This week, we're talkin' self-dealing in the subdivision: how Lennar appears to have sold more than a thousand finished houses to a company its own chairman controls, and why some of those houses are sitting empty in the Nashville area. On the eve of Election Day, I walk through Florida's property-tax cut, who may end up paying for it, and why Oracle's Nashville groundbreaking matters more to local investors than the headlines suggest. I'll also explain why, despite some echoes of the dot-com era, this is not 2000.

Let's get into it.

Today's Interest Rate: 7.27%

(👇 .06% from this time last week, 30-yr mortgage)

Mortgage rates spent October quietly giving back ground. The Mortgage News Daily 30-year index closed Friday at 7.27%, its lowest in more than a month and down from a 7.61% peak on October 5 (Mortgage News Daily).

Things are good! Initial jobless claims fell to 194,000, hovering near their lowest level in 57 years (Reuters via Times of Oman). Yes, gas prices still stink. The national average is $4.17 a gallon, down from $4.40 a month ago but well above the $3.04 of a year ago (AAA). I think prices keep falling, with plenty of volatility along the way, and get back to pre-conflict levels within six months.

The Weekly 3 in the News:

1) Americans vote tomorrow in midterm elections for all 435 House seats, 35 Senate seats and 36 governorships. The Senate total includes 33 regular contests plus special elections in Florida and Ohio (Mappr). Housing costs are on the ballot directly in several states: voters in seven states will decide 13 property-tax measures this year, according to Ballotpedia (Ballotpedia)(Florida's and Tennessee's get their own section below, keep reading).

2) International Story - We can learn a lot from this here in the US…Dutch landlords are paying tenants tens of thousands of euros, and in rare cases more than €100,000, to move out so they can sell vacant. Jelmer Schortinghuis, owner of Proranje Vastgoed, a firm that buys occupied homes, told NL Times that more than half of the private investors who contacted his company in the third quarter had already tried paying tenants to leave, and that seller inquiries rose more than 31% from a year earlier (NL Times). The driver is a wide gap between what vacant and occupied homes sell for, after a 2024 “Affordable Rent” law extended rent caps into the middle of the market (DutchReview), and uncertainty from a new wealth tax grew. California, Seattle and New York, are you listening?

3) The Federal Reserve held the benchmark rate at 3.75% - 4.00% last Wednesday. Chair Kevin Warsh told reporters that while financial conditions have tightened, he was "unwilling and unable to call them tight" (National Mortgage News), adding that the economy was "expanding at a strong pace". Fed funds futures now put the odds of a quarter-point hike at the December 9 meeting at 74.6% (CME FedWatch). I still expect inflation to cool when/as oil prices fade.

Bonus: Apartment lenders pulling back. The National Multifamily Housing Council's October survey of apartment executives found its debt-financing index collapsed to 10 from 46 a quarter earlier, and its equity index fell to 17 from 44; 81% of respondents said now is a worse time to borrow, and 61% saw fewer property sales (NMHC). That is the weakest reading for capital since late 2023, and it lands just as rents are on the upswing. Historically, the best apartment purchases have this setup. Anecdote: I’m starting the search for my next apartment building.

Source: NMHC Quarterly Survey of Apartment Market Conditions, January 2022 to October 2026. Readings below 50 mean conditions worsened.

A Few Fun Things Happening in Nashville This Week

• "Star Songs" at Vanderbilt's Dyer Observatory, Saturday, Nov. 7. Jazz and stargazing on the hilltop above Radnor Lake, timed for the new early sunsets. (Dyer Observatory).

• Iron Fork, Thursday, Nov. 12, 6 to 9 p.m., Harken Hall, 514 Madison Station Blvd., Madison. The Nashville Scene's chef showdown, with tastings from area restaurants; proceeds support Celebrate Nashville, the Centennial Park Conservancy and Second Harvest Food Bank (Big Tickets). It tends to sell out!

Self-Dealing in the Subdivision: Lennar Sold 1,233 Homes to a Company Its Chairman Controls

Circular dealing appears to have spread to the homebuilders.

