JTBD RE Pulse Week 6/18/26
The tide that lifted every deal for forty years is going out.
Now everything reprices from hope to reality, at once.
For three issues we have followed a war, an inflation that won't quit, and a world that trusts the dollar a little less each month. Step back and they are one story. Four tailwinds that blew for forty years — cheap money, globalization, an unquestioned dollar, and the reflex to bail out every mistake — are expiring at the same time, while a brand-new force, the electricity that artificial intelligence consumes, switches inflation on from a direction it has never come from before. Your Massachusetts sellers are simply the first to reconcile a 2021 price with a 2026 reality. They will not be the last.
- This is a regime, not a dip. Four forty-year tailwinds are expiring at once, so there is no cycle to wait out and nothing upstream coming to rescue a thin deal.
- Technology switched sides. For the first time in forty years the leading technology raises prices instead of lowering them, because AI runs on scarce electricity. It lands in everyone's electric bill, and hardest on the landlord who pays the building's.
- The Fed has no room and no rescue. A hot energy headline keeps it restrictive even as the core cools, and cash now loses to inflation. Plan on rates in the 6s.
- Massachusetts is reconciling first. Homes priced to today sell in five to six weeks near asking; homes priced to 2021 sit three months and get pulled.
Four tailwinds, expiring together
The tide was never one thing. It was four separate tailwinds, each with a birth date, that blew the same direction for forty years and got mistaken for normal. They are going out at the same time, which is why everything seems to reprice at once. A fifth force, new to the era, is blowing the other way.
| The tailwind | Born | Going out as |
|---|---|---|
| An unquestioned dollar | 1971 gold window; 1974 petrodollar | Central banks trade dollars for gold; oil increasingly settles outside the dollar |
| A forty-year fall in the cost of money | 1980, Volcker's peak | Rates cannot fall from a floor; the 30-year fixed sits at 6.52% |
| Globalization as a disinflation machine | 1989; China's 2001 WTO entry | Tariffs, reshoring, and "self-reliance" switch the machine off |
| The reflex to bail out every mistake | 2008 QE; 2020 stimulus | Inflation, absent since the 1980s, now blocks the bailout |
| NEW — technology as an inflation source | 2023 onward | AI's hunger for electricity makes the leading technology inflationary for the first time |
The tide that is going out
The world is walking back toward gold
The dollar has been backed by trust rather than gold since 1971, and by the world's need to buy oil in dollars since 1974. That arrangement is quietly running in reverse. Gold sits near $4,250 an ounce, up about a quarter over the year, and a record 43% of central banks say they intend to add more this year while trimming dollar reserves. They are buying it for a bundle of reasons that all point the same way. A record 90% of central banks cite gold's performance in a crisis and 84% its role as a long-term store of value, while the 2022 freezing of Russia's reserves sharpened the case for diversifying out of the dollar by showing that dollar reserves can be frozen and gold cannot. The gold rush is not a fad. It is the 1974 machine that recycled the world's dollars into US debt, running backwards.
Two details make this structural, not a momentum trade. The old anchor has snapped. For two decades, higher real yields reliably pulled gold down; that link broke in 2024–25, with gold clearing $3,000 even as real yields held above 1.5%, because price-insensitive central-bank buying replaced rate-sensitive ETF flows as the marginal buyer. The dollar is confirming the move: the dollar index had its worst first half in over fifty years in 2025, and gold and the dollar still trade inversely at roughly −0.5 to −0.8. The honest wobble is that reported central-bank net buying fell sharply in early 2026 — yet OTC and Swiss-refinery flows imply the real buying continued, the tell that this is reserve managers on multi-year mandates, not traders chasing a price.
