Clear Lake Coffee Roasters: Political Economy Series: Shelter as a Security: How Housing Became an Asset Class, Who Was Bailed Out the Last Time, and Who Is Holding the Bag This Time
Shelter as a Security: How Housing Became an Asset Class, Who Was Bailed Out the Last Time, and Who Is Holding the Bag This Time
On record prices, a frozen market, and a political economy that has made it nearly impossible to let a house be worth less than it cost
Clear Lake Coffee Roasters · Clear Lake, Iowa · November 2026
A comodity market is supposed to be an exchange of value. Buyers say what they can pay, sellers say what they will take, and the price that emerges carries information about cost, scarcity, about neighborhoods, about what households can bear. The American housing market is still producing rigged prices. It is no longer clear that it is producing reliable, actionable, trustworthy information for the consumer.
Consider the present moment. In September 2026, existing-home sales ran at a seasonally adjusted annual pace of about 3.99 million units; the Associated Press noted that sales were essentially flat last year at a 30-year low. The median sale price that month was $417,500, an all-time high for a September, and August's $429,100 was a record for its month. As of early October, Bankrate's survey put the 30-year fixed rate at 7.53%, a three-year high, with the 10-year Treasury above 5.3%.
Record prices, near-record-low volume, and the most expensive money in years. In an ordinary market those three facts do not coexist for long. Either prices fall to meet the transaction costs, or volume returns. Neither has happened. This post argues that the reason is not a single villain but a premise we rarely question: that a house is an asset class, that its price should rise, and that when it threatens to fall, someone with a printing press or a Treasury account will step in.
I. A Market That Has Stopped Moving
For decades the typical American homeowner sold and moved on after roughly six to seven years; HUD's summary of the National Association of Realtors' annual survey puts the historical norm there. In NAR's 2025 Profile of Home Buyers and Sellers, median tenure before selling hit an all-time high of 11 years. Redfin's own data put it at 12, nearly double the 6.5 years it measured in 2005.
The cause is the lock-in effect. Freddie Mac research found nearly six in ten borrowers with rates at or below 4%, and roughly 60% of all mortgages carry rates below 5%. A household on a 3.5% loan that sells and buys a comparable house today is not making a lateral move; it is trading a cheap liability for an expensive one. On a $400,000 loan, principal and interest at 3.5% is about $1,796 a month. At 7.53% it is about $2,805, roughly $1,000 more per month, a 56% increase, for the same house.
The result at the buying end: first-time buyers made up 21% of purchases in NAR's latest survey, the lowest in 44 years of data, at a median age of 40. (A fair caveat: the Mortgage Bankers Association, using FHFA loan records rather than a survey, puts the median first-time buyer nearer 32 or 33. The direction is the same in both datasets even if the level is disputed.) The median buyer overall is now 59.
The market has not priced its way to a new equilibrium. It has seized up. What holds the price in place while volume collapses?
II. The Original Sin: Treating Shelter as an Asset Class
Begin with a distinction we almost never draw. A house does two jobs. It is a place to live, a consumption good whose value is the shelter it provides. It is also a leveraged, collateralizable, tradable store of wealth. Public policy has spent roughly a century insisting these are the same thing, and that the second is the measure of the first: a "good" housing market is one in which home values rise.
Look at what that definition requires. If prices rising is the goal, then each cohort of owners profits at the expense of the next cohort of buyers. Home equity is, in significant part, a claim on the future wages of whoever comes next. A society cannot coherently promise both that shelter will be affordable for the young and that the home of the old will appreciate faster than incomes. One of those promises has to be broken, and we have chosen which.
The political economy of the choice is not mysterious. Economist William Fischel called it the "homevoter" hypothesis: homeowners, whose largest asset is their house, vote locally and consistently to protect its value, through zoning, permitting, and opposition to new supply. Renters and would-be buyers have no comparable organized constituency. So every downturn produces a defense of prices, and no upswing produces a defense of access.
Then came the financial layer. Mortgages were pooled and securitized, which made shelter into collateral for bonds. In the 2010s, as we describe below, rent itself was packaged into yield. Once the price of housing underwrites trillions in securities, a falling price stops being a household problem and becomes a systemic one, and the state steps in to prevent it. That is the ratchet. An asset class that cannot be allowed to fall is not a market in any meaningful sense; it is a managed commitment, and the people who manage it are not neutral about who benefits.
