than buried under value-add.