At the end of Emerson Lane in Westmoreland, Tennessee, a town a little north of Gallatin in Lennar's Pleasant Grove Farms community, three new houses were deeded on the same day this summer, to the same buyer.

The buyer was not a young family or a retiree downsizing from Hendersonville. It was MRP Liberty LLC, a rental entity of Millrose Properties, the land-banking company Lennar spun off in February 2025.

According to a public dataset assembled from county deed records and rental listings, the three homes sold for between $313,632 and $325,330, between 5% and 8% above their last list prices, and within weeks all three were advertised for rent at $1,755 to $1,850 a month (Public-CRE-Data).

Three houses in Sumner County would be a footnote, but multiply them across 18 states and you have one of the more instructive stories in housing this fall. In the final days of Lennar's fiscal third quarter, which ended August 31, Millrose began buying completed Lennar homes in bulk. By the time it reported earnings on October 27, it owned 1,233 of them, purchased for about $333 million (Millrose Q3 release). Lennar, meanwhile, reported 20,840 deliveries for the quarter, just 340 above the bottom of its guidance range of 20,500 to 21,500 (Lennar Q3 8-K).

Here’s the part that starts to get fishy.

Stuart Miller is Lennar's executive chairman and chief executive. Through a class of high-vote Millrose shares, he also controls roughly 43% of Millrose's voting power, according to my calculation from the company's proxy, which says he "had the power to cast 117,545,904 votes" (Millrose proxy). When one company sells inventory to another company whose largest voting holder runs the seller, finance has a name for it: a related-party transaction. When the money loops back like this, people start calling it circular.

I want to be careful here. Nothing I have read shows that anyone broke a law, and Lennar and Millrose have strong arguments on their side, which I lay out in full below. But as an investor and a broker who prices homes in these exact subdivisions, I think the arrangement may matter to you in three concrete ways: what it says about the real health of new-home demand, what it does to appraisals and comparable sales in Lennar communities, and what it adds to the supply of rental homes competing with yours.

This looks like numbers padding to me. And this type of self-dealing boosts financial exposure.

How the Deal Was Built

Let’s go another layer deeper.

The paper trail begins with a Securities and Exchange Commission filing on September 1. In it, Millrose disclosed that on August 27 it had amended its founding agreement with Lennar. The filing's cover text described the change in dry terms, as adjustments to a "Priority Amount" and to Lennar's consent rights over collateral; the words "rental" and "build-to-rent" appear only in the attached exhibits (Millrose 8-K).

The exhibits are where the substance lives. The amendment created a new asset class called "BTR Properties," defined as "completed single family homes" bought by Millrose and "intended to be operated … as rental properties." Money raised to buy them was carved out of the capital pool on which Lennar holds priority rights, and Millrose was permitted to pledge the homes for separate financing (Exhibit 10.1). A companion amendment with Millrose's external manager, an affiliate of the investment firm Kennedy Lewis, added rental operating costs to the expenses that shareholders reimburse (Exhibit 10.2). Neither document says what price Millrose would pay for the homes.

The buying started the next day. Hunterbrook, a research outlet whose affiliated fund disclosed that it is short both Lennar and Millrose, combed through 6,080 Lennar deeds across 47 counties and counted at least 356 sales to MRP Liberty in the last week of August alone (Hunterbrook). The independent deed dataset I mentioned tells the same story with more precision: of the 1,107 Millrose rental homes it has matched, about 93% were recorded in the three weeks surrounding Lennar's August 31 quarter-end, and recordings had slowed to a trickle by early October (calculated from Public-CRE-Data).

Millrose's own description, when it finally arrived on October 27, was measured. The company called the portfolio a pilot "underwritten as a standalone rental investment," funded with about $120 million of cash and $213 million of limited-recourse debt at the secured overnight financing rate plus 1.20%, and it put the average price at "approximate $270k" per home (Millrose release; Millrose deck). On the earnings call, management said about 10% of the homes were leased, at rents about 10% above what it had underwritten, and that it expected full lease-up to take about six months (call summary). Chief executive Darren Richman described the approach as "crawl, walk, run," and framed the homes as a test for a future rent-to-own product rather than the start of a large landlord business.

