Do you underwrite every certificate as a property purchase, or price it purely for yield?

Question

Reading this room for a few weeks and there are two clearly different practices in it and both are defended by people who've done more than I have.

The first says: most certificates redeem, so this is lending. Screen for redemption probability, price the interest, keep positions small enough that any single one going long doesn't matter, and if a parcel ever comes to you, treat it as a windfall you didn't underwrite for. The advantage is speed and volume. You can look at 300 parcels in a weekend if all you need is a redemption signal, and the statutory rate is the same whether the collateral is a nice house or a strip of dirt. The problem is the tail. Threads here have people holding positions they can't enforce economically and parcels nobody would want, plus you're competing hardest on exactly the parcels everyone else has screened as safe, which is where the bid-down grinds the rate to nothing.

The second says: assume you might own it, so underwrite the asset. Full parcel work, access, structures, code cases, whatever the state's rules are about what survives the sale. You'll bid on twenty parcels instead of three hundred and you'll pass on most auctions. The advantage is that the tail stops being a threat and becomes the reason you're there, and you're bidding where competition is thinner. The problem is that if 90-plus percent redeem, you've done full property diligence hundreds of times to buy a bond, and your cost per dollar deployed is enormous.

There's an in-between that I see less often stated openly, which is running two books on purpose with different rules and different money.

I don't have a position. I'd like to know which one people who are actually placing money use, and whether the answer changes with the size of the check.

angela_burke · Accepted Answer

The split you've drawn is real, and the thing that decides it isn't philosophy, it's the pricing mechanism in the state you're bidding in.

In a bid-down-the-rate state, the market prices redemption safety directly. The safest collateral clears at rates that barely cover fixed per-certificate costs, which several people here have found out with actual money. So underwriting purely for yield in a competitive bid-down county puts you in the worst part of the curve. In a premium-bid state, or over the counter at the statutory rate with no bidding, the same yield-first approach works fine because you're paid a fixed rate and the only question is whether the owner pays.

So my practice: yield-only where the rate is fixed and I'm not competing it away, asset-underwritten where the rate gets bid down, because in those counties the parcels left at a decent rate are the ones with a problem and the problem is what I'm being paid to evaluate.

Check size changes it too, and not the way people assume. Small money can afford full asset diligence because you're buying few positions. Volume forces you toward screens, and screens push you back toward the crowded end. That's the ceiling on this strategy for individuals, and it's procedural rather than financial.

Running two books works if the money is genuinely separate and the deadlines are tracked per position, which is where the failures in this room come from. One book with two purposes turns into neither.

tashaford30 · Answer

@cobalt's split by pricing mechanism is more useful than the framing I came in with, which was that asset underwriting is just the careful version.

For my first purchases I'm going to do full asset work on a handful even though I expect them all to redeem, because I want to find out what I miss while the stakes are small. Calling that tuition rather than a strategy.

Do you underwrite every certificate as a property purchase, or price it purely for yield?

2 replies