The Things That Aren’t Really Markets
The Things That Aren’t Really Markets
Some things look like markets. They have prices, sellers, contracts, all the vocabulary. But they don’t behave like markets, and pretending they do has cost us forty years.
Start with a question you’ve probably asked yourself
Why does everything essential keep getting more expensive?
Not everything, actually. Televisions are cheaper than they’ve ever been. Clothing, in real terms, costs less than it did when your parents were young. Airfare, computers, toys: the market did what markets are supposed to do, and prices fell or held steady while quality rose.
But healthcare. Housing. Childcare. College. Water bills, electricity, internet. These didn’t just rise with inflation; they rose at multiples of it, decade after decade, until a middle-class life that one income could cover in 1980 now strains two.
So here’s the real question: why those categories and not the others? What do healthcare, housing, childcare, education, water, electricity, internet, and food access have in common that televisions don’t?
It’s not that they’re run by worse people. It’s not bad luck. It’s structure, and once you see the structure, you can’t unsee it.
The four conditions
Economists have a technical definition of a “public good”: something you can’t fence off (like clean air) and that doesn’t get used up when one person enjoys it (like a lighthouse beam). By that definition, healthcare and housing aren’t public goods. You can absolutely be locked out of a hospital or an apartment. So mainstream economics files them under “private goods with some market failures,” and the policy conversation follows from there: tweak the incentives, add a subsidy, run an information campaign.
That filing is wrong, and it’s been producing wrong answers for forty years.
What actually matters isn’t whether something fits the textbook definition. It’s whether four conditions show up in the same place at the same time:
First: you can’t walk away. Economists call this inelastic demand. You cannot opt out of healthcare when you’re sick. You cannot opt out of shelter, or food, or the electricity that keeps your insulin cold. When the price goes up, you still have to buy. The seller knows this.
Second: there’s nowhere else to go. One broadband company serves your address. One hospital system runs every ER within an hour. A handful of landlords own the rentals in your part of town. The whole point of a market is that competition punishes overcharging. When competition is gone, the punishment is gone, whether or not anything technically qualifies as a monopoly.
Third: being priced out isn’t an inconvenience; it’s a harm. Skip the streaming service and you miss some shows. Skip insulin, or heat in January, or a home, and you are in genuine danger. The severity of the harm removes your exit option in a second, deeper way: it’s not just that you don’t want to leave. You can’t survive leaving.
Fourth: we’ve already run the experiment. This isn’t a prediction about what private provision might do. It’s a record of what it has done. Forty years of results are in, and by any honest reading, private provision in these sectors has not delivered fair access at sustainable cost.
Here’s the thing to hold onto: where all four conditions stack up in the same sector, overcharging isn’t a risk. It’s a guarantee. Not because the people involved are greedy monsters, but because the conditions make it the easiest path available. Which brings us to the mechanism.
Nobody has to be a villain
This is the part where most political arguments go wrong, so let’s slow down.
Organizations behave a lot like people: they conserve energy, chase rewards, avoid pain, and settle into whatever equilibrium requires the least friction. In a genuinely competitive market, the lowest-friction path includes treating customers well, because customers who feel mistreated leave. That’s the whole trick of markets. It’s a good trick.
But change the conditions and you change the path. In a sector where customers can’t leave, the lowest-friction path shifts: charge more, maintain less, and let the costs land on people who have no exit. No conspiracy required. The hospital administrator who approves an aggressive billing policy isn’t necessarily a bad person. She’s making a locally reasonable decision inside a system that rewards it and, crucially, shields her from ever seeing what it does.
That shielding matters more than it sounds. The executive who sets hospital prices will never have medical debt. The utility board that approves rate hikes will not lose heat in February. The landlord who owns 85,000 houses will never experience housing instability. The distance between the decision and its consequences isn’t a side effect of power in these sectors. It’s what the power is. And when decision-makers never feel the results of their decisions, the people harmed become steadily harder for them to see. Not through malice. Through insulation.
One more idea, borrowed from the philosopher Philip Pettit, sharpens the picture. Pettit argues that domination isn’t just the use of arbitrary power over you; it’s the existence of that power when you have no alternative. Think of a bully. A bully doesn’t have to be actively hitting you to change your behavior. You walk different hallways. You watch the clock. You shrink a little, preemptively, because you know what he could do. Now, with an ordinary bully, there’s at least an answer: people can choose not to be around bullies. You change schools, change jobs, change neighborhoods. But you can’t do that here. There is no hallway that routes around the only hospital system in your county, no schedule that avoids the rent. Structural extraction is bullying you cannot walk away from, and it changes your behavior the same way. You don’t fight the bill. You don’t push for the referral. You ration your own care preemptively. The power works even while it rests.
This is why the usual fixes fail. Consumer awareness campaigns, choice initiatives, marginal regulations: they all assume the system is malfunctioning and just needs better information. It isn’t malfunctioning. It’s responding correctly to its incentives. If the outcomes are wrong, the architecture is wrong, and the architecture is what has to change.
The accelerant: private equity
Everything described so far happens under ordinary ownership. Private equity is what happens when a business model is purpose-built to run that dynamic at maximum speed.
