Summary

Bill Chen’s central post-NAREIT takeaway is that blue-chip public REITs are entering the next cycle with balance sheets private owners cannot match. The management teams he met reported no distress, interest coverage generally ran 3-8x—7.2x at Mid-America and 6.8x at Camden—and seven-year unsecured debt was available around 4.9% to the high-5% range. Private-market new-debt underwriting, by contrast, often starts at 1.25x debt-service coverage.
A 70-80% collapse in market-rate multifamily starts is setting up a supply-light window through at least 2028-29. Affordable and mission-driven projects may still proceed, while AI data centers are a separate category; Bill thinks rent growth could run at least 3% in 2026-28 if market-rate supply remains constrained. Public REITs can fund development from diversified, roughly 95%-occupied portfolios while private owners have their “hands tied behind their backs.”
Bill still underwrites roughly 18% three-year and 15% four-year IRRs for Mid-America and Camden without heroic assumptions. His model uses 3-4% rent growth and a 5% exit cap rate, while Andrew’s pushback is that a new property trading at a 5% cap does not prove an older portfolio deserves the same mark. Bill’s rebuttal: scale, liquidity, diversification and acquisition optionality arguably warrant a “liquidity premium,” not today’s public-market discount.
New York residential fundamentals are unusually strong precisely because regulation and physical constraints suppress new construction. Bill separated that operating strength from political risk: proposals for indefinite rent freezes or price controls could damage affordable-housing providers, as Washington, D.C.’s eviction restrictions illustrated, yet they might also increase the scarcity value of existing high-end, largely unregulated portfolios. “You can’t really build anything here.”
Grocery-anchored shopping centers are Bill’s next major theme after more than 15 years of underbuilding. Annual supply growth fell from roughly 3% in 1999-2008 to below 0.5% recently, while Sun Belt portfolios can combine 3% contractual escalators, 10-20% renewal lifts and roughly 20% new-lease increases into 4-4.5% NOI growth. Competing projects may require rents to double, making existing centers bought at 7-9% cap rates especially compelling.
Management quality remains the largest caveat to cheap REIT valuations. Andrew highlighted controlled structures and missed buybacks; Bill contrasted Camden’s modest $40 million repurchase with Eurofins buying roughly 5% of its shares in five months, and called Hudson Pacific’s buyback-near-$20/issuance-in-the-mid-$2s sequence “doing everything wrong.” Face-to-face meetings remain crucial because defensiveness, frustration and capital-allocation intent do not appear cleanly in filings.
Bill now sees enough dislocation to construct an entire roster rather than rely on one or two ideas. Mid-America, Camden and FRPH are the “offensive linemen”; grocery centers are running backs; Bill later identified Clipper and COPT as 2%-3% wide-receiver allocations, while also discussing a small Seaport position; and Dream Residential is special teams, with a potential sale against a stated $13.30 NAV. “In today’s environment, we could build a whole team.”

Deep dive

1. Public REIT balance sheets bear little resemblance to private-market distress

Across roughly a dozen NAREIT management meetings, Bill heard no company describe debt distress or difficulty covering interest. Coverage generally ranged from 3x to 8x, including 7.2x at Mid-America and 6.8x at Camden—levels that leave little plausible path to trouble absent a large, self-inflicted acquisition or development mistake.
The private-market contrast is stark: new debt underwriting may begin around 1.25x debt-service coverage, with anything above that deemed healthy. Bill has also watched private multifamily LP distributions disappear when floating-rate debt reset, while the major public apartment REITs continued funding dividends, acquisitions and development.
Public balance sheets also retain financing access. Blue-chip REITs were issuing seven-year, fixed-rate unsecured bonds from roughly 4.9% through the mid- or high-5% range, with lenders underwriting diversified, approximately 95%-occupied portfolios rather than treating the proceeds as financing for one speculative project.
Bill’s framing: these are “clearly the more structurally advantageous players”—often around 25% loan-to-value, roughly 4x net debt to EBITDA and 7x interest coverage—yet investors can still buy them below private-market asset values.

