When the exit cap assumption is carrying the whole model on a 12 house package

Consider a package from a retiring owner: twelve houses within about nine miles of each other, built between 1958 and 1979, sizes from 1,050 to 1,480 sq ft. Asking 2,640,000, so 220,000 a door. Current in-place rent across the twelve totals 21,300 a month, average 1,775, roughly 8 percent under market on eight of them from renewals that haven't been pushed in years. The seller's T-12 gross is 251,000 with about 96 percent collections. Reported operating expenses are 87,000, giving NOI of 154,000, a 5.8 cap on ask. That expense number excludes management, since the owner self-manages, and a maintenance line of 4,900 a house for that vintage doesn't hold up. A more realistic rebuild adds 10 percent management on collected, bumps maintenance and capex to 1,700 a door annually against roll, and adds a real vacancy factor of 7 percent rather than the seller's 4. That drops NOI to about 122,000, a 4.6 cap. Pushing the eight under-market units to market over 24 months adds maybe 14,000 of additional gross, call it 133,000 stabilized, a 5.0. Financing is the tight part. Debt at 65 percent, quoted around 7 percent on a five year term, runs 5,600 a month or so on 1,716,000. Coverage on 133,000 is about 1.18, which clears but barely. The model only produces an acceptable IRR with an exit cap at or below entry in year five. Every basis point of cap expansion eats roughly 26,000 of value across the package, and fifty basis points wipes out the entire rent lift and then some. That's a familiar problem, just unusually exposed here: the whole outcome rests on an exit cap nobody can forecast. The honest paths are pricing it as a 4.6 today and offering around 2,400,000, likely losing the seller, or passing.