III. Eighteen Years Ago: The Bailout, and the Lesson Everyone Should Have Learned
The last time housing overheated to this degree, the answer to who bears the loss was settled in a few weeks in the fall of 2008.
The facts are well documented. Bank of America agreed in January to buy Countrywide, once the country's largest mortgage lender, for $4 billion in stock. In March, JPMorgan bought Bear Stearns at $2 a share, with the Federal Reserve assuming about $30 billion of its riskiest assets to make the deal work. In September, Lehman Brothers filed the largest bankruptcy in American history; less than two weeks later regulators seized Washington Mutual, the largest savings and loan, and sold it to JPMorgan the next day. Merrill Lynch went to Bank of America, and Wells Fargo bought Wachovia for $11.7 billion. On October 3, Congress passed the Emergency Economic Stabilization Act, authorizing the $700 billion Troubled Asset Relief Program, which ultimately put money into nearly a thousand companies. Citigroup and Bank of America received $45 billion each, AIG $40 billion, JPMorgan and Wells Fargo $25 billion each.
Meanwhile, from September 2008 through September 2012, roughly four million American households lost their homes to foreclosure, more than 800,000 in 2008 alone (estimates vary by definition and period, and some reporting runs higher). Unemployment rose from 5% to 10% by October 2009, with roughly 8.7 million jobs lost.
A fair accounting requires two caveats. First, most TARP money was repaid, with interest, by 2012; the direct fiscal cost of TARP was a small fraction of the headline number. Second, the largest support was not TARP at all. Bloomberg's analysis of Federal Reserve records, obtained through a Freedom of Information Act request, showed emergency lending to the banks on a scale that dwarfed TARP, extended at below-market rates, and concluded that the six largest banks earned substantial extra profits from it. The reporters' summary of the effect: the Fed's secret lending "helped America's biggest financial firms get bigger."
This is the point at which the economics term of art becomes unavoidable. Moral hazard is what happens when a party is insulated, indemnified from the downside of the risk it chooses to take. The lesson of 2008, taught to every lender, servicer, and investor in the country, was this: if the risk is large enough, the losses will be absorbed by someone else. One of the key contradictions in capital accumulation in this political economy is the ability to privitize gains and socialize cost, loss. No chief executive of a major bank was criminally prosecuted for crisis-era conduct. The institutions that survived did so larger. The institutions that originated the loans were rescued; the borrowers who held them mostly were not.
We do not mean this as a charge of malice. A bailout in a panic may well have been necessary to prevent something worse, and reasonable economists argue it was. We mean it as a description of incentives. An industry that has watched the downside be socialized once will price the next cycle accordingly, with more leverage, thinner capital, and a bet that the same rescue arrives again. And in 2020 it did. The Fed's holdings of mortgage-backed securities roughly doubled, to about $2.7 trillion, and its balance sheet grew from about $4 trillion to nearly $9 trillion by spring 2022. The purpose, preventing a pandemic-era financial collapse, was defensible. The effect on asset prices, housing included, was not a side issue.
IV. Who Benefited From the Consolidation
If the first question is who was rescued, the second is who emerged with more.
The banks. The crisis turned a fragmented mortgage-banking landscape into a concentrated one. The firms that absorbed Countrywide, Washington Mutual, Merrill Lynch, and Wachovia acquired their rivals' franchises at distressed prices, with the government's balance sheet standing behind the transactions. Contemporary trade press noted that the pool of major lenders shrank in 2008 as a series of deals consolidated it, and the title industry followed: Fidelity National Financial's acquisition of LandAmerica's underwriting companies gave it control of roughly 45% of the national title insurance business. The firms regulators called too big to fail became bigger.
The distressed-asset buyers. The wave of foreclosures created an inventory of cheap homes, and a new kind of buyer to acquire them. Blackstone began buying in 2012 through Invitation Homes, reportedly spending about $10 billion on roughly 50,000 homes, in one case 1,400 houses in Atlanta in a single day, at a pace of around $100 million a week. Other large buyers took bulk portfolios, including thousands of homes sold in bulk by Fannie Mae, by then under federal conservatorship. Blackstone brought Invitation Homes public in 2017 and exited its stake by 2019, reportedly earning on the order of $7 billion.