Why the Timing Matters

Context explains the urgency. Lennar's third quarter was difficult by any measure. Net earnings fell to $284 million from $591 million a year earlier, orders of 20,879 homes missed the company's own guidance, and the average sales price slipped to $372,000 from $383,000 (Lennar Q3 8-K). Gross margin on home sales was 15.8%, down from 17.5%, and that was with incentives running at about 12% of the sales price. Miller noted demand from "single-family for rent" and "build-to-rent" buyers and said Lennar had cut its completed, unsold inventory to 1.8 homes per community from 2.1. A call summary put the drop in finished, unsold homes at roughly 400, from about 3,500 to about 3,100 (MarketBeat), which appears to be in the neighborhood of Millrose's quarter-end purchases.

Nor is Lennar alone in struggling to move houses. PulteGroup said in its third-quarter call that it "elected to walk away from approximately 8 thousand lots" it had held under option, with incentives at 10.4% (Pulte call transcript). Nationally, the median new-home price in September was $400,400, down 3.9% from a year earlier, and the inventory of new homes for sale equaled 8.4 months of supply, a level that historically signals a buyer's market (Mortgage News Daily, citing Census). I wrote last month about how falling building costs had not translated into cheaper homes ("As Building Costs Fall, the Price of Owning a Home Keeps Rising"); this is the other half of that story, where the discounts finally show up and the builders scramble to protect their numbers.

Sources: Lennar quarterly 8-K earnings releases, compiled in Public-CRE-Data/mrp-liberty-homes (secondary); Q3 FY2026 checked against the 8-K. Selected quarters, not evenly spaced.

Did the Millrose purchases count toward Lennar's delivery total? Lennar has not said. I could not find any disclosure of the sales, their number or their revenue in Lennar's quarterly report filed October 2, and an independent review of that filing reached the same conclusion (Travis Valuation). If they did count, the math is uncomfortable: subtract the 356 last-week sales Hunterbrook identified from 20,840, and Lennar lands at 20,484 deliveries, 16 homes short of its own floor (calculated). That is a hypothesis, not a finding, and it rests on Hunterbrook's count and on an assumption about accounting that Lennar has not confirmed. But it is the question any analyst on the next earnings call should ask.

A Word for This?

There are several, and they range from the mundane to the notorious.

The financial term is a related-party transaction, a deal between two entities with common control or influence.

Millrose's own annual report concedes that its agreements with Lennar "were not negotiated at arm's length" and that it "might have received better terms from unaffiliated third parties" (Millrose 10-K). Related-party deals are legal and common; they are why public companies have independent audit committees, and Millrose says its committee must approve them.

When a related-party transaction transfers value from one set of shareholders to the person who controls both sides, economists call it self-dealing, or, in a more colorful term coined by Simon Johnson, Rafael La Porta, Florencio Lopez-de-Silanes and Andrei Shleifer in a 2000 paper in the American Economic Review, tunneling: "the transfer of resources out of a company to its controlling shareholder." The authors stressed that much tunneling is perfectly legal (NBER). The direction of any tunnel here would depend on price. If Millrose overpaid, value would flow from Millrose's public shareholders toward Lennar's; if Millrose got a bargain, the reverse.

The terms that make auditors nervous describe sales arranged to make a quarter look better than demand would. Channel stuffing means pushing inventory onto friendly buyers ahead of real demand; Bristol-Myers Squibb paid $150 million in 2004 to settle SEC charges that it had loaded its two largest wholesalers with $1.5 billion of drugs while covering their carrying costs (SEC). Bill-and-hold sales book revenue on goods that never leave the seller's orbit; the SEC once described Sunbeam's as "little more than projected orders disguised as sales" (SEC). Round-tripping is the most extreme form, in which money cycles back to its source as revenue; Homestore.com, the operator of realtor.com, was charged with recognizing "its own cash as revenue" in 2001 (SEC).

