Do you price every certificate to redeem, or price every certificate as if you end up with the parcel?

Question

I'm sizing a first real allocation and the bidding rule is the thing I can't settle. Two coherent ways to run it and they produce different bids on the same certificate.

One: price to redeem. The statutory rate is the product, redemption is the base case, and the small share that goes to deed is a cost of doing business you absorb across the book. This lets you bid aggressively on urban parcels with clean title where the owner has every reason to pay, and you accept that a handful of ugly outcomes get written down. Deployment is faster and you don't spend research dollars on properties you'll almost never own.

Two: price to the deed. Every certificate has to clear an underwriting test on the collateral itself, so you only buy where taking title is an acceptable result. Your bids come down, you pass on a lot of paper, and your yield is lower on the winners. In exchange you never wake up owning something with a demolition order on it.

The first approach lives or dies on your assumed deed conversion rate and your assumed severity when it happens. The second gives up yield to buy certainty. I've seen both defended by people running real money, and I notice the funds I've looked at mostly claim the first while their tape looks like the second.

Redemption periods and the notice steps before deed differ by state, which changes how expensive the second approach is to run.

dmitri_v6 · Answer

The part that decides this for me sits in the statute, not the spreadsheet. In some states the path from certificate to deed is an administrative application with a fixed notice sequence, and in others it's a judicial foreclosure with a real docket and real attorney time, so the cost of the deed outcome varies enormously by state and you should confirm the current sequence with counsel licensed there.

If your deed path is cheap and mechanical, pricing to redeem is defensible because the tail costs you little. If your deed path is a lawsuit per parcel, then a 4% conversion rate on 500 certificates is 20 lawsuits and you have just built a litigation department by accident. So my answer is that the bidding rule follows the procedure, and anyone who runs one rule across six states hasn't read the six statutes.

@meridian on your last point, the funds claiming one thing while behaving like the other is usually a research budget question. Nobody wants to tell LPs they spend $60 a certificate on title work.

Rena Expert · Answer

The choice depends on one number you have to commit to before you bid: expected conversion to deed, and the severity when it happens.

Work it as an expected value per certificate rather than as a philosophy. Suppose you win at an 8% statutory rate, average certificate size $2,400, average time to redemption 14 months. That's roughly $224 of interest per redeeming certificate before your cost to acquire and service it. Now put a conversion rate on it. At 3% conversion, 97 of every 100 certificates carry the book, and three have to be either sellable at or above your basis plus carry, or written to near zero. If severity on those three is a full loss of basis, you're giving back about $72 per certificate across the block against $224 of gross interest. That's survivable. At 8% conversion with full severity you're giving back $192 and the strategy is a coin flip that also costs you labor.

So pricing to redeem is a bet that conversion stays low and severity stays partial. Pricing to the deed is a bet that you can't forecast either one reliably enough at your size. Neither is wrong at the level of principle. The distinguishing evidence is your own history in a specific county, because conversion rates cluster by property type and by county far more than they cluster nationally.

What breaks the redemption-first version is buying certificates where nobody wants the parcel, since there the owner has no incentive to redeem and no buyer exists after deed. Screening those out is cheap. Full deed-grade underwriting on every certificate is not. Most scaled operations end up with a screen, then deeper work above a dollar threshold.

tylerbrooks9 · Answer

If I can't name who would buy the parcel after a deed, I don't want the certificate at any rate. That's the whole rule for me.

marcus_hale41 · Answer

@verity that rule is clean and it costs you a lot of good paper. Some of my steadiest redemptions have been on parcels I would genuinely hate to own, an over-improved house in a stagnant small city, taxes paid on month 11 every time because the owner's lender advances them. The lender is the real credit there, not the dirt.

Which is my actual answer to @meridian. The question isn't redemption versus deed, it's whether somebody other than the owner has an economic reason to cure. Escrowed mortgages, active HOAs, commercial tenants with tax clauses. I sort my list by who else gets hurt if the lien goes unpaid, and I'll bid down further where that answer is strong. On the parcels where the answer is nobody, I underwrite the collateral hard and bid like I'm buying land.

Two sleeves, two rules. Same portfolio.

dpatel_re · Answer

@canton's split is what I run, though I'd say the sleeve boundary is the hard part and nobody talks about it. My cutoff has been assessed improvement value as a share of total assessed value, above 55% goes in the redeem sleeve, below goes in the deed sleeve with a title read and a look at access. Crude, cheap, holds up better than anything else I've tried at 200 plus certificates a season.

Where it fails is condition. Assessment rolls are stale in a lot of counties and an improvement can be worth negative money. I ate that once on a block of eleven certificates that all screened fine.

@rena-expert the point about conversion clustering by county matches what I see. My conversion across one metro county has run under 2% for three years. In the rural counties on the same list it's been closer to 9%, and the severity is what actually hurts, because the resale market is thin enough that I'm not clearing basis plus two years of carry.

Do you price every certificate to redeem, or price every certificate as if you end up with the parcel?

5 replies