Put those two facts side by side. Families that lost their homes in the crisis did not have access to the same bulk discounts, financing, or time horizon as the funds that bought them from the banks. The foreclosure wave moved ownership from households to capital, and public institutions, through conservatorship and bulk sales, were part of the transfer. Critics have argued, with some force, that many of those evicted could have been kept in their homes by more aggressive loan modification, and that servicers found foreclosure more profitable than modification. We cite this as an argument others have made, not as a settled finding.
The securitizers. Having acquired the homes, the next step was to turn rent into a financial product, with single-family rental securitizations packaging lease payments into bonds. Shelter became a yield instrument at a second remove, which is part of why the "asset class" framing is not merely rhetorical. There are now investors whose fiduciary obligation is to maximize the rent a family can be made to pay.
The incumbent owners. The broadest beneficiaries are the least visible: the households and investors who owned housing before 2012 and whose wealth was then lifted by a decade of falling rates and then a second round of stimulus. That is not a villain. It is a distribution, and it is the same distribution that Piketty describes.
V. Algorithmic Pricing
The strongest legal evidence that algorithms can substitute for competition in housing comes from rentals, and it deserves a precise telling.
In August 2024, the Justice Department and several states sued RealPage, whose software gave landlords daily rent recommendations. The government alleged that landlords fed confidential data into a common model that aligned pricing among competitors, and that the company pushed landlords to auto-accept its recommendations. In November 2025 RealPage settled, paying no damages and admitting no wrongdoing, agreeing to stop using real-time nonpublic data to train its models. The company has consistently denied breaking any law and argues its aggregated data led to lower rents and fewer vacancies.
Private litigation told a harsher story about how defendants valued the risk. Greystar, the largest landlord in the country, agreed to pay $50 million to settle a class action. By May 2026, settlements by 11 landlords exceeded $218 million. A 2024 White House estimate put the cost to renters at $3.8 billion in 2023 alone. California and New York have since passed laws restricting rent-setting software, as have cities including Philadelphia and Seattle.
Automated valuation models, listing-price estimators, and instant-offer engines now sit upstream of nearly every transaction. When buyers, sellers, agents, and lenders anchor on the same few model outputs, the models stop reflecting the market and begin shaping it. That is a hypothesis worth testing, not a proven cartel, and it is the sort of question a well-staffed financial press would be asking. (See our earlier post on where those reporters went.)
VI. The Institutional Investor Question
UBS Financial Services Inc., citing Bank of America research, puts institutional holdings at about 0.35% of the housing stock and about 3% of single-family rentals. Realtor.com's March 2026 deed analysis, defining institutional buyers as those with more than 350 purchases since 2015, found they account for roughly 1% of single-family purchases nationally, and no more than 4.4% in even the most active metro, Memphis. Blackstone says institutional purchases are down over 90% since 2022. According to CNBC reporting summarized by MortgagePoint, institution-owned homes listed for sale were more than double their February level by July, with 54% of those listings carrying price cuts against 38.7% overall (Parcl Labs). Small investors, not institutions, accounted for 66.2% of investor purchases in the first half of 2026.
Locally, the concentration is real. Atlanta has roughly 72,000 institutionally owned single-family rentals, about 30% of that market and ten times the national share, according to the American Economic Liberties Project. In one 12-month stretch beginning July 2021, investors bought roughly a third of the homes sold in metro Atlanta, mostly in cash at asking price. SFR Analytics found ZIP codes in Atlanta, Charlotte, Phoenix, and Jacksonville where institutions own more than 10% of single-family homes, and neighborhoods where it is upward of 50%. The Urban Institute found that 45% of institutionally owned single-family homes sit in just six markets.
Why does a small national share matter if it is concentrated? Because price formation is local and set at the margin. When an all-cash buyer, with no mortgage, no appraisal contingency, and a rental yield to underwrite, wins a bid at or above asking, that sale becomes a comparable. Comparables feed appraisals, appraisals feed lenders, and lenders set the next buyer's maximum bid. A few percent of transactions can reset the price per square foot of a street.
Investors still use cash far more often than ordinary buyers: 55.5% of investor purchases in the first half of 2026 versus 30% for typical buyers, down from a 68% peak in 2021 and 2022. NAR's survey found all-cash buyers at 26% of all purchases, tied for the highest on record. Higher rates penalize everyone who needs a loan and spare the buyer who does not. That is not a conspiracy; it is what a rate shock does when some bidders are exempt from it.