The closest analogue, and the one already on most investors' minds, is vendor financing and its modern cousin, circular financing. Before the 2001 telecom bust, Cisco and Nortel extended financing equal to more than 10% of revenue to customers who used it to buy their own equipment (Fortune). In September 2025, Nvidia announced a $100 billion investment in OpenAI, which would in turn buy Nvidia chips; by early 2026 that pledge had reportedly stalled, Jensen Huang said it was "never a commitment," and the figure under discussion shrank to about $30 billion (Tech-ish; ITdaily). Oracle, whose order backlog is more than half tied to OpenAI, suffered its worst week since 2001 in June (AI Weekly). When I argued in my October 20 Take that today's AI spending is not a replay of 2008 ("Inflation Is Still Too Damn High"), the circular deals were the piece I flagged as worth watching. The common thread is a seller that finances, or controls, its own buyer, so that reported demand is partly a mirror.

Which label fits Lennar and Millrose? The fairest answer today is "related-party transaction, with features worth scrutinizing." The features are the timing, the control, the absence of price terms in the governing documents, and the earlier disclosures. If you want to learn to spot these patterns in a 10-Q yourself, the best guide I know is Financial Shenanigans by Howard Schilit, Jeremy Perler and Yoni Engelhart (Financial Shenanigans on Amazon), which devotes whole chapters to sales that are not quite what they seem; I keep it on the shelf next to the closing binders.

What Millrose Said Before

The earlier disclosures are the reason this has reached a courtroom. On October 9, Millrose shareholders filed a proposed class action in federal court in the Southern District of Florida, accusing the company and two executives of misleading investors (complaint). The complaint quotes Millrose's March 2025 annual report, which said, "We do not, and will not, have any lease agreements with respect to the Transferred Assets," and its 2025 prospectus, which said the company did "not intend to enter into any lease agreements in the foreseeable future." The suit claims Millrose shares fell 8.5%, from $24.61 to $22.51, after Hunterbrook's report on October 2. These are allegations, not findings; the case is at its earliest stage, and the deadline for lead plaintiffs is December 8 (PR Newswire). Lennar's stock fell 6.7% on October 5, to its lowest close since October 2022 (Tiger Brokers).

Did Millrose Pay Too Much?

This is the crux, and the evidence conflicts. Hunterbrook matched Millrose's purchases to sales to ordinary buyers in the same communities and found nearly identical deed prices, about $290,661 for Millrose against $291,112 for individuals. Its argument is that individuals also received Lennar's roughly 12% incentives, such as closing-cost credits and mortgage-rate buydowns, so their effective price was closer to $256,000, which would mean Millrose paid about 13% more than a retail buyer (Hunterbrook). One vivid example came from Wesley Chapel, Florida: a house listed in April at $490,990, cut to $420,490, then sold to MRP Liberty for $409,300.

Source: Hunterbrook (Oct 2, 2026), Hunterbrook estimates; "net" subtracts Lennar's ~12% average incentive. Hunterbrook Capital disclosed short positions in LEN and MRP.

The independent deed dataset is less damning. It finds Millrose paid a median of 3.2% below last asking price and roughly the same as comparable sales in the same community (Public-CRE-Data). The two analyses differ mainly on one assumption: whether a bulk buyer should be judged against the price on the deed or against the price net of incentives that the bulk buyer may or may not have received. Neither company has disclosed incentive terms on the Millrose homes. Until they do, I would call the overpayment case plausible but unproven.

The return math is clearer. Millrose's management described an initial yield of about 5% on the homes, rising to an 8% unlevered return over time with 3% annual income growth (call summary). The debt on the rentals costs about 5.12%, by Millrose's own estimate (Millrose deck), and in September Millrose sold $1 billion of corporate bonds at 6.5% and 6.75% (Millrose release). Meanwhile its core land-banking business earns about 8.5%. In plain terms, a company that can earn 8.5% lending against land chose to put $120 million of equity into houses that yield about 5% before corporate overhead, financed at a little over 5%.

Sources: Millrose Q3 2026 release and deck (SEC); Hunterbrook (Oct 2, 2026); call summary for the ~5% cap rate; Middle Tennessee gross yield calculated from 10 homes in the Public-CRE-Data deed set.

