You feel something. A lump, a cough that won’t quit, a pain that’s still there after two weeks. Before you do anything else, you do the math: what’s my deductible, have I hit it yet, is this probably nothing, and what happens to my finances if it isn’t. Most Americans have run this calculation. It isn’t really a medical question. It’s an actuarial one, and you’re the one doing it, alone, with no training, because the alternative is finding out the hard way what your plan actually covers.
Now picture the other version of that same morning. Something feels off. You call your doctor’s cell phone, because you have that number, and you’re seen that afternoon. Nobody asks what your deductible is. Nobody asks if it’s covered. The only question is whether it’s serious, and if it is, you get referred to whoever is best, not whoever happens to be in network.
That’s the two-tiered healthcare system this country actually runs, even though almost nobody says it out loud. One tier fights for care and hopes the bill doesn’t ruin them. The other tier just gets care. I don’t think that’s a coincidence or a problem waiting on the right insurance regulation to fix it. I think it’s what happens when you build a system that treats staying alive as a product to be sold instead of something a decent country owes the people in it.
The System You’re In: Prove It, Then Pay For It Anyway
If you have insurance, here’s what “covered” actually means in practice. A 2025 KFF poll found that 62% of insured adults who needed specialized care, things like surgery, imaging, a specialist referral, or a specialty drug, were required to get prior authorization first: your doctor has to ask your insurance company’s permission before treating you. More than half of that group said their care was then delayed or denied outright, roughly split between the two. Seventy-three percent of the public calls this a major problem, and that opinion crosses party lines: 57% of Republicans, 79% of independents, 84% of Democrats. That’s about as close to consensus as anything gets these days, and what everyone agrees on is that insurance companies are standing between patients and their doctors.
Ask actual physicians and the picture gets darker. In an American Medical Association survey, 93% of doctors said prior authorization delays care, and 82% said it causes patients to abandon treatment altogether. Twenty-nine percent said prior authorization has led to a serious adverse event for one of their patients, including hospitalization, permanent damage, or death. Nearly a third of physicians said the criteria insurers use to approve or deny claims are rarely or never evidence-based. And six in ten doctors are worried AI is making it worse: insurers increasingly run these decisions through automated systems that can batch-deny claims with little or no human review before anyone with an actual medical degree looks at the file. Medicare Advantage insurers alone made nearly 50 million prior authorization determinations in 2023.
This is the part I keep coming back to. It’s not that insurance doesn’t exist for most Americans. It’s that having it doesn’t mean what you think it means. You can pay a premium every month for years and still have a claims reviewer who has never examined you, and increasingly isn’t even a person, decide your MRI wasn’t medically necessary.
Insurance Doesn’t Mean You’re Covered
Even when a claim gets approved, plenty of Americans still can’t afford what’s left over. In 2024, more than 31 million Americans, most of whom had some form of insurance, borrowed a combined $74 billion just to cover healthcare costs. Depending on how it’s measured, somewhere between roughly 20 million and over 100 million American adults are currently carrying medical debt or struggling with a medical bill. This isn’t mainly a story about the uninsured. Most people with medical debt are employed, insured, and wouldn’t describe themselves as poor.
Ask people who’ve actually been through something serious. A 2025 survey found 51% of cancer patients and survivors reported medical debt from their treatment, and nearly all of them had insurance going in. Having a plan didn’t spare them. It just meant they found out later, after the diagnosis, what their plan’s limits actually were.
The predictable result is that people delay care they need. Recent surveys put the share of American adults who’ve postponed necessary treatment because of cost at around 36%, including more than a third of people with employer-sponsored insurance, the “good” kind everyone’s supposedly grateful for. Compare that to the European Union, where a combined 3.6% of adults report unmet medical needs due to cost, distance, or wait times put together. We don’t wait in line over there. We just don’t go, and we tell ourselves it was probably nothing, because we couldn’t afford to find out otherwise.
Then there’s the in-network question, which turns every referral into a research project. Before 2022, about 1 in 5 emergency room visits and up to 16% of in-network hospitalizations resulted in a surprise bill from an out-of-network anesthesiologist, radiologist, or specialist the patient never chose and often never met. Congress passed the No Surprises Act specifically because this had become common enough to need a federal law, and it has genuinely helped: insurers reported it was preventing roughly a million surprise bills a month within its first year. It didn’t make the underlying absurdity go away. Basic medical care in this country still requires you to personally audit a hospital’s staffing contracts, or hope you never have to.
Right Now, Losing Coverage Is Getting Easier
This is getting worse this year, not better. The enhanced ACA premium tax credits that had held marketplace premiums down since 2021 expired at the end of 2025. KFF found that premiums for the average marketplace enrollee more than doubled in 2026, a 114% jump. By March 2026, 9% of people who’d had ACA coverage the year before were already uninsured, and 55% of the people who managed to re-enroll said they’d started cutting spending on food and other basics just to keep paying for health insurance. The Urban Institute projects 4.8 million people will lose coverage entirely because of this.
I live in Texas, which already has the highest uninsured rate in the country, 16.7% overall and 21.6% among working-age adults, largely because it’s one of ten states that never expanded Medicaid. Texas alone accounts for over 40% of the national “coverage gap,” people who earn too much for Medicaid but too little to qualify for marketplace subsidies. That gap isn’t an accident. It’s a policy choice, renewed every year the legislature doesn’t act.
And it isn’t just about premiums. Between 2023 and 2024, states “unwound” pandemic-era Medicaid protections and reviewed everyone’s eligibility again. Over 25 million people were disenrolled, and 69% of them lost coverage over paperwork problems, not because they’d actually become ineligible. It’s the same story as the safety-net paperwork I’ve written about on this site before. The maze isn’t a side effect of the system. It’s the point.