Our conclusion: the claim that institutions hold a nationwide pricing floor is not supported by the numbers, and anyone making it should be asked for theirs. The claim that all-cash and investor bidding distorts price discovery in specific, identifiable submarkets is well supported. And the deeper point is that today's institutional share is the residue of the 2012 transfer, not a separate phenomenon.
VII. Intermediation: What the Middleman Costs, and What It Buys
The traditional commission was an arrangement in which sellers paid both agents, buyer-agent pay was set through the listing, and the total hovered near 5 to 6% regardless of effort. In March 2024 NAR agreed to a $418 million settlement, with new rules effective August 17, 2024. The expectation, shared even by industry critics, was that commissions would fall.
They have not. Clever Real Estate reports that the national average total commission rose from 5.32% in 2024 to 5.46% in 2026, with the buyer's agent portion up from 2.58% to 2.70%. Only 28% of agents in its August 2026 survey said commissions had compressed in their market. Redfin's data show the buyer-agent fee dipping to 2.36% when the rules took effect and rebounding to about 2.42% by late 2025. Because commissions are a percentage, rising prices raise agents' pay in absolute dollars even when the rate edges down.
Commission is only one component. Sellers commonly cite all-in selling costs, including both agents, closing costs, concessions, repairs, and staging, of 8 to 10% of the sale price. That is the basis of the thesis that intermediation adds roughly a tenth to the cost of housing. We present it as a thesis, not a measured fact: 10% is a commonly used upper-range estimate of seller friction, not a regression coefficient, and economists disagree about how much is passed into list prices versus absorbed by sellers.
What is easier to defend is the question of utility. A generation ago an agent controlled access to the listings. Today every listing is public within hours, and the agent's marginal contribution on a standard transaction has shrunk toward negotiation, paperwork, and risk management. A fee set when the agent held the information is still being paid now that the information is free.
VIII. The Valuation Chain: Who Is Paid to Get the Number Right?
In a typical purchase, the listing agent, the buyer's agent, the loan officer, and the lender are all paid when the deal closes. The appraiser, whose job is to say whether the price is defensible, is the one participant whose independence is supposed to serve as the check.
We are describing a structure in which every party benefits from a closing and no party's income depends on whether the price is right 5 years later. In such a structure, drift toward the contract price is the path of least resistance.
Appraisers say so themselves. One lender's guide for buyers states bluntly that "appraisers correlate to contract price," explaining that the signed contract is often the best available indication of value. That is sensible practice, and also circular: the contract price becomes the evidence for the value that justifies the loan that funds the contract price.
There is academic evidence of something worse at the margin. A paper presented at the American Economic Association's annual conference examined new purchase appraisals and documented systematic misreporting consistent with intentional inflation of property values, concluding that post-crisis reforms had not eliminated bias in the collateral values attached to mortgage loans.
Meanwhile the appraisal is being automated away. At the March 2021 peak, waivers accounted for nearly half of loans backed by the government-sponsored enterprises, per the AEI Housing Center; by October 2025 they were around 23%. But FHFA has raised the maximum loan-to-value ratio for waiver-eligible purchase loans from 80% to 90%, and waiver use on higher-LTV loans jumped from about 2% to about 15% in a few months of 2025. A waiver means no human looked at the house.
The consequence of an overstated number does not fall evenly. The agent has been paid, the lender has sold the loan, and the borrower, who put down the median 10% that NAR reports for first-time buyers, holds the risk.
IX. The Arithmetic of the Last Buyer
Take a buyer who purchases a $450,000 home with 10% down: a $405,000 loan at 7.53% and a payment of about $2,840 before taxes and insurance. Two years in, the balance has barely moved, to about $397,000. If prices fall 10%, the house is worth $405,000, and after a typical 5.5% selling cost the owner walks away roughly $14,500 short. A 15% decline leaves them about $36,000 short, and 20% about $57,000.
Lock-in cuts both ways. The seller on a 3.5% mortgage is insulated from a decline; they can simply stay. The buyer who paid today's record price at today's record rate has no such cushion, and if circumstances force a sale, the market's depth is exactly what has been missing for three years. A thin market is easy to hold up and hard to exit.