2. The construction cliff is creating a multiyear rent-growth runway

Bill’s late-2023 thesis was that already planned and budgeted projects would finish, but high rates and inadequate development returns would then cause starts to “fall off an absolute cliff.” Eighteen to 24 months later, Bill sees market-rate multifamily starts down 70-80%, with declines of at least 50% across self-storage, warehouses and most other property categories.
Affordable or mission-driven projects may still be greenlit, but conventional market-rate supply has largely stopped receiving the green light. AI data centers remain a separate category because they are “a whole different animal.”
Bill’s conditional forecast is at least roughly 3% annual rent growth in 2026, 2027 and 2028, with meaningful new supply unlikely before the second half of 2028 or 2029.
The longer rates stay high, the longer that window may remain open. Public REITs can therefore develop into scarcity while leveraged private owners remain sidelined—the precise asymmetry Bill expected when the construction pipeline was still obscuring the eventual shortage.

3. Cheap unsecured capital lets the public REITs restart development first

Mid-America and Camden can borrow against hundreds of stabilized buildings, then direct a small portion of total capitalization toward ground-up projects. Bondholders are not lending 100% against one risky development; they are lending to a large, occupied portfolio that happens to allocate perhaps 5-6% toward construction.
That structure creates a cost-of-capital arbitrage unavailable to most private developers. Bill even asked why Mid-America should stop near a $1.2 billion development pipeline instead of pushing toward $2 billion, though he acknowledged limits for a company with roughly $22-23 billion of enterprise value.
The same advantage applies to acquisitions. Management teams would buy newly built multifamily assets around a 5% cap rate, hold them for 25 years and compound from the coming rent-growth cycle; forced sellers accepting that price often lack the option to wait for a better exit.
At roughly 4x net debt to EBITDA, Mid-America could potentially add another turn of leverage—about $1.5 billion by Bill’s estimate—for acquisitions or development. The opportunity is therefore not merely defensive survival: excess balance-sheet capacity can become offensive capital allocation.

4. The 15-18% apartment-REIT return case survives Andrew’s cap-rate challenge

Bill’s base underwriting produces roughly an 18% IRR over three years or 15% over four, assuming 3-4% rent growth and a 5% exit cap. Andrew emphasized how unusual that is for liquid, dividend-paying large caps with relatively narrow operating outcomes: “That’s the stuff investors make a career on.”
Andrew’s pushback—worth keeping—is that a REIT paying a 5% cap for a brand-new building does not establish the same value for its older portfolio. Bill conceded the quality difference but argued that forced development sales are poor evidence of equilibrium pricing, especially when most private buyers are absent.
Bill’s broader claim is that large, liquid, diversified operators with superior funding and acquisition optionality should eventually trade above private value. Public REITs historically did receive that “liquidity premium”; today, investors are “not paying anything at all” for the optionality.
A postmortem on the late-2023 thesis reinforced his confidence. Buying in the stated roughly $115-$130 range generated annualized outcomes beginning just under 15% and extending above 20%; selling during the more recent $155-$170 range could have produced roughly 17-38%, depending on entry and exit.

5. New York’s political risk is also its supply barrier

At NAREIT, Bill heard broad enthusiasm for New York-area residential fundamentals from operators such as Equity Residential and AvalonBay. The attraction is simple: demand is healthy, rent growth is strong and “you can’t really build anything here.”
Regulation deepens that barrier by making development economics unusually uncertain. High-end portfolios have limited direct rent-regulated exposure—Clipper is a notable exception—so stricter rules might perversely increase incumbent asset values by preventing competing supply.
Bill nevertheless reacted sharply to proposals for indefinite rent freezes and government-backed supermarkets, drawing on his experience growing up in a communist country: “If we’re going back to another price-control environment,” the long-run consequences matter even if existing owners initially benefit.
His concrete warning came from Washington, D.C., where rules that effectively prevented eviction encouraged some tenants not to pay. Affordable and mission-driven housing providers then faced financial distress or bankruptcy—the opposite of the policy’s intent and, in Bill’s telling, a wake-up call for politicians.

6. Multifamily operations barely cracked under the supply wave

At Mid-America and Camden, the maximum NOI decline was less than 2%, while occupancy remained around the mid-95% range—roughly 95.5%. Bill credits disciplined pricing: management conceded some rent to protect occupancy rather than allowing the delivery wave to create destabilizing vacancy.
Affordability also remains healthier than coastal anecdotes imply. Rent consumes roughly 21-23% of resident income across these portfolios, a ratio Andrew contrasted with the far heavier burden faced by New York renters.
Mid-America increased its annual dividend from about $5.60 to $6.06 per share during a period when many private multifamily LPs lost distributions altogether. Cash generation continued funding dividends, purchases and developments rather than servicing a floating-rate rescue.
Bill has made multifamily roughly 50% of one hard-asset portfolio because he sees narrow downside ranges and durable demand. Unlike an operating company whose value can unexpectedly migrate to one speculative project, the underlying exposure remains hundreds of apartments across a dozen Sun Belt cities.