Here’s the model, plainly. A private equity firm buys a hospital chain, a nursing home network, a childcare company, a housing portfolio, or a water system, using mostly borrowed money. The debt doesn’t go on the firm’s books. It goes on the acquired company’s books. The hospital now owes the money that was borrowed to buy the hospital.
So the hospital must generate enough cash to service that debt and deliver the returns the fund’s investors expect, all within the typical five-to-seven-year window before the firm sells and moves on. In a competitive market, that pressure would be impossible to sustain; customers would flee. In a sector where customers can’t flee, the pressure gets passed straight through: onto patients through billing, onto tenants through rent hikes and skipped repairs, onto children through thinner staffing, onto towns through closed service lines. And whatever long-term damage the squeeze causes (the deferred maintenance, the hollowed-out workforce, the fragile balance sheet) becomes the next owner’s problem, or the community’s. The firm is gone by then. The community isn’t.
The evidence here is not vibes. It’s peer-reviewed.
In healthcare: a 2023 study in JAMA found that private equity acquisition of hospitals was followed by a roughly 25 percent increase in hospital-acquired complications: falls, infections, central line problems. A National Bureau of Economic Research study of nursing homes found PE acquisition associated with about a 10 percent increase in short-term deaths among Medicare patients. PE-owned physician staffing firms became the main engine of “surprise billing,” a practice so pervasive that Congress passed a law, the No Surprises Act, specifically to stop it. And when Steward Health Care collapsed in 2024 after years of PE ownership, communities in eight states faced sudden hospital closures. The pattern repeats: buy, load with debt, extract, exit, and let someone else absorb the wreckage.
In housing: during the foreclosure crisis, when ordinary buyers couldn’t get mortgages, institutional investors with cash bought tens of thousands of single-family homes at depressed prices. Invitation Homes, the largest single-family landlord in the country, was assembled from a Blackstone portfolio this way (Blackstone has since sold its stake, but the crisis-era origin isn’t disputed). The industry likes to point out that institutional owners hold under one percent of homes nationally. True, and beside the point: housing markets are local. A company invisible in the national statistics can still be one of the two or three biggest landlords in your metro area, which is exactly where pricing power lives. Federal Reserve research has linked institutional purchases to higher local rents and lower homeownership rates. And in apartment buildings, the Justice Department sued the software company RealPage for enabling nominally competing landlords to coordinate rents through a shared pricing algorithm; RealPage settled in late 2025, agreeing to stop using competitors’ private data in its recommendations. Call it what it structurally is: cartel behavior in a sector where the customers can’t leave.
In childcare: the playbook is especially clean. Buy centers, cut staffing to the legal minimum, hold worker pay down, then sell the buildings to a real estate trust and lease them back at inflated rates, saddling the operation with occupancy costs designed to maximize what can be pulled out. The results are visible on both ends: childcare workers earn a median of about $14.60 an hour, among the lowest wages of any licensed profession, while families pay costs that in many states exceed college tuition. The gap between poverty wages on one side and unaffordable prices on the other is not a mystery. It’s the margin.
Notice what this evidence lets us say. Ordinary investment means putting money into something so it grows; the investor wins when the company thrives long-term. What PE does in these sectors is the opposite: pull value out fast and leave before the bill arrives. Same legal paperwork, opposite function.
What actually works
Here’s where the argument turns constructive, and where it gets persistently misrepresented, so let’s be precise about what’s being claimed.
Calling something a common good sector is not a call to nationalize it, ban private business, or put the government in charge of everything. The claim is narrower and, honestly, more market-friendly than its opponents admit: where all four conditions hold and private provision has demonstrably failed, there must be a genuine public alternative. Not a subsidy. Not a regulation. An actual alternative that people can choose.
Why an alternative rather than just rules? Two reasons.
First, it restores the exit. Markets discipline sellers only when buyers can leave. A public option at cost-based pricing recreates the exit that concentration destroyed. Private providers can still operate, still compete, still profit. They just can’t charge captive-audience prices anymore, because the audience is no longer captive. This is the least intrusive fix available: it doesn’t tell private firms what to charge. It simply makes overcharging unprofitable again, which is what competition was supposed to do all along.
Second, it makes the truth measurable. When Medicare exists next to private insurance, we can compare overhead. When a city runs its own broadband, we can compare speeds and prices. Without the public benchmark, the private market gets to claim its prices reflect the true cost of provision, and nobody can prove otherwise. This, incidentally, is a big part of why incumbent industries fight public options so ferociously: not just the lost customers, but the lost cover.
And the track record? The standard objection is that government provision is inherently inefficient. That’s an empirical claim, so check it empirically:
Healthcare. Every wealthy peer nation covers everyone at 40 to 60 percent of what Americans pay per person, with better average outcomes. Their systems differ (single-payer, regulated multi-payer, hybrids), but they share one constant: some public structure sets a floor under pricing. On administrative costs, the famous “Medicare 2 percent vs. private insurance 12–18 percent” comparison is genuinely contested; Medicare’s older, sicker population inflates the denominator, and fairer per-person accounting narrows the gap. But even under the most conservative accounting supported by CBO and MedPAC work, private insurance carries meaningfully higher overhead, worth on the order of $80 billion a year across the system, and possibly double that. That money buys billing departments, prior-authorization queues, and denial letters. It does not buy healthcare.