In my book, The 5 Ways Real Estate Investors Make Money and Build Wealth, I break a rental's return into five engines: cash flow, appreciation, loan paydown, tax benefits and leverage. A 5% yield financed at 5.12% has a cash-flow engine that is barely turning over and a leverage engine running in reverse, which leaves appreciation and rent growth to do nearly all the work. That can still be a perfectly good investment for a patient owner. But it is a long-duration bet on exurban rents, and it is not the kind of deal a land banker would typically choose if it were shopping in an open market. If you want to see how quickly negative leverage eats a return, run a $270,000 house with $1,850 rent and a 5.12% loan through DealLab.

The Homes in Our Backyard

Millrose's deck shows 27 of its rental homes in Tennessee (Millrose deck). The deed dataset places ten of them in Middle Tennessee: one in Hermitage, one in Clarksville, three in Rutherford County, four in Sumner and one in Lebanon (Public-CRE-Data). They average about $424,000 in purchase price and about $2,303 in asking rent, for gross yields of 6.0% to 7.0% before taxes, insurance, maintenance and management (calculated). After a typical 40% expense load, that is roughly a 4% net yield, which is below what Millrose pays to borrow (calculated).

Source: Public-CRE-Data/mrp-liberty-homes (county deeds and Evergreen Residential rental listings; secondary). Gross yield and price versus list calculated.

The Westmoreland homes carry a twist worth noting. According to the dataset's title notes, the land under them was owned at the start of the year by "Millrose Properties Tennessee LLC c/o Lennar," then conveyed to Lennar in 2026, built on, and sold back to a Millrose rental subsidiary (Public-CRE-Data). That is how land banking is designed to work for the first two steps; it is the third step that is new. I have not seen the Sumner County deeds myself, and the dataset warns that some list-price matches may be imperfect. Davidson County's register of deeds charges $50 a month for online searches (Metro Nashville), but anyone can search the state's free property assessment site for owner "MRP Liberty" (Tennessee Comptroller), and I plan to check it again after Lennar's fiscal year ends on November 30.

Why It Matters for Investors

First, comparable sales. When a related buyer pays full deed price for a new home, that price enters the county record and the MLS-adjacent data that appraisers rely on, while the incentives a retail buyer received often do not. In a subdivision where Millrose bought a large share of recent sales (Hunterbrook found it took more than one in four of Lennar's sales in Princeton, Texas, according to its report), the recorded comps may overstate what a typical buyer actually paid. If you are buying or refinancing in a Lennar community, ask your appraiser how he handled concessions.

Second, competition for tenants. Hunterbrook's reporters visited 36 Millrose homes in Marion County, Florida, and found none occupied (Hunterbrook). Hundreds of brand-new rentals hitting the same exurban subdivisions at once can soften rents for every small landlord nearby, at least until they lease up. If you own single-family rentals in Gallatin, Westmoreland or Smyrna, price your next renewal with that in mind.

Third, the policy angle. In my October 26 issue I argued that Washington's new limit on large institutional buyers of single-family homes would do little ("As Washington Chases Big Landlords, a Quieter Apartment Shortage Takes Shape"). Treasury's proposed rule bars investors with 350 or more homes from buying more after January 7, 2027, but exempts homes bought within 18 months of their certificate of occupancy and offered for rent (Treasury; HousingWire). Millrose already owns more than 350 homes, and every one of them was bought new from a builder. My hypothesis is that the rule may channel large-scale capital toward builders' unsold inventory rather than away from single-family housing, and that the builder-to-affiliate model could become a template.

The contrast with Invitation Homes, the largest single-family landlord, is instructive. In the third quarter Invitation sold 658 homes at an average of $483,782 and raised its 2026 sales target to $900 million to $1 billion, while buying back its own stock (INVH supplemental). By my rough calculation, its share price values its homes at about $285,000 apiece, net of debt, so the company appears to believe its houses are worth more sold one at a time than held in a portfolio (calculated from INVH 10-Q and the Oct 26 close). The most experienced landlord in the business is selling single-family homes into the retail market. Millrose is buying them at retail prices. One of them is likely to be wrong.

The Steelman: The Case That Lennar and Millrose Did Nothing Wrong

Again, I could be wrong; how could I be misreading this?