7. Technology strengthens scale, but AI is not yet moving resident demand

Andrew asked whether AI employment is redirecting migration toward a few technology hubs or away from them. Bill’s honest non-answer was that management teams were not reporting an identifiable AI-driven population shift, and the absolute number of AI-development jobs may be too small to move Sun Belt apartment demand.
The immediate AI and software benefits sit inside operations: automated tenant screening, paperless leasing, self-guided tours and fewer on-site leasing employees. Leak-detection systems offer another mundane but valuable example of technology that a 100,000-unit operator can spread across its portfolio.
A 300-unit private owner cannot rationally spend millions developing the same systems. Public REIT NOI margins are materially higher than those in many private deals Bill sees, although he cautioned that scale cannot be isolated cleanly from their larger, more institutional-quality assets.
Andrew pressed for a per-unit cost figure, but management teams had not quantified one. Bill added it to his next NAREIT question list rather than pretending precision: the advantage is visible, but its exact dollar contribution remains unproven.

8. Governance separates genuine compounding from merely surviving a rerating

Andrew’s recurring concern is that REITs are effectively controlled companies: activism is difficult, insider ownership can be limited and management may prize asset growth because it supports prestige and compensation. When shares trade far below NAV, failing to repurchase them can be a major per-share opportunity cost.
Bill found only modest evidence of aggressive buybacks. Camden repurchased roughly $40 million, negligible relative to its size; by comparison, Eurofins bought about 5% of its shares in five months after complaining that the stock traded near half of private value. “They just went out and bought back 5%.”
Retail Opportunity Investments Corp. supplied the better capital-allocation outcome. At the prior NAREIT, Bill read the CEO’s visible frustration with repetitive Kroger-Albertsons questions as a sign the company might sell; Blackstone called on June 19, shortly after the conference, and a full process followed.
ROIC had been a roughly 15% position and contributed about 4% to fund performance. Bill wished the sale price were higher, but the episode reinforced why he calls NAREIT “the Super Bowl for what we do”: facial expressions, defensiveness and frustration can reveal what transcripts cannot.

9. AI accelerates preparation, while human meetings preserve the edge

Bill’s firm has built internal AI tools that systematically pull developments from earnings calls and prepare outlines before meetings. Bill expects such capabilities to become commonplace within 12-24 months, so the current informational advantage will decay.
His advice to younger investors is therefore to combine those tools with repeated management contact. Comparing a team’s answers and body language over two or three years can reveal whether a thesis is progressing—or whether a CEO becomes unexpectedly defensive when asked a simple question.
Andrew agreed from his own company calls: a press release may sound ambiguous until management either explains an alarming expansion plan or plainly reiterates that capital will be returned, perhaps because an 85-year-old chairman has estate-planning needs.
The shared conclusion was not that meetings produce forbidden information, but that they clarify intention and temperament. That remains one way to “AI proof” investment work after everyone can summarize the same filings.

10. Grocery-anchored centers combine scarce land with embedded rent resets

The market still carries an “over-retailed America” narrative rooted in the 1999-2008 period, when shopping-center supply grew around 3% annually. After the financial crisis that fell near 0.75%, and during the last four or five years it dropped below 0.5%.
Bill distinguishes grocery-anchored strips from troubled B/C malls and power centers. A supermarket surrounded by Chinese takeout, pizza, Pilates, urgent care, physical therapy or radiology offers enduring convenience: “You pull up, you park, you go in.”
In infill Sun Belt locations, a competing center may require 10-15 acres at a busy intersection. Operators told Bill that rents would need to be roughly double current tenant rents before a competing center worked, while population has continued growing around the existing stock.
Lease economics carry their own delayed mark-to-market. Portfolios often have 3% annual escalators, 10-20% renewal increases and roughly 20% increases on new leases; together with mid-90% occupancy, that supports 4-4.5% annual NOI growth on assets sometimes purchasable at 7-9% cap rates.