The households with the least equity and the highest payments bear the largest share of the tail risk, and they are the least able to hold through it. If 2008 taught the system anything, it is that there is a hierarchy of who gets rescued. Borrowers who bought at the top were not at the top of it.
X. The Canary: Condominiums, the Hidden Liability, and the Dues That Ate the Price
There is one corner of the housing market where prices have been allowed to fall, and it tells us something about every other corner. A condominium buyer does not simply buy a unit. They buy a share of a building: its roof, its elevators, its pipes, its insurance policy, and its unpaid maintenance. The price on the contract is only part of the cost. The rest arrives monthly in the form of association dues, and occasionally as a lump-sum special assessment.
For years the second number was easy to ignore. Now it is the number that sets the first.
What the data show, and where. The national picture is mixed, and we should say so. The median condo price nationally has been roughly flat to slightly up in recent months: Wolf Street's analysis of NAR data put it at $363,000 in September, up 1.2% from a year earlier, and NAR's chief economist has described the condo market as showing signs of stabilization in some metros. But the volume tells a different story. The same analysis shows condo sales at a record low and supply at or near a record high. And the aggregate hides severe local declines. Using the same data, median condo prices were down 27% in Fort Myers, 24% in Sarasota County, and 22% in both Tampa and Garland, Texas. In Southern California, Attom data reported by The Real Deal showed condo prices falling by the biggest year-over-year amount in 14 years in April, with the median single-family list price there 22% above the condo price. Zillow's data showed Atlanta-area condos down 4.6% over the prior year.
The trend was visible earlier. In May 2025, Redfin found the national median condo price had fallen 2.2% year over year to $354,100, the second-largest annual drop in its records back to 2012, while single-family prices still rose 0.5%. Condo sales fell nearly 12%, more than three times the decline in single-family sales, and in Deltona, Florida, the median condo price dropped 32.2% in a single year. Redfin attributed the slump in part to rising association fees and insurance costs.
The other side of the ledger. While prices softened, the monthly cost of owning went the other way. A Foundation for Community Association Research survey in May 2026 found that 54% of responding condominium communities were raising regular assessments to meet reserve funding expectations, and 14% were using special assessments. Redfin's earlier analysis found HOA dues rising 12.8% in West Palm Beach, with typical Miami dues around $835 a month, and faster-than-average increases in Fort Worth (14.6%), Sacramento (13.1%), and Atlanta (12.2%). Special assessments of $100,000 or more per unit have been reported in Florida. The Movoto research group puts the national median condo HOA fee at $421 a month among listings that report one. A mortgage-broker trade association white paper claims that roughly 40% of owners have faced a special assessment in the past three years; we cite that as an advocacy group's figure, not an audited one.
Why Florida is the epicenter. The collapse of the Champlain Towers South condominium in Surfside in June 2021 led Florida to require milestone structural inspections and funded structural-integrity reserves for older buildings, while the state's property-insurance market tightened sharply. Both pushed dues up at once. Reporting by NewsNation found the fees hitting seniors on fixed incomes especially hard, and a source quoted by the Business Observer described a "massive influx" of listings from owners who can no longer afford their payments.
The mechanism. The arithmetic is unforgiving. At today's 7.53% mortgage rate, every $1 of additional monthly dues reduces the loan a buyer can qualify for by roughly $143, because lenders count the dues in the buyer's debt-to-income ratio. A $500 increase in monthly dues therefore removes roughly $71,000 of purchasing power from a buyer who is at their lending limit. A $500 jump is exactly the kind of increase described above. Dues do not sit alongside the price; they are subtracted from it.
The financing channel compounds this. Fannie Mae and Freddie Mac decline to buy loans in condo projects with underfunded reserves, heavy delinquencies, or unresolved litigation, so buyers in those buildings need cash or a specialty lender. The mortgage broker association's white paper says more than 1,400 associations are on Fannie Mae's unavailable-project list. A thin buyer pool means longer time on market and deeper price cuts.
Then there is the feedback loop, which we offer as reasoning rather than measurement. A building's costs are fixed by its structure and insurance, so when owners leave or default, the same bill is spread across fewer solvent owners, which raises dues, which pushes out more owners. The unit does not lose value for lack of walls and a roof. It loses value because the liability attached to it has become visible.
