Does the tax benefit ever change what you'll pay for the dirt, or is that how people talk themselves into bad basis?

Slow question. If the benefit only pays off after ten years, how does anyone put a present number on it at all? You'd have to guess the appreciation over a decade and then discount it back, and both of those seem like numbers you can make say anything.

@juniper that's exactly the objection. You can build the model and the answer moves 40 percent if you change the exit cap by 50 basis points. Which is why I voted for zero. Anything I can't defend to myself in year one I don't pay for in year one.

Voted for the required return version because it's the only one where I can see myself cheating and catch it.

Everyone votes zero and then bids like it's 5 percent. I've watched it happen at three auctions.

Looking at my own behavior, the benefit doesn't change my price and it does change whether I chase the deal at all. If there are two sites and I can only underwrite one this month, the one in the tract gets the weekend. That's a real premium, it just shows up as attention rather than dollars, and attention is how you end up emotionally committed and paying up on Tuesday.

The premium isn't wrong in principle. It's wrong because of what you're buying when you pay it.

A benefit conditional on a ten year hold, on the tract being designated at the time you make the investment, on the fund passing its testing dates every six months, and on you actually hitting the substantial improvement standard within the window. Each of those is a condition you can fail, and paying a premium up front converts a conditional benefit into an unconditional cost. Whether any particular fact pattern satisfies those conditions is a question for a tax professional looking at your documents, and I've read enough offering material asserting things that turned out to depend on facts nobody had checked.

If you're going to pay a premium, the least bad way I've seen it done is paying it in the contract as a price bump contingent on the tract appearing on the final designated list, with the deposit refundable if it doesn't. Sellers hate it. Some of them sign it anyway.

@fathom that contingent bump is what I keep sketching and never proposing. What's the actual language people use to define "appears on the final list" in a way a title company and an attorney will both accept? The designation is at census tract level and the parcel to tract mapping isn't always obvious on a boundary lot.

On the passive side I see the premium already baked in before I ever get to vote on it. By the time a fund shows me a deal, the sponsor has bought the site and the price reflects whatever they were willing to pay for tract status.

So the diligence question I ask is what the site traded for and what comparable dirt outside the tract traded for in the same period. In a dozen or so states sale prices aren't public at all, which makes that check hard to run and is by itself a reason to ask the sponsor for the settlement statement rather than take their word for the basis.

Zero premium, and I don't find it a close call. The whole appeal of this to me was that it's a tax outcome sitting on top of an ordinary building that has to work like an ordinary building. The moment I'm paying extra, I've bought a tax product.

Let me put numbers on it so people can tell me what's wrong with them.

Say a 1.9m all in project, 8.2 yield on cost, and you hold ten years and exit at a 7 cap on rents that grew 2.5 percent a year. Rough exit equity gain over basis of maybe 700k. If the ten year treatment applies and eliminates the tax on that appreciation, and you'd otherwise be looking at a combined rate somewhere in the mid twenties on it, you're talking about something in the region of 175k of value, before any state treatment, which varies by state and in some places doesn't conform at all.

Discount 175k back ten years at 10 percent and you get about 67k. On a 1.9m project that's 3.5 percent of total cost, and if the land is 280k of that, the same 67k is 24 percent of the land price.

So the answer to bramble's question depends completely on whether you express the premium against land or against total cost. Both of those are real numbers and they sound wildly different in a negotiation. My discount rate is the soft assumption and I'd argue for higher than 10 given how many conditions @fathom just listed.

@ledger that framing is the most useful thing in the thread for me. Someone saying "I'll pay 20 percent more for the dirt" and someone saying "I'll pay 3 percent more for the project" can be the same person and they sound like opposites.

The bit nobody's priced is that a premium paid on land also raises the building basis you have to double under the improvement test if the allocation moves with it. Pay up for the dirt and your improvement number can go up too. Worth having whoever does your allocation say out loud how the extra dollars split before you agree to them.