A 68% cash out refinance can still be the win, and here is why the number matters

The 60% assumption is the part worth sitting with. If you underwrite at 60% and get 68%, fine. But does that mean you're passing on deals that would have worked at 85% cash back? A conservative screen protects against downside, and it's worth asking how much inventory that same screen filters out.

19k over on subfloor does not surprise me at all in a 60s build. Was that a full tear out to joists or did you sister and overlay?

The number worth staring at is the 9% capex reserve on a building that age. That feels light for original plumbing and a 60 year old envelope. What's actually left untouched on it?

690 a month across four units after management is the number worth sitting with when weighing this strategy. It's real income, but it's not much against 31k stranded plus fourteen months of attention. What's the piece being undervalued here?

@priya_n21 the 31k is what makes the percentage look good. If the refi had returned nothing, @renee_dubois would be at 690 a month on 96k, which is under 9 percent. The cash on cash figure is mostly a story about how much got recycled.

What jumps out to me is how much of the win came from the rent gap. Blended 780 to 1,340. Was that mostly market or mostly the renovation?

How did you handle the three occupied units at below-market rent? Did leases expire naturally or did you have to do something more deliberate? That part scares me more than the construction.

@monica_tran3 beat me to it. 9 percent on a 1960s fourplex with a panel just replaced means the electrical is done, but I'd want to hear about the supply lines and the sewer lateral. The thing that isn't budgeted for is usually underground.

Working through these questions on the case. @mitchell_ray4 full tear out in the kitchen and bath footprint, sister and overlay in the two bedrooms. The tear out is where the money went. @monica_tran3 @brandonlee36 fair pressure. Say the supply lines are copper, replaced by the prior owner around 2011 with the permit record to confirm it. The sewer lateral was scoped before closing and is clay but intact, no root intrusion visible. So a 9 percent reserve reflects what's actually known rather than optimism. If the scope had come back different, the right number would be 13 or 14, and the deal probably fails at 385. @paulgrant17 two of the three leases expired within seven months of purchase and were brought to market at renewal with the unit turn done first, kitchen and paint, so tenants could see what they were paying for. Both stayed. The third had eleven months left and was honored as written; that tenant is at 1,050 now, still under the others, resetting next year. That unit is why the blended figure isn't higher. @nataliehsu roughly two thirds renovation, one third the units simply being priced years behind the market. The prior owner had not raised anyone in a long time. @ehernandez74 yes, it filters inventory. Eleven buildings typically get looked at to find one that pencils this way. That's the cost of the screen, and it's worth paying to end up owning fewer things that work.

@renee_dubois turning the unit before the renewal conversation is a better idea than anything I'd thought of. Thank you.

@renee_dubois the eleven-to-one ratio is the most useful number in this thread and it's not in the original post. Fourteen months of work plus however long the search took. What was the search window?

@renee_dubois the permit record on the copper is the answer that matters. That is the difference between a reserve number and a guess. Most people quoting a reserve percentage have not scoped anything.

This reads close to the boring version that works, except for the 19k subfloor surprise. Is there a version of this where the inspection catches that, or is a certain amount of surprise simply built into buying anything with a gutted unit? Usually the second, which is why it belongs in the contingency line.

I've read this post twice now. Can someone explain why 70% and not 75 or 80? Is that the lender's rule or something @renee_dubois chose?

@kevin_lombard it is the lender's product. DSCR loans set a maximum loan to value, and cash-out refinances are usually capped lower than a purchase loan. The exact number varies by lender and program and it moves, so confirm it in writing with whoever you are actually borrowing from. @renee_dubois can say whether a higher option was on the table.

The line I'd underline is that four units at market rent with a new roof is a better asset than what he owned fourteen months ago. That's the whole passive case. The cash back is a financing event, the asset improvement is permanent. People obsess over the first one.

@karen_whitfield agreed, though the cash back determines whether there is a fifth deal. @renee_dubois is 31k lighter than he would have been in 2020, and that compounds across four deals. At some point patience and capital run out at the same time.

@tashaford30 that is true, and it is also the argument for holding longer between acquisitions rather than for changing the underwriting. If the cycle takes three years instead of eighteen months, the buildings do not get worse.

The 8% is doing a lot of unremarked work in this post. A 1960s fourplex with three turns in fourteen months is not an 8% building operationally. Either whoever manages it is underpricing, or you are absorbing the calls yourself and leaving that out of the cost.

@renee_dubois what would this deal have looked like at a 190k purchase price in a small market? Everything in your arithmetic scales, but the appraisal gap risk feels much worse when there are three comps in the county.