What this reveals about the asset class. Here is why we call condominiums the canary. A condo is the clearest case of a point we made earlier: a home is not only an asset, it is a stream of obligations. In a single-family house the owner can defer the roof and the furnace, and the deferral stays invisible to the next buyer until the inspection. In a condominium, deferred maintenance is institutionalized in the association's reserve account, and the gap between what has been saved and what is owed is, in principle, a number someone can read. A common critique, which we cannot document building by building, is that for years dues were kept low because low dues kept units easy to sell, and the true cost was pushed forward to whoever owned the unit when the bill came due. If so, it is the structure the 2008 section described in miniature: the downside lands on someone else later.
Later has arrived. And notice what happened when it did: unlike single-family housing, where lock-in, cash buyers, and scarcity have held prices up, the condo market let the carrying cost reach the price. It is the one place where the full cost of ownership is printed on the page, and where, as a result, the market is doing what markets do. The rest of housing has been spared, not because it is healthier, but because its liabilities have been easier to hide.
The caution runs the other way too. Condo markets in many places are stable or recovering, and the worst declines are concentrated in older, coastal, heavily regulated Florida buildings and in markets with new-construction oversupply. We are not predicting that every condominium reprices. We are observing that where owners' obligations became visible, prices followed.
XI. The Other Half of Real Estate: The Office Collapse, and the Housing Boom That Is Underwriting It
When people say "real estate," they usually mean houses. But the commercial market is the larger and, at the moment, the sicker half, and its condition cannot be understood without the residential boom we have been describing. One half is at record prices. The other is setting records for distress. The interesting question is not how both can be true but whether one is holding the other up.
The state of office. According to Trepp, the delinquency rate on office loans in commercial mortgage-backed securities hit an all-time high of 12.34% in January 2026, above the peak reached after the financial crisis, and was still around 12% in August. Much of that is not borrowers who stopped paying; analysts note that many of the newly delinquent loans are maturity defaults, loans that came due and could not be refinanced at today's rates against today's building values. The wall of paper still to come is large. Office loan maturities of roughly $289 billion fall due through 2028, and about 59% of that volume was originated before 2021, at valuations that predate both the rate shock and the remote-work shift. Morningstar DBRS expects more than half of the $100-billion-plus of CMBS loans maturing in 2026, across all property types, to fail to repay at maturity. Yardi expects office values to remain under pressure until the maturity wave subsides.
Vacancy tells the same story, with a measurement caveat. Moody's Analytics recorded a national office vacancy rate of 20.7% in August 2025, a record, while other trackers using different methods put it near 14%, itself the highest on record and above the 13% mark that was rarely breached before 2020. Moody's pointed to economic uncertainty and pressure from AI on white-collar demand. The picture is also bifurcated: in Manhattan, Class A buildings had a 9% vacancy rate against 14.3% for Class C. The distress lives in commodity Class B and C buildings with dated systems in the wrong locations. Trophy towers are doing fine. That distinction matters for what follows.
Channel one: the tax base. The most concrete way housing is propping up commercial real estate is also the least glamorous: the property tax roll. American central cities have long relied on downtown office towers to pay for services. As office values fall and home values rise, the arithmetic shifts the burden onto owners of houses.
In Minneapolis, commercial property fell from 34.7% of the city's net tax capacity in 2016 to 24.9% in 2026, while the residential burden rose. The city assessor reported a 20% drop in commercial building values and a further 2-point shift to residential, which the city said would cost the average homeowner about $240 more for 2026. In Boston, where state law restricts how rates can be adjusted between sectors, homeowners are now shouldering the largest share of the tax load in more than 40 years, with office values down 5% last year and on track to fall another 6%. Washington, D.C.'s commercial assessments fell by $8 billion, about 8%. Municipal budget forecasts cited by Propmodo anticipated office property tax receipts falling nearly 10% in 2025 and another 12% in 2026.
The honest caveat is that the effect is uneven. Twin Cities analysts say the office decline is not necessarily the largest factor behind rising residential tax bills, Ramsey County projected a shift of under 1% for St. Paul, and Spokane's assessor did not expect a noticeable shift. Assessments also lag the real decline in office values. But the logic is simple and it runs in one direction: a house assessed at a record is a house taxed at a record, and the surplus covers what the tower no longer pays. A homeowner's celebrated equity is in part funding the municipal bills that office value used to cover.