The forty-year fall in the cost of money has hit its floor
From Volcker's 1980 peak, interest rates fell one direction for four decades, and almost everything people call appreciation was that falling discount rate inflating assets underneath them. That engine is mathematically spent — and the proof is that the Fed has been cutting while the long end refuses to follow. Since the Fed began easing in September 2024 (100 basis points and counting), the 10-year Treasury has risen about 100 basis points, almost entirely through higher real yields and a rebuilding term premium. The cause ties straight back to 1.1: the same central banks trading dollars for gold have trimmed their dollar reserves by roughly $113 billion, and research suggests every one-percentage-point fall in foreign Treasury holdings relative to GDP lifts yields by more than 30 basis points. Apollo's chief economist Torsten Sløk flags the rising term premium as a core fiscal-sustainability risk, noting the Treasury is leaning on short-term bills precisely to avoid pushing long rates higher still. Then add a second floor your mortgage sits on: the spread between the 30-year fixed and the 10-year Treasury is wide by historical standards — roughly two points here, toward the top of its 1.5–2.5 range — after years of quantitative tightening pulled the Fed out of the mortgage market. The 30-year fixed sits at 6.52%. You cannot fall from zero again — and this time, the Fed cutting does not even guarantee your mortgage follows.
The war is cooling — and that is the cyclical counter-note
Not every current points the same way, and an honest read names the off-ramp. A US–Iran interim agreement reached in June has the Strait of Hormuz reopening and crude back in the mid-$70s, down roughly 38% from its April peak. That is the cyclical good news sitting on top of the structural story: if cheaper energy holds, the inflation scare of the spring fades faster than the Fed expects. It is the single most plausible way the gloomier version of this thesis is wrong for a while. The war that started the inflation is winding down, but the four structural tailwinds it sat on top of are not coming back with the peace.
A hot headline, a cooling core, and a new kind of inflation
The 4.2% scare is an energy mirage on a cooling core
May headline inflation hit 4.2%, the hottest reading since 2023, and the wire ran with it. It is the least informative number in the release. Energy alone drove more than 60% of the monthly increase, while core inflation actually halved from April, shelter reverted once a one-off adjustment dropped out, and core goods went slightly negative. The danger is not that inflation is broadening. The danger is that an energy-driven headline keeps a three-week-old Fed chair restrictive while the underlying trend is already cooling.
The inflation that is sticky is the new kind
The new inflation reaches a building through technology. For forty years technology was deflationary; chips and software pushed prices down. AI inverted that, because the bottleneck moved from cheap, copyable bits to scarce, physical electrons. Electricity rose 5.9% over the year as data centers drove the fastest power-demand growth in a generation, and you cannot build a transformer with a rate cut. Tariffs, meanwhile, are now running disinflationary, with pass-through largely complete and core goods falling, so the inflation keeping the Fed restrictive is not coming from trade. It is coming from energy and the grid. And the pressure is not spread evenly: more than 70% of the grid nodes posting price increases sit within fifty miles of major data-center load, and wholesale power near those hubs has run as much as 267% higher over five years. Technology has switched sides in the inflation fight, and real estate is where the new bill arrives — hardest on whoever shares a substation with a hyperscaler.
Cash now loses, which is the point
Money-market funds yield about 3.5% and Treasury bills about 3.8%, both now below 4.2% inflation, so the real return on "safe" cash has gone negative. That is not an accident. With debt this size, an administration, a Congress, and a Fed can all live more comfortably with inflation in the 3 to 5% range, because that is what quietly erodes the debt. In a fiscal-dominance world, savers are taxed by inflation so the sovereign can stay solvent, and that is a reason to own scarce, productive assets, from real estate to equities to gold, rather than hold cash.
The reads that carry the argument
| Indicator | Current read | Signal |
|---|---|---|
| CPI headline y/y (May) | 4.2% | Energy mirage |
| CPI core y/y (May) | 2.9% (core m/m halved to +0.2%) | Cooling underneath |
| Electricity y/y (May) | +5.9% | AI-driven, structural |
| Gold | ~$4,250/oz (+~25% y/y) | De-dollarization |
| Gold–dollar correlation | ≈ −0.6 (DXY worst H1 in 50+ yrs) | Same coin |
| 30-yr fixed / 10-yr Treasury | 6.52% / ~4.5% | Floor in the 6s |
| 10-yr since Fed began cutting | +~100bp (while Fed cut −100bp) | Term premium |
| Real yield on cash (MMF − CPI) | ~ −0.7% | Cash loses |
| Small-business hiring plans | net 9% (lowest since 2020) | Labor cooling |
Where the tide reaches the closing table
Bring the whole story down to one deal. Illustrative: at 6.5%, a buyer putting 20% down on the April median Massachusetts single-family home (about $645,000, The Warren Group) pays roughly $3,260 a month in principal and interest. At 3%, where rates sat in 2021, that same house cost about $2,175. That roughly $1,085 monthly gap is the tide going out, made personal, and it is why the market below has split in two.