The Other System: Cost Is Never the Question
Now here’s what healthcare looks like when money genuinely isn’t a constraint.
Concierge and direct primary care practices, where patients pay a membership fee on top of or instead of insurance, grew 83% between 2018 and 2023. A typical primary care doctor manages a panel of 2,000 or more patients. Concierge doctors typically manage a few hundred. Some ultra-premium practices, concentrated in places like Manhattan, Beverly Hills, and Silicon Valley, keep panels under 500 and charge retainers into the tens of thousands of dollars a year, on top of whatever insurance still covers. One doctor who made the switch went from managing roughly 4,500 patients to 450, with exam times stretching to 90 minutes instead of the standard rushed visit. No prior authorization fights. No network to check. If something feels minor, you get it looked at anyway, because why wouldn’t you. If something feels major, you get referred to whoever is actually the best, not whoever your plan happens to cover.
Congress quietly made this easier last year, too. The same 2025 reconciliation bill that added new work-verification paperwork to Medicaid also let people with high-deductible plans pay for concierge and direct primary care memberships under $150 a month using their tax-advantaged health savings accounts. So one side of the ledger got a new bureaucracy to prove it deserves basic coverage, and the other side got a tax break that formalizes paying your way out of that same bureaucracy entirely. That’s not a coincidence of timing. That’s what it looks like when a system is built around who can pay.
To be fair, not all direct primary care is a luxury good. Some lower-cost versions genuinely serve middle-income patients for a modest monthly fee. But the fastest-growing, most exclusive layer of this market is unmistakably a wealth product, and it’s growing precisely because the standard system has gotten this hard to deal with.
We Pay the Most and Get the Least
Zoom out to the country level and the same pattern holds. The United States spent roughly $14,800 per person on healthcare in 2024, about $5,000 more per capita than the next-highest spender, Switzerland, and nearly $7,000 more than the average of other wealthy nations. Despite that, the US has the lowest life expectancy of any peer country. Americans don’t even use more healthcare than people in other wealthy nations. Utilization rates here are actually lower. The extra money isn’t buying more care. It’s buying higher prices for the same care, plus an administrative apparatus, insurance overhead, billing departments, prior authorization staff, that costs more per person here than in any comparable country.
The Commonwealth Fund’s 2026 global comparison put a number on where this is heading: the Medicaid and marketplace cuts working their way through the system now are projected to push the uninsured population up by 17 million people by 2034 and could lead to more than 50,000 additional preventable deaths a year. That’s not a rhetorical flourish. That’s a mainstream health policy research organization’s projection, based on what happens when people lose coverage and stop getting care early enough for it to matter.
We are, by a wide margin, the most expensive healthcare system in the developed world, and simultaneously the one most likely to hand you a form instead of a diagnosis.
Why Healthcare Doesn’t Fit in the Capitalism Box
Here’s the argument I actually want to make, underneath all the statistics: healthcare doesn’t behave like a normal market, and no amount of tweaking around the edges is going to make it act like one.
Markets work reasonably well when a buyer can compare options, walk away from a bad deal, and choose when to buy. None of that holds for medical care. The Nobel-winning economist Kenneth Arrow made this case back in 1963, in a paper that’s still the starting point for how economists think about healthcare: demand for medical care isn’t planned the way demand for a car or a television is. It’s triggered by illness, which is unpredictable by definition. You don’t decide you want a triple bypass the way you decide you want a new phone. It’s worth saying plainly that Arrow’s paper is more careful and more contested than the simplified version people usually invoke online, and he didn’t hand down a verdict that markets always fail and government must always win. But the core observation holds up: healthcare has features, uncertainty, urgency, and a massive information gap between the person providing treatment and the person receiving it, that ordinary consumer markets don’t have to deal with.
You can’t comparison shop during a heart attack. You can’t get a second opinion while unconscious. You usually can’t even tell whether the test your doctor ordered is the right one, because if you had that knowledge, you wouldn’t need the doctor in the first place. That’s not a flaw in how any particular company runs its business. It’s baked into what medical care actually is. A grocery store that overcharges you loses you as a customer next time. A hospital that overcharges you while you’re in an ambulance still gets paid, because you never had a choice to begin with.
That’s what I mean when I say healthcare doesn’t fit in the capitalism box. Markets allocate goods by willingness and ability to pay, and they treat the buyer’s freedom to walk away as the thing that keeps sellers honest. Take away the ability to walk away, which is exactly what illness and emergency do, and you’ve taken away the mechanism that makes a market a market. What’s left isn’t competition. It’s leverage, and the hospital, the insurer, and the drug manufacturer all have more of it than the person lying in the bed.
Taking the Other Side Seriously
I want to take the strongest version of the other argument seriously, because it isn’t a stupid one.
Critics of single-payer or universal systems point to real wait times in countries that ration care openly instead of privately. Sweden and Spain report median waits of two to three months for some procedures, and some countries do worse. That’s a genuine trade-off, not a myth, and anyone promising a universal system with zero downsides is selling something.
But the “at least American care is fast” defense doesn’t hold up, because we have real wait times too, we just don’t count them the same way. AMN Healthcare’s 2025 survey of new-patient appointments found the average wait to see a doctor across six common specialties in 15 major US cities is now 31 days, the longest it’s been since the survey started tracking this in 2004, up 19% in just three years. Family medicine, the specialty closest to a routine physical, averages 23.5 days, and that’s for a brand-new patient calling cold. I can vouch for this personally: I tried this month to reschedule an annual checkup that was