Channel two: the conversion exit. A dead office needs a new use, and the only use with durable demand and strong rents is housing. One industry analysis counted a record 70,700 office-to-apartment conversion units. Conversions only work if the apartments they produce can rent or sell for enough to justify the construction cost, which is another way of saying that high housing prices are what make the exit possible. And residential property is generally taxed at lower rates than commercial, so conversions soften the municipal blow rather than offset it.
Channel three: the bank balance sheet. Commercial real estate loans make up roughly a quarter of average bank assets, about $2.7 trillion by one NBER estimate, with regional banks alone holding about $1.3 trillion by another analyst's count. A New York Fed staff report found that the share of bank capital tied to maturing CRE debt had grown to 27%, up from 16% in 2020, and warned that banks delaying recognition of troubled loans, so-called extend-and-pretend, created a financial stability risk. The same banks carry large residential mortgage and home-equity books whose collateral is at record highs and performing well. It is an inference, not a measured fact, that this strength lets banks absorb office losses gradually rather than all at once. If residential values fell, those banks would face losses on both sides of the ledger together.
We should be fair to the other view. A 2026 Federal Reserve Board paper, "Pretend or Amend?", found little evidence that banks broadly used extend-and-pretend, and large banks in particular appear not to. Predictions of mass regional bank failures have not materialized; only one small bank has failed. Several regional banks have been reducing office exposure and reporting improved CRE loan performance. The doom forecasts of the last three years have been wrong so far. One reason they may be wrong, and this too is an inference, is precisely that the rest of the balance sheet is sound, which makes housing a load-bearing wall.
Channel four: the household balance sheet. The most diffuse channel is the wealth effect: owners who feel richer spend, and spending supports retail, hospitality, and logistics space. We do not have a measured estimate of how much, and we flag it as reasoning rather than evidence.
Is housing "one of the only things" holding up commercial real estate? As stated, that is too strong. Other supports exist: strong demand for the best buildings, lower short-term rates in 2026, and industrial and data-center space, the latter arguably a beneficiary of the AI capital spending we examined in an earlier post. That makes two bubbles, housing and AI, holding up the commercial market's overall picture, which is hardly reassuring. What the evidence does support is a narrower and still troubling claim. The tax roll and, plausibly, the banking system are being cushioned by record residential values, and the cushion works only as long as housing holds.
That is the structural risk. It is not the 2008 story run backward, but it rhymes with it. The healthy half of real estate is subsidizing the sick half through municipal tax rolls and bank capital, which means the two halves are not diversified; they are correlated. If rates stay high and housing softens just as the maturity wall peaks, the support and the load arrive together. And the incentive to delay recognizing losses, to extend and amend rather than write down, is the same incentive we described in the bailout: when the downside is shared, the temptation is to wait for someone else to bear it.
XII. What Honest Price Discovery Would Require
If the aim is a market where price carries information, several things follow.
Stop treating appreciation as the goal. The policy target should be affordability and stability of shelter, not the rising value of a stock of houses. That means saying aloud that falling real prices, relative to income, are a success, not a crisis.
End the implicit guarantee. If lenders and investors are to be exposed to the risk of the assets they hold, the rescue cannot be a certainty. Capital requirements, resolution regimes that actually impose losses on creditors and executives, and clawbacks of compensation tied to failed institutions would restore the downside that moral hazard removes.
Transparency and antitrust in valuation. Disclose when automated models rather than people set the number, and extend the logic of the RealPage settlements, audited common data inputs and independence of pricing, to wherever it applies.
Independence in appraisal. Pay for accuracy, not accommodation, and revisit waiver thresholds when the cycle is at a record high, not only when it is low.
Unbundled intermediation. The post-settlement data show that rule changes alone do not move prices. Real unbundling requires enforcement and consumer awareness.
Unlocking the lock-in and the supply. Portable or assumable mortgages would let households move without surrendering their rate; zoning and permitting reform would address the scarcity that makes every other distortion more profitable.
Make carrying costs visible before the contract. Require sellers and associations to disclose reserve funding levels, pending assessments, insurance trends, and delinquency rates up front, and require reserves to be funded honestly from day one rather than deferred to a later owner.