1) Builder-to-rental sales are routine. Lennar has sold homes to single-family rental operators for years; one analyst estimates such sales have historically been a mid-single-digit share of its deliveries (Tiger Brokers, citing KBW). Taylor Morrison built an entire build-to-rent brand, Yardly, and in 2025 signed a $3 billion financing facility for it with Kennedy Lewis, the same firm that manages Millrose (Taylor Morrison).

2) It is small. At $333 million, the pilot is a sliver of a company with $5.85 billion of equity, and Millrose excluded the rentals from its manager's fee base, which removes one obvious conflict (Millrose deck). Miller is not a Millrose director, its five directors are independent, and related-party deals require audit committee approval.

3) The prices look like market prices. The deed data show Millrose paying about what other buyers paid, and slightly below asking. The 13% overpayment figure depends on an assumption about incentives that its author, a short seller, has an interest in.

4) The rents are beating plan, and serious money is betting on Lennar. Management says early rents are 10% above underwriting. Berkshire Hathaway, which I wrote about in August when it began buying builders ("Warren Buffett Just Bet Big on Real Estate"), added roughly $540 million of Lennar stock (calculated) in late September and early October, after the Millrose purchases were underway, and now owns about 12% of the company (Parameter; IBTimes).

Where I land: I'm not accusing anyone of any wrongdoing. But this type of financial engineering boosts risk, so I’m calling it out. Purchases clustered in the final days of Lennar's quarter, why the governing documents left price out, and how the deals squared with earlier statements that Millrose would not lease its assets… this is out of the ordinary for a healthy company. We shall see. If Millrose's buying resumes in the last week of November, around Lennar's fiscal year-end, and goes quiet again afterward, the pattern looks less like a rental strategy and more like quarter-end inventory management. If the buying is steady through the year, or the homes lease up quickly at the promised rents, the trend looks less concerning.

Florida Votes Tomorrow on a Tax Cut With a Bill Attached

Tomorrow, Florida voters decide whether to raise the homestead property tax exemption for locals from $51,411 today to $150,000 in 2027 and $250,000 in 2028 (indexed to inflation after that), and to cut the annual cap on assessment increases, which includes rentals and commercial buildings, from 10% to 5% (Tax Foundation).

It needs 60% to pass and is polling between 50% and 55% (News4Jax, RISMedia).

The fine print contains a feature I think deserves more attention than it has received: a toll for newcomers.

Only owners who are permanent Florida residents by December 31, 2026, get the full exemption. Anyone who establishes a homestead on or after January 1, 2027, receives $50,000 of non-school exemption for five years before graduating to the full amount, and local governments can shorten that wait only by a two-thirds vote, starting in 2030 (Kaufman Rossin).

In practice, two neighbors in identical houses could pay very different tax bills based on the year they moved in. Using 2025 non-school millage rates, I estimate the gap at about $1,800 to $2,700 a year for a home assessed at $300,000 or more, or roughly $9,000 to $13,000 over the five-year wait (calculated).

Calculated from Amendment 3's terms (Tax Foundation; Kaufman Rossin) and 2025 adopted non-school millage (Miami-Dade Property Appraiser; Hillsborough and Duval county summaries, secondary). Assumes an assessed value of at least $300,000; ignores inflation indexing and millage changes.

The second question is, who fills the hole?

The state estimates the measure would cost local governments about $4.9 billion in its first year and about $11.8 billion a year by 2031 (Tax Foundation). S&P Global surveyed 90 rated Florida cities and counties and found they expect to lose an average of 14% of revenue after the second year, up to 48% in some cases, and most said they would turn to fees, user charges and special assessments rather than spending cuts (Boca Raton Tribune). An analysis by tax economist Jared Walczak for the Florida Housing Coalition estimated that if localities raise rates to replace lost revenue, the property-tax bill would rise by about $406 per apartment and $1,081 per single-family rental in 2028, and by $544 and $1,471 by 2031 (Tampa Bay 28). Landlords pass most such increases through in rents over time, so renters, who cannot vote themselves a homestead exemption, may end up paying for part of their neighbors' tax cut.