11. Blackstone’s ROIC purchase validates the theme, but the best names remain undisclosed

Blackstone bought ROIC at roughly a 6.1% cap rate, gaining a 97%-occupied West Coast portfolio that private-market participants described as almost impossible to assemble. If NOI grows around 4% and financing costs sit around 5-6%, Bill thinks “that math just works really well.”
Bill said there had been only three deals in the grocery-shopping-center space over the preceding two or three years, including ROIC’s sale, a Kimco stock-for-stock transaction and Urstadt Biddle Properties. He would welcome consolidation where duplicate G&A disappears and two smaller companies become large and liquid enough for bigger REIT funds to own.
He floated a hypothetical 40-50% stock premium only where the acquirer’s shares were also cheap, allowing investors to roll into a better-scaled platform. Andrew laughed at the casual wish list, but the underlying logic was scale, liquidity and eliminated overhead rather than cashing out indiscriminately.
Bill’s firm was still building positions and withheld the names: one grocery-center idea underwrote to a mid-30% three-year IRR, with either a sale or removal of an overhang as the catalyst; others modeled in the mid-20% range. Bill wanted to surface the asset-class mechanism without front-running unfinished work.

12. The roster pairs stable compounders with smaller, event-driven torque

Bill’s football construction puts Mid-America, Camden and FRPH on the offensive line: large, durable holdings designed to anchor the portfolio. Grocery-center ideas are running backs capable of mid-20% to mid-30% outcomes; he later described Clipper and COPT as 2-3% “wide receiver” positions with more volatility and torque, while discussing Seaport as a separate small position that might be viewed as a running back.
Clipper has been frustrating because of New York rent stabilization, but management maintained the dividend and avoided meaningful dilution. If rates ease and investors re-engage, Bill thinks the retained per-share upside could support an $8-$10 stock—quite different from rescuing a company by multiplying its share count.
Seaport’s assets may be valuable and its cash was roughly $18 per share, yet Andrew learned that neither figure creates a hard floor for a small, controlled, cash-burning company. Progress on the Meow Wolf lease and 250 Water Street matters, but the new CEO must drive enough traffic to create restaurant and event operating leverage.
The separation of the MPC assets from Seaport was strategically sound in Bill’s view: Ward Village, The Woodlands and the Summerlin assets fit one operating box, while Seaport requires a focused turnaround. More concerts, events and foot traffic are the mechanism; asset value alone is insufficient.

13. Office recovery is class-specific, and Hudson Pacific shows the cost of bad timing

Bill believes New York has moved past the “office is dead” narrative for well-located Class A buildings, not for B/C properties. Alexander’s became a 12% position because its Bloomberg global headquarters has a lease through 2040 and a tenant whose credit and headquarters commitment make the asset unusually defensible.
Bill also bought Vornado preferred shares near $0.45 on the dollar at roughly an 11.5% yield after another investor correctly called the common-stock bottom. Outside trophy assets, B/C offices can still transact if long leases and strong tenants let buyers DCF the remaining cash flows and assign a conservative residual.
Hudson Pacific raised about $600 million of equity while its stock traded in the mid-$2s, eliminating much near-term bankruptcy risk but increasing the share count by more than 150%. Andrew’s illustrative NAV fell from roughly $10 to $4.50 per share after dilution: the call option lived longer, but each share owned far less upside.
Bill’s governance verdict was severe but hedged: buying shares near $20, investing in studios and issuing heavily in the mid-$2s looked “likely to just preserve your job.” AI-generated video adds another uncertainty to the studio thesis, even if San Francisco office fundamentals eventually recover.

14. Dream Residential offers a bounded catalyst that broad REIT ETFs miss

Dream Residential, a small Canadian REIT holding 15 Class B apartment properties across three U.S. markets, announced a strategic-alternatives review. Bill’s firm called brokers—including one who had sold the properties to the REIT—and concluded there should be local buyers across the age and quality spectrum.
The stated NAV was about $13.30; Bill initially found units near $7.70-$7.80 and built an average around $8.10. At roughly $9.15, he allowed for liquidation friction and modeled proceeds nearer $11.50-$12, plus an annualized monthly dividend yield around 5%.
From Bill’s firm’s entry, that implied approximately 40-60% one-year total return if a sale completed. The underlying 7-8% cap rate offered support, but Bill stressed that allocation still depends on “forward IRRs” and deal probability; he had already rotated capital from a lower-return holding.
Broad ETFs cannot express that judgment. VNQ held only about 8.8% in multifamily while market-cap weighting concentrated exposure in towers, Prologis, Equinix and Simon; Bill’s firm held roughly 50% residential. Bill and Andrew’s closing puzzle: liquid public assets should historically earn a premium, yet investors can currently choose the desired property exposure at a discount.

Embryology of AI: How Training Data Shapes AI Development w/ Timaeus' Jesse Hoogland & Daniel Murfet Vol.65 AI’s New Era: Is Google Back?