Let commercial losses be recognized. Office values will reset, and delay only moves the loss to whoever is least able to see it coming: homeowners through the tax roll, and depositors through bank capital. Transparent loss recognition, honest stress tests, and municipal budgets that show homeowners how much of their bill is backfilling vacant towers would turn a quiet subsidy into a visible choice.
A seat for the missing reporter. None of these problems is hidden; they are simply unstaffed. A local housing beat, with someone reading the deed records and asking who bought the block, would do more than any single reform to restore a basic fact of markets: that the person who sets a price should have to explain it.
XIII. From the Roastery
We sell a product whose price is easy to explain. The green coffee costs what it costs, we publish what we paid, and the customer can check our work. We would not know how to run a business in which the party setting the price was paid whether or not it was right, in which the losses were shared with the public when the bet went bad, and in which the one bearing the consequences was the one with the least information.
Housing is not a luxury good. It is where families live, and its price is the largest number most people will ever put their names to. We spent 18 years treating that number as an investment that must never fall, and we paid for the privilege once. The question for the next 18 is whether shelter can be allowed to be what it always was: a place to live, priced by what households can actually bear.
Clear Lake Coffee Roasters LLC · 15068 Hill St · Clear Lake, Iowa 50428 hello@clearlakecoffeeroasters.com · clearlakecoffeeroasters.com
Sources and further reading: National Association of Realtors, 2025 Profile of Home Buyers and Sellers; NAR existing-home sales, September 2026 (via the Associated Press, October 13, 2026); Bankrate national mortgage survey, October 7, 2026; HUD PD&R National Housing Market Summary, 3Q 2025; Redfin homeowner tenure report, March 2026; Freddie Mac research on rate lock-in; Mortgage Bankers Association/FHFA analysis of first-time buyer age; Realtor.com institutional investor deed analysis, March 2026; SFR Analytics, January 2026; American Economic Liberties Project, Atlanta single-family rental report (2026); Urban Institute analysis of institutional single-family ownership; UBS Wealth Management and Bank of America research; MortgagePoint and CNBC reporting on institutional listings, July 2026; RISMedia, October 2026; U.S. Department of Justice and Associated Press reporting on the RealPage settlement (November 2025); The Real Deal, May 2026; Clever Real Estate and Redfin commission data, 2024 to 2026; AEI Housing Center appraisal waiver data via Working RE; American Economic Association 2021 conference paper on appraisal inflation; FHFA waiver LTV changes (2025); Britannica and EBSCO summaries of the 2007 to 2008 financial crisis and TARP; Inman, "A Grim 2008 in Real Estate" (December 2008); Bloomberg News reporting on Federal Reserve emergency lending (summarized in The Week); Fox Business and RealtyTrac foreclosure data; reporting on Blackstone and Invitation Homes, including Laura Gottesdiener (2013) and coverage of the 2017 IPO; William Fischel, The Homevoter Hypothesis (2001); Federal Reserve balance sheet data; Thomas Piketty, Capital in the Twenty-First Century (2014). Figures reflect sources as of early November 2026. Where we cite reporting or advocacy we have attributed it as such. This post is commentary, not financial or legal advice.
Six reasons for making Clear Lake Coffee Roasters - CLCR - your go-to coffee roaster:
☕️ We are a local family-run business located in the heart of Clear Lake, Iowa.
☕️ We go to great lengths to find only the finest and ethically sourced coffee around, from the top 2% of coffee beans in the world.
☕️ We only source 100% certified Arabica coffee beans, carefully hand-selecting each coffee based on specific quality and taste attributes.
☕️ Our roasting process has been refined over the years and each roast profile is individually designed to complement the nuances of the coffee we source, from Cup of Excellence (COE) award-winning producers.
☕️ By roasting in smaller batches, we can ensure our coffee is ALWAYS fresh, in fact, we roast your coffee only after you place an order - the same day your order ships out.
☕️ At CLCR, we are dedicated to a single mission: the unyielding pursuit of coffee perfection in every cup.
We would give you more reasons, but rather than reading it's better if you visit our website, purchase a bag or two, and experience a unique caffeinated or half-caff journey for yourself 😊!
Explore goodness. Click. Buy. Smile.
Clear Lake City Park Beach - Named one of USA Today's top 50 Beaches in the United States 2025
Leave a comment
Please note, comments must be approved before they are published