Tax liens at scale: is the business bid discipline or servicing

Servicing, and in rural counties it is not close. Rural certificates redeem slowly and a real share go all the way to deed, which means the operator inherits parcels and has to do something with them. A tight bid on a parcel nobody will buy at any price is still a bad outcome. Ask any fund pitching this what happened to its deed pile.

I'm new enough that I had to look up what bidding the rate down even means, so take my vote as a beginner's one. What convinces me on the servicing side is that the bid is a single decision made once and the servicing is hundreds of decisions made every month for years. More chances to be wrong. Though I suppose a disciplined bidder never gets into the bad positions in the first place.

@bmoreno23 that's the part I'd push on. The bid isn't one decision, it's one decision repeated across every certificate in the sale, and the failure mode is systematic rather than random. If someone's ceiling is set two points too high they overpay on the whole book at once, and no amount of clean calendar work recovers it. Servicing errors are usually a slice. Bidding errors are the whole thing.

So I'd vote bid discipline, with the caveat that I want to see the walk-away rate. How many lots did they bid on and lose. An operator who wins everything they bid on is telling me their ceiling is soft.

The walk-away rate is a nice ask, @jordan_ellis2, I'm stealing that. What I'd add is that the two halves feed each other. An operator with strong servicing can bid on paper in counties where the process is slow and messy, because they can carry the timeline. An operator with weak servicing has to stay in the easy fast counties, which are exactly the ones where competition is worst. So the servicing capacity sets which auctions you can compete in at all. That's why I'm voting equal weight.

The invisible until it fails part is real, and it fails in a specific way. It does not fail on the certificate that redeems. It fails on the twenty percent where something is unusual, a bankruptcy filing, say, or a competing lien, or a notice address that bounces. Those need a person who knows the statute in that state, and states differ a lot on notice and redemption mechanics, so you want somebody licensed reviewing it locally. The question worth asking any operator is what the exception process is, rather than what the normal process is. Same question here. Find out who touches a file that goes off script and how fast.

@angela_burke yes, and the numbers hide there. I've watched a small block of certificates where the gross statutory rate looked great and the net came in about four points lower once you counted the legal work on the handful that didn't behave. Nobody's marketing deck breaks out cost per exception file, but that's the line that separates a real operation from a hobby with a lot of money in it.

I'm voting servicing. But my honest reason is that I can verify servicing by asking specific procedural questions and I can't verify bid discipline without seeing every sale they didn't win.

@samantha_ruiz that last sentence is the practical answer to the poll. Verifiability is not the same thing as importance, and the room seems to be drifting toward voting for whichever skill can actually be inspected. Jurisdiction selection is where my vote went, and nobody has picked it yet. The statutory rate structure, the redemption period and the notice burden are all set by the state, and they vary enormously. An operator who picked six workable states has already decided most of the outcome before any bidding or servicing happens. The other two skills get exercised inside a box someone chose earlier.

@tashaford30 the next question is how they chose the six, because the honest reason is usually that a person on staff already knew those courthouses. That is fine. It is just not portfolio construction. On the poll, equal weight. @tonyv88, the thing to ask for is the redemption curve by vintage, month by month. An average hides whether a bad quarter came from overpaying or from missing a deadline. If they can produce that curve on demand it also tells you their systems are real, which answers half your question sideways.