Sellers get honest: the reconciliation, in the data
68% were active withdrawals or cancellations, not quiet expirations. Sellers are choosing to walk rather than meet the market.
Of original asking, in about five to six weeks. The buyers are there for homes priced to today.
Versus 3.5 statewide. Boston is the most stuck market in the state, more than double any other county.
The homes that came off the market were over-priced from the start. Even after a price cut they were still asking more per square foot than comparable homes were clearing at, about $435 against $407 for single-family, $662 against $570 for condos, and $351 against $310 for multifamily, and they sat about three months. A fresh statewide pull confirms the spread on large samples, 1,046, 838, and 287 delistings against 3,842, 1,859, and 502 recent sales. The gap is not uniform, and that is the investor's edge. Condos were asking roughly 16% above clearing and multifamily about 13%, against just 7% for single-family, which works out to about $125,000 over market on the average delisted condo and $166,000 on the average three-decker. The segments that ran furthest from reality are exactly where a write-to-comps offer is most defensible, and where the seller, already proven willing to walk, has the most room to come down. A home priced to today sold near asking in five to six weeks; one priced to 2021 cut, waited twice as long, and still ended above what the market would pay. Sellers are reconciling a 2021 price with a 2026 reality, the same thing the dollar and the Treasury are being forced to do, and the buyers come back the moment the price does.
The long-term magnet still pulls, even as its near-term drivers wobble
The bull case and the bear case sit side by side, and both are true. In June the Financial Times and Nikkei ranked Boston the number-one US city for foreign multinationals to invest in, citing its talent and research depth even as they flagged its high costs, and the region's most credible new growth story is the longevity economy built on its hospitals, universities, and aging research. At the same time the engines that anchor premium rents are wobbling: Greater Boston lab vacancy sits near 28%, biotech layoffs continued into June, and Massachusetts has roughly 5,000 fewer international students than a year ago, thinning the university-area renter pool. The long-term demand thesis is intact and externally validated; the near-term drivers are in a cyclical correction, which is exactly the gap a disciplined buyer underwrites.
The rent-control question is still open
The statewide rent-control question is still on track for the November 3 ballot, with advocates collecting the second-round signatures due July 8 after the Legislature declined to act by its early-May deadline. Advocates have offered to drop it if the Legislature passes a weaker local-option law, but no deal has been reached and the Supreme Judicial Court has heard but not yet ruled on a challenge. The decisive detail is unchanged: an owner-occupied building of four units or fewer would be exempt, the same building under non-owner-occupied ownership would not. Until it resolves, every Massachusetts multifamily deal carries two scenarios, and the owner-occupied four-unit line is what separates them.
The counter-tide: a supply cliff that favors the Northeast
Name the bullish current too, because it is real. The same rate shock repricing sellers today has frozen new construction: US apartment starts fell more than 40% from 2023 to 2025 and hit their lowest quarter since 2011, with completions sliding from about 550,000 units in 2025 toward roughly 360,000 by 2027. Supply is being strangled just as demand holds — and the Northeast, which barely built through the boom, sits on the favored side of the split, with industry forecasters projecting 4–5% rent growth here in 2026 against 1–2% in the oversupplied Sun Belt. Rental economist Jay Parsons puts the caveat plainly — with supply already low, it is "really all about demand" — but the trough is set. The tide going out reprices your entry today; the building drought quietly supports your rent line tomorrow. Underwrite the trough, and confirm the local rent number against your own MLS pull, not a national average.
The most honest price in Boston
If you want to see the tide going out in a single transaction, look downtown. In June an 11-story office building at 18 Tremont Street sold for $30 million, a 71% loss for an owner who paid $103 million for it in 2019. That is not a distressed outlier; it is the new clearing price for older space, with Boston office vacancy around 18.5% against under 5% in 2019. The reconciliation is the same one your home sellers are doing, just larger and more brutal, and it is feeding the housing story directly: nearly 30 downtown buildings are now approved to convert to apartments. The office market is repricing a decade of cheap-money optimism in one print, and the overflow is becoming housing supply.
