Who bought HVLS 2026-1 — and what the bidding tells you

Most coverage of a HUD loan sale stops at the headline price. HVLS 2026-1 cleared at 69.4% of broker price opinion, and that number is worth knowing. But the results summary contains something more useful: a bidder-by-bidder breakdown showing exactly who won what, and in which states.

Read that way, the sale is not one market. It is at least three, and the buyers in each are doing different things with the same collateral.

Two buyers took half the sale

GITSIT Solutions won 309 loans. CFS15 Grantor Trust won 200. Between them that is 509 of 1,061 — 48% of everything sold, before the other twenty buyers took anything.

The tail is long and thin. Eight buyers took ten loans or fewer. Three took five. Whatever else this sale was, it was not an even distribution among equals.

The 18% bid

Bids ranged from 18% of BPO to 78%. The median was 69.5%. Twenty of the twenty-two buyers landed between 55% and 78% — a tight band, and roughly what you would expect for vacant collateral with title work ahead of it.

Then there is WFL Homes LLC, which won 15 loans at 18% of broker price opinion. The next lowest bid in the sale was 55%. That is a 37-point gap from the nearest buyer and more than 50 points below the median.

On $1.2M of BPO across 15 loans, an 18% bid implies roughly $220,000 — about $15,000 a loan. For comparison, GITSIT paid an average near $135,000 a loan against the same kind of collateral.

The split that actually matters: aggregators and specialists

The most useful column in the results summary is not price. It is the count of states each buyer ended up holding loans in — because that determines what they can realistically do next.

Carrington Mortgage Services won 58 loans in exactly two states: 39 in Florida, 19 in Maryland. Phalanx Capital Series won 7 loans, all of them in Puerto Rico. Fairworth 365 took 5, four in Maryland and one in DC. These are deliberate, concentrated positions in markets the buyer presumably already works.

GITSIT Solutions won 309 loans across 41 states. Its largest single state, Maryland, accounts for 11% of its book. It holds one loan each in Alaska, Hawaii, Vermont, Wyoming, Colorado and New Hampshire.

Both strategies are rational. The specialist knows one foreclosure timeline, one set of county clerks, one bench. The aggregator is buying a statistical outcome and does not need to care where any individual asset sits.

But they produce very different problems eighteen months later, when the loans have resolved and the collateral is a house.

What happens to a scattered book

A buyer holding 39 Florida loans can build a process: one attorney, one title company, one contractor, one listing relationship. The thirty-ninth file costs a fraction of what the first one did.

A buyer holding one loan in Wyoming cannot. There is no process to build and nothing to amortise it against. Every step — local counsel, a property inspection, a broker who will take a single listing four hundred miles from their office — is priced as a one-off.

Five of the twenty-two buyers in this sale hold fewer than two loans per state on average. Several of the largest hold long tails of single-asset states beneath a concentrated core. GITSIT’s 41-state book contains a substantial Maryland, Texas and Alabama position and then a scattering of individual files in places it will never build a process for.

The nonprofit presence

Two mission bidders appear in the results. LLACG Community Investment Fund won 73 loans — the fourth largest position in the sale — at 73% of BPO, above the median. The Headlands Foundation won 9 at 64%.

LLACG’s book is notably concentrated: 41 of its 73 loans are in Texas, with Florida and Georgia making up most of the balance. That is a specialist pattern, not a scattered one, and it is consistent with an organization working defined communities rather than buying a national statistical outcome.

Together the two nonprofits took 82 loans, 7.7% of the sale. HUD’s stated ambition for mission participation is considerably higher, which is the subject of a separate piece in this series.

What to take from it

Three things are worth carrying out of this results summary.

Concentration is high and rising in importance. When two buyers take 48% of a sale, the disposition behavior of those two buyers determines what the secondary market looks like for everyone else.

Price dispersion is enormous at the edges. A 60-point spread between the lowest and highest bid on nominally similar collateral means the market has no shared view of what these files are worth — only of what each buyer can do with them.

Geography predicts the tail. If you want to know which assets will come loose in eighteen months, look at which buyers ended up with one or two loans in states where they have no operation. That list is available in the results summary, and it is the same list every sale.

All figures from the FHA Office of Asset Sales HUD-Held Vacant Loan Sale 2026-1 Sale Results Summary, sale date December 9, 2025, transaction specialist Falcon Capital Advisors LLC. Loan counts, bid percentages of BPO, and state-level distributions are as reported in Tables 2.1, 3.1 and 3.2–3.23. Per-loan dollar figures are derived by applying each bidder’s reported bid percentage of BPO to their reported BPO value and dividing by loan count; they are approximations and are rounded. Loans-per-state figures are calculated from the per-bidder state tables. Actual loans delivered at settlement may be fewer than offered, per HUD’s note on qualifying status. Nothing here is investment, legal or tax advice.