Source: Jared Walczak analysis for the Florida Housing Coalition, via Tampa Bay 28 (Sept 23, 2026). Assumes local governments raise millage to replace lost revenue.

Florida is not alone. Ballotpedia counts 13 property-tax measures in seven states this year, including Wyoming's proposal to exempt half the assessed value of a primary residence and Oklahoma's plan to lower assessment caps (Ballotpedia; Kiplinger). An Ohio drive to abolish property taxes fell about 108,000 signatures short (calculated) of the ballot and is now aiming for 2027 (Ohio Capital Journal). The frustration is understandable: the average single-family property-tax bill reached $4,427 in 2025, up 3% (ATTOM).

Tennessee's version, Amendment 2, is mostly symbolic. It would write into the state constitution a ban on a state property tax, which Tennessee has not levied since 1949; county and city property taxes are untouched (Ballotpedia; WSMV). It passed the legislature by wide margins, and no organized opposition has formed. If it passes, it may matter mostly as a signal to the out-of-state investors and relocating families who compare Tennessee and Florida on carrying costs, which brings me to the point for investors. A Florida newcomer penalty, if it passes, would widen the cost gap in Tennessee's favor at the margin, and a 5% cap on rental assessments could make Florida apartments look cheaper to own than they will be once a sale resets the assessment to market value, which existing law already requires.

As you head to the polls tomorrow, be careful of those trying to pull one over on you. Tune out the polemics and read the fine print, the way Florida's newcomer toll hides in the details of a tax cut. The game of special interests is to litigate, then legislate. Each rule written to protect whoever is already inside adds a little more dirigisme, and markets ossify under it. Let's rebuff those efforts, and force them to innovate.

Nashville: Oracle Is Hiring Before It Has a Building

Oracle's Nashville headquarters is moving from renderings to dirt. Axios reported last week that the company will break ground on its Nashville headquarters this month (Axios), and it is hosting a public open house at 2300 Bowline Ave. on November 12 and 13 from 3:30 to 7:30 p.m., followed by a community day on November 14 (NewsChannel 5). The campus is planned at about 1.3 million square feet across roughly 23 buildings, with completion targeted for the end of 2031 (Construction Review Online).

What gets lost in the coverage is that Oracle did not wait for the campus to start hiring. The company reported 91 Nashville jobs to the state at its 2021 baseline, and by the end of 2025 it had added several hundred more: 782 across the annual figures the Nashville Scene lists, or 827 by the Scene's own total (Nashville Scene). That puts Oracle's local headcount at roughly 870 to 920, close to ten times where it started (calculated), all of it working out of leased space while the East Bank site was still being permitted and graded. Hiring has also picked up after a slow 2023, rising from 71 new jobs that year to 168 in 2024 and 188 in 2025.

Sources: Nashville Scene (Aug 14, 2026), from Oracle's annual job reports to TNECD; cumulative totals calculated. The Scene reports 827 new jobs; its listed annual figures sum to 782.

The scale of what comes next is the bigger story. Oracle has said it intends to employ about 8,500 people in Nashville by 2030 or 2031. The state's incentive agreement, worth as much as $240 million through a $65 million grant and up to $175 million in property-tax reimbursements for East Bank infrastructure, is built on 5,989 jobs, with 90% due by the end of 2029 (Nashville Scene). A company spending billions of its own money on 1.3 million square feet earns a return on it only one way, by filling the buildings. For investors, that means the East Bank, Germantown, Neuhoff and inner East Nashville have a large, well-paid employer arriving in phases: construction crews first, then thousands of office workers as the buildings open. I would underwrite Oracle demand as a story that starts now and accelerates as the first buildings come online.

Nashville's unemployment rate is 2.8%, well below the nation's.

Oracle is not carrying Nashville alone, jobs are plentiful here. The metro's unemployment rate was 2.8% in September, below Tennessee's 3.4% and the nation's 4.2%, and the area added jobs at a 1.0% annual pace, to 1,204,600 (BLS; TN Dept. of Labor; FRED). Davidson County's population grew 1.3% last year, to 745,904, again the fastest in the state (UT State Data Center). Corporate commitments keep coming, from Starbucks' $100 million office expansion with up to 2,000 jobs (Axios) to Sennheiser moving its Americas headquarters here (Sennheiser).