The off-ramps from the base case
A view with conviction names what would prove it wrong. These are the four dates and signals that confirm or break the thesis over the next several weeks.
The cleanest test. A second straight cool core confirms the April spike was a one-off and the disinflation is real. A re-acceleration forces a rethink.
The biggest off-ramp. If the peace holds and crude keeps falling, the energy headline drops fast and the Fed gets room to cut. Watch whether the truce sticks.
Whether the hawkish guidance hardens into a hike signal, and whether he treats 4.2% as an energy outlier to look through or a headline to hold against.
The signature deadline and the ballot. Plus a pending SJC ruling. Together they decide whether the multifamily two-scenario problem resolves this year.
One more sits underneath the rate path itself: the mortgage-to-Treasury spread. It is unusually wide, so if quantitative tightening winds down and that spread normalizes, the 30-year fixed could drift toward the low 6s even if the 10-year Treasury does not budge — the one mechanical way mortgages ease without the Fed first winning the inflation fight. (Wolf Street)
We will revisit each of these in Issue 5 in July.
If you are tracking a deal, a refinance, or a development timeline, these are the points that move your math.
Underwrite for a world with no tide
If the tide is going out, the whole old playbook, which assumed the tide would do the work, has to be rewritten. Here is what changes in the next deal you write.
Audit your assumptions for tide assumptions
Appreciation bailing out the price, a refinance bailing out the rate, time bailing out a thin deal: each is a bet that one of the four tailwinds comes back. Strip them out. Underwrite the deal as if nothing upstream will rescue it, because for the first time in forty years, nothing upstream will. If it only works at 5.5% or with 4% annual appreciation, you are not buying real estate, you are buying a forecast the world is pricing against. The refinance bailout in particular is gone, not late: roughly $310 billion of multifamily debt matured in 2025 — the largest slice of a CRE maturity wall S&P sees cresting near $1.26 trillion in 2027 — and owners who locked 3–4% are refinancing into nearly double. Plan for the loan that does not pencil at renewal, not the one that does.
Put the electron in your operating statement
Technology has switched sides, and the bill lands on your utilities line. The Massachusetts twist is counterintuitive: the state already carries some of the nation's highest power prices because ISO-New England is gas-set, and data centers have been slow to arrive here precisely because power is so dear — so your real risk is less a new local hyperscaler than a structurally high, gas-driven bill that the national AI surge keeps lifting from underneath (ISO-NE wholesale ran up 16% year-over-year this spring). Where data centers do cluster, proximity is now an underwriting input: buildings sharing a grid node with major load have seen the steepest increases. Model 5%-plus operating-cost growth, and favor the revenue-generating capital improvements — in-unit laundry, heat-pump conversions — that let the rent line keep pace.
The pulled and expired lists are your pipeline
A delisted home is not a vanished bargain. It is an over-priced listing whose seller has proven they want out and is now sitting off-market with no public competition. Track withdrawn and expired listings and approach those owners at the clearing price for the type, roughly $407 per square foot for single-family, $570 for condos, $310 for multifamily, adjusted for submarket. The condo and multifamily pools are where the delisting-to-clearing gap is widest, so that is where a write-to-comps offer has the most room. In Boston, where 7.9 homes are pulled for every 10 sold, that off-market pool is deepest. And watch the office-to-residential conversions for a new supply channel.
Underwrite multifamily in two ballot scenarios
The rent-control question is unresolved and the owner-occupied four-unit exemption is the line that matters. Underwrite as if it passes and as if it fails. The owner-occupied, house-hackable triple-decker carries a structurally protected return the same building under non-owner-occupied ownership does not. If it passes, that gap widens permanently; if it fails, you still bought a house-hackable triple-decker at 2026 prices. The asymmetry is the trade.
Don't wait for the tide. Underwrite for its absence.
The deals that are honest on day one are the ones still standing when nothing upstream comes to help.