Housing. In Vienna, about 60 percent of residents live in municipal or cooperative housing, maintained at good quality, at roughly a third of market cost. Singapore houses about 80 percent of its people in public housing that residents own and build wealth through. The shared design principle: keep a chunk of the housing stock permanently outside the speculation cycle, so it anchors prices for everyone. America’s closest homegrown versions, community land trusts and limited-equity co-ops, work fine; they just don’t exist at scale yet. (And no, the failed American housing projects of the last century don’t refute this. Those failed by concentrating poor families in isolated towers, a design failure, not proof that public housing can’t work. Vienna is the counterexample, sitting in plain sight.)
Childcare. France’s crèche system, the Scandinavian models, and Canada’s $10-a-day program differ in detail but converge on the result: more parents working, better outcomes for kids across income levels, and net positive returns to the public purse. The American research base agrees. James Heckman won a Nobel Prize partly for methods that catch statistical bias in imperfect studies, and when he turned those very methods on the flagship Perry Preschool data (whose original randomization had real problems), the effects held: an estimated $7 to $12 back for every dollar invested in high-quality early childhood programs, through lower crime, less remedial education, higher lifetime earnings. That’s not a projection. It’s measured lifetimes.
Utilities and broadband. Rural electric cooperatives built under the New Deal have beaten investor-owned utilities on cost and reliability for eighty years. Mature municipal broadband networks routinely deliver higher speeds at lower prices than the incumbents. The record has failures, and they’re instructive: the networks that struggled were underfunded, under-scaled, or built in states where incumbent lobbyists had already passed laws designed to make them fail. More than twenty states restrict or ban municipal broadband. Ask yourself why an industry confident in its own efficiency would need laws forbidding the comparison.
Water deserves a special word, because it’s the case where the argument should already be over. Demand is perfectly inelastic (you die without it), the pipe network is a natural monopoly (nobody sane builds competing pipes to your house), and the harm from shutoff is immediate. Yet privatized systems keep producing the same pattern: rate increases of 50 to 100 percent within five years, quietly deferred maintenance, and regulators gradually captured by the companies they oversee, until the harm surfaces as lead in the pipes or shutoffs in poor neighborhoods, by which point the oversight machinery has already been compromised. Public and cooperative water systems, the norm in peer nations and in most of America before the privatization wave, don’t have this pattern, because they don’t have the extraction motive that drives it.
Across every sector, the same finding: public or cooperative alternatives deliver better outcomes at lower cost than private provision operating without competition. The “government is inefficient” claim survives in political talk shows. It does not survive the data. It persists because the industries that benefit from it can afford to keep it alive.
The other half: keep private equity out
Public options are half the design. The other half is a prohibition, and it follows from everything above.
If common good sectors require long time horizons (water infrastructure is planned in fifty-year increments; a child’s development can’t be paused for a fund cycle), accountability to consequences, and debt loads compatible with the mission, then the PE model, which is defined by short horizons, exit-before-consequences, and maximum leverage, is structurally incompatible with these sectors. Not morally suspect. Structurally incompatible, the way a drag racer is incompatible with a school bus route.
This isn’t an argument against private ownership of hospitals or childcare centers. Family-owned, physician-owned, cooperative, conventional corporate: all can work. It’s an argument against one specific ownership model, and banning an ownership model from a sensitive sector is not some radical novelty. Glass-Steagall barred banks from combining commercial and investment banking for sixty years. The FCC caps media ownership concentration on public-interest grounds. The tools already exist: ownership disclosure requirements, limits on debt loading in covered sectors, mandatory long-term operating commitments as a condition of acquisition, personal liability for fund managers whose portfolio companies collapse after exit, and regional market-share caps.
Neither half works alone. Public options without PE limits get strangled politically by incumbents with extraction money to spend. PE limits without public options leave the concentrated market intact, just slightly slower. The design needs both, because both aim at the same target: reconnecting decisions to their consequences.
What remains when the arguments run out
Every version of this proposal gets met with the same words: overreach, socialism, inefficiency, killing innovation. Notice what those words have in common. None of them engages the evidence. The efficiency claim has been tested in every peer nation and every sector reviewed here, and it lost. The accountability claim has been tested everywhere private equity has operated in these sectors, and it lost too. When a position keeps losing on evidence but keeps winning in politics, you’re no longer looking at an argument. You’re looking at an interest, wearing an argument’s clothes.
And that, finally, is why the naming matters. “Common good sectors” isn’t a slogan; it’s a diagnosis, and a testable one. Any of these eight sectors could, in principle, fail the four-condition test. None of them do. They are not eight separate problems needing eight separate bills. They are one structural pattern appearing eight times, and one pattern can be designed against, once we stop calling it a market and start calling it what it is.
The evidence has been in for decades. The open question was never what works. It’s whether we can build the political will to act on what works faster than the people profiting from the current design can prevent it.