One number in that same population report deserves attention. Net international migration into Davidson County fell to 5,887 in 2025 from 14,102 a year earlier, a 58% drop (UT State Data Center). Immigrants filled a meaningful share of the city's new apartments during the building boom, and that source of demand has more than halved. Strong local hiring may offset it, but if you own Class B and C apartments in Antioch, Madison or along Nolensville Pike, watch your renewal rates closely this winter.

Sources: BLS (Nashville MSA, not seasonally adjusted, preliminary); Tennessee Dept. of Labor (state, seasonally adjusted); BLS via FRED (U.S., seasonally adjusted); UT Tennessee State Data Center (Mar 30, 2026).

And now my final thoughts.

My Skeptical Take:

To put it more succinctly, every deal has two sides.

Most of the time we only hear from one of them.

But some of these self-dealing shenanigans rhyme with what companies were doing in the dot-com era.

None of this makes me less bullish on real estate. But it does remind me to be more vigilant when I’m running numbers on a potential investment.

The market is handing out discounts right now, things like: builder incentives, sellers caught in a loan coming due and facing a refinance out of 4% loans into 7.3% loans, and through a capital market that has slammed shut on apartment owners. Capital is expensive and more scarce today for real estate investors. But if you can afford it and are creditworthy, this is the exact setup I like to pick up a deal. (I know I know I’ve been a broken record on this for months, but it’s still true). Discounts are where long-term, generational returns come from.

So here's how I'm operating:

• If you own rentals near new-build subdivisions: expect more competition from institutional rentals over the next six months, and price renewals to keep good tenants rather than to chase the last $50.

• If you're buying new construction (I wouldn’t, you pay a premium for someone else’s work.. but if you must…): ask the builder in writing what incentives were offered to other buyers in the community, and ask your appraiser how he adjusted for concessions. The recorded price may not be the real price.

• If you're buying resale: the builders' troubles are your leverage. A seller competing with a brand-new house at a 12% discount down the street has to sharpen his pencil. Underwrite at today's 7.3% rate and flat rents, and make the builder's inventory problem your negotiating advantage.

This is the time to pick up a deal. Hell, I’m getting all bulled up just thinking about it!

A little over a century ago, a Boston lawyer named Louis Brandeis wrote a series of articles for Harper's Weekly about bankers who sat on both sides of the deals they financed. He collected them into a book with a title that could headline this issue, Other People's Money, and he offered a remedy:

To be clear, this is in no way a 2000 systemic issue. Not in the slightest.

Here's why this is not 2000:

• It's small. At the height of the telecom boom, Cisco and Nortel extended financing equal to more than 10% of revenue to their own customers (Fortune). Millrose's 1,233 homes equal about 6% of the 20,840 homes Lennar delivered in a single quarter, and the $333 million pilot is about 6% of Millrose's $5.85 billion of equity (calculated from the Lennar 8-K and Millrose deck).

• The debt is fenced off. The rentals carry $213 million of limited-recourse debt, so a bad outcome stays inside the pilot instead of spreading through the company (Millrose release).

• The collateral is real. Much of the money in 2000 went into fiber and equipment that sat unused for years. These are finished houses that people can live in, and the first ones are leasing at rents above plan.

• The economy is strong. GDP grew at a 3.4% pace last quarter (BEA), and jobless claims are near 57-year lows.

But it reminds me as an investor to be vigilant when underwriting each investment.

Sunlight highlights, but it takes a little perspicacity to look for it: in the exhibits, the deed records, the chain of title, the footnotes nobody reads.

In 2000, the warning signs were sitting in the footnotes.

Pay attention to the minutia.

Then hold your nose and just keep buying.

Until next time. Stay Curious. Stay Skeptical.

Herzliche Grüße,

-The Skeptical Investor

I'm a Nashville-based investor and broker who writes about real estate and markets each week. Happy to answer questions in the comments.