Where do you put the downside in a hotel model: the operating case or the exit?

Question

Working through a five-year hold on a repositioning, upper-midscale, 160 keys, secondary market with one anchor demand driver. The sponsor's base case has RevPAR growing every year of the hold and a downside case that's just the base case with 200 basis points shaved off growth. Nobody in either version has a recession in it.

I've seen three approaches to fixing that and they produce different deals.

One is to model the trough explicitly, put a 15 to 20 percent RevPAR decline into year three or four with the margin compression that goes with it, and see whether coverage survives it. That's honest and it wrecks most five-year IRRs because the recovery eats the back half of the hold.

Two is to leave the operating case alone and take the pain at the exit, widen the exit cap 75 to 100 basis points over going-in and stop pretending you know when the cycle hits. Cleaner, but it lets the deal keep distributing all the way through a period where it plausibly wouldn't.

Three is to stop adjusting the projection and adjust the balance sheet instead, lower going-in leverage, longer or more flexible debt, real liquidity held at the fund level so a soft year is survivable rather than fatal.

The first two are about being right. The third accepts you won't be and pays for the ability to wait. I don't think they're substitutes, and the sponsors I've talked to treat them as if they are. What do you actually make the deal answer for?

bmoreno23 · Answer

I'm new enough that most of this is over my head, so I'll ask the dumb question. If option one wrecks the IRR on almost every deal, does anyone ever use it? Or does modeling a recession just mean you never buy a hotel, which can't be what people do either.

The part I did follow is your last line. If the sponsor's downside case is the base case minus a bit, they haven't modeled a different world, they've modeled the same world going slightly worse. Even I can see those aren't the same thing.

tylerbrooks9 · Answer

@ember, it does get used, and the way it stays usable is that you don't judge the deal on the IRR that comes out. You judge it on whether the property is still yours at the end. Run the trough, look at debt service coverage in the worst quarter, look at whether the loan has a cash trap or a covenant that trips, look at what the sponsor has to do if the lender sweeps cash for six quarters. If the answer is a capital call from investors who may not fund, that's the real finding and the IRR number is irrelevant.

@pike, I'd vote for your third one, and I'd say the exit-cap approach is the one I trust least. Widening the exit cap is a single number that makes a spreadsheet feel prudent and changes nothing about whether the deal survives year three. You can lose the asset before you ever get to test the exit assumption.

luis_carrera · Answer

@verity is right that the exit cap is doing cosmetic work in most decks, and I'd still keep it, because the two failures are different. The trough case tells you about survival. The exit assumption tells you about the price you're paying today, and hotels get bought at going-in caps that assume the buyer five years out is looking at the same travel demand you are.

What I've settled on after too long staring at these is stress the operating case and the exit at the same time, then stop tuning. One trough year, margin compression heavier than the revenue drop because the fixed cost base doesn't move, exit at a cap wider than going-in, and no bailout from a refinance. If it clears at moderate leverage, the segment and the operator have to carry the rest. If it only clears when you assume a good economy the whole way, you've written a bet on the cycle and priced it as if you hadn't.

On your 160 keys with one anchor demand driver, I'd want to know a lot more about that anchor than about any of the three modeling choices.

Where do you put the downside in a hotel model: the operating case or the exit?

3 replies