Millrose Properties, Inc. (MRP) stock intrinsic value fair value estimate & owner earnings analysis

Millrose Properties is a land banker for homebuilders. It was spun out of Lennar in February 2025 and at June 30, 2026 held about $9.6 billion (at cost) of finished and in-development homesites that builders have the exclusive option to buy on set schedules. Instead of owning the land, a builder pays Millrose a monthly option fee on the capital invested (about 8.2% a year across the book in the first half of 2026) and then buys the lots at cost, returning Millrose's capital. Builders also post deposits, about $1.0 billion in total, which they lose if they walk away. Millrose is externally managed by Kennedy Lewis and pays it 1.25% a year of tangible assets. Lennar was 72% of revenue in the first half of 2026, down from 84% in 2025, as other builders were added. Millrose has also committed to fund about $6.9 billion more of development on its existing contracts, 74% of it with Lennar. How we measured earnings. Reported operating cash flow includes the cost of lots builders bought back, which is a return of Millrose's own capital rather than profit, so we removed it. On that basis 2025 owner earnings were about $417 million, or $2.51 per share, for roughly ten and a half months as a public company, and $2.94 per share before interest. That is a ramp-up year, so we do not use it as the level. The first half of 2026 is the better guide: option fees of $381 million, net income of $249 million and $79 million of interest, which is roughly $627 million a year of cash before lenders after the management fee and taxes, or about $3.78 per share. Two consecutive quarters at that pace, and capital that was only partly deployed during 2025, support the step-up. There is no earlier history: before 2025 the business was a carve-out inside Lennar with no revenue. What we credited. Only what is contracted: the $9.6 billion of option contracts on the books, the $6.9 billion of development Millrose is already committed to fund (it earns option fees on that money as it goes in), and capital returning through takedowns at the pace seen in the first half of 2026, about a third of the book each year. We did not credit Millrose reinvesting the returned capital into new contracts beyond those commitments. It has done that for about 20 months, never through a housing downturn, and it depends on finding new builder deals, so it earns no value in the base case. Base case: about $38.60 per share. The contracts produce fees after management costs and taxes, plus the return of capital, worth about $9.2 billion. Millrose owes about $2.8 billion: $1.25 billion of 6.375% notes due 2030, $750 million of 6.25% notes due 2032, $485 million drawn on its bank line at June 30, small property mortgages and seller notes, a $100 million holdback owed to Rausch Coleman, and the dividend and interest accrued at June 30, less cash. That leaves about $6.4 billion, or $38.60 per share, against book equity of about $35 per share. Interest on this debt runs close to 6%, so no adjustment was made for its cost. The base case includes the roughly $250 million of finished Lennar homes Millrose bought in September 2026 to rent out (see below), which earn less than they cost to finance and trim value by about 10 cents a share. Bear case: about $24.70 per share. Housing has cycled before, and land is where the losses concentrate. In 2007-2009 finished lot values fell about 40% from peak and raw land 60% or more, and Lennar itself sold a large land portfolio at about 40% of book in November 2007. The bear case runs that episode starting in the second year. Lennar uses both of its six-month pause rights, during which option fees are cut to half; takedowns slow to well under half the current pace for three years; and builders walk away from 40% of the capital, keeping Millrose's protection to their forfeited deposits. The abandoned land earns nothing, costs money to carry, and is sold over years four to seven at 55% of its cost. On the remaining contracts fees continue. The Dream Finders/Beazer commitment is funded just ahead of the downturn and suffers the same way, and the rental homes are sold at a 30% loss. The business is then worth about $6.9 billion, debt is $2.8 billion, leaving about $4.1 billion. Cash flow before lenders stays above roughly $350 million a year at the worst point, comfortably over the interest bill, and returning capital would cover the 2030 notes and the March 2030 bank line maturity. Debt would approach the 1:1 debt-to-equity limit in Lennar's agreements, which would stop new investment but not force a sale of shares. Bull case: about $44.75 per share. Millrose keeps recycling takedown proceeds into new option contracts for five years, as it did in 2025 and the first half of 2026, at a slightly higher blend as non-Lennar builders (who pay about 11%) grow; Lennar uses its right to extend takedown dates so capital earns fees about a year longer; and the up to $1.25 billion Dream Finders/Beazer commitment closes and earns the non-Lennar rate. Business value is about $10.2 billion; after the same $2.8 billion of debt, about $7.4 billion. Even this case adds only about $6 a share, because new capital earns only a little more than it costs to own. What changed since our previous analysis. Three things happened at the business. First, a short-seller report in early October showed that in September 2026 Millrose bought roughly 700 to 1,100 finished Lennar homes, about $200 to $290 million, and listed nearly all of them for rent through a property manager, while management had described the program to analysts as a rent-to-own pilot with tenants often lined up in advance. Lennar's agreement with Millrose was amended on August 27, 2026 to allow rental assets. The purchases helped Lennar meet its delivery target at a time it was struggling to sell homes; on the report's estimates the rents barely cover taxes and insurance and fall short of Millrose's borrowing cost. The amount is small against a $9.6 billion book, but it shows Millrose's capital can be used to absorb Lennar's unsold inventory, and none of this has yet appeared in a Millrose filing. Shareholder lawsuits followed. Second, Millrose committed up to $1.25 billion to support Dream Finders' purchase of Beazer Homes, its largest non-Lennar deal. Third, on October 6 it sold $1.0 billion of notes ($500 million at 6.50% due 2029 and $500 million at 6.75% due 2031, the latter to be redeemed if the Beazer deal does not close by May 13, 2027) and drew its $500 million term loan, largely to repay its bank line and fund the Beazer commitment. Our previous estimates were built on a narrower set of assumptions and are not directly comparable. Margin of safety price: $22. We are fairly sure of the earnings level, because it is set by contracts on the books and the first half of 2026 confirmed it. The worst plausible outcome is a 2008-style land bust in which builders walk from a large share of the pipeline; the structure would survive it, because the debt is modest against a book of land carried at cost, the notes do not mature until 2029-2032, and deposits absorb the first losses, but shareholders would lose about a third of book value. The things we are most likely to be wrong about are how much the relationship with Lennar is run for Lennar's benefit (the rental purchases are the first evidence), how fast capital comes back, and how deep a land downturn would cut; if all three go against us the shares are worth roughly the bear case, mid-$20s. Over the holding period, Millrose pays a dividend of $0.77 a quarter, about $3.08 a year, though that would shrink in a downturn. At $22 the dividend alone represents about 14% a year, so only a modest rise in value is needed for the cases to imply 20% a year, and even a value near the bear case within about two years implies close to that. Above that price, a single bad turn in housing or in the Lennar relationship would leave too little room. What would make this wrong: builders walking from options faster or at deeper losses than the 2008 record; Lennar continuing to steer Millrose's capital into low-return uses such as rentals; slower takedowns that keep capital tied up longer; and the external management fee, which is charged on assets whether or not they earn. We could not fully reconcile the 2025 schedule of contracted option fees, which implies capital returning more slowly than the first-half-2026 takedown pace; each extra year of average life moves value by roughly $1 per share. The rental purchases have not yet been reported by Millrose, so their size and returns rely on a third-party estimate until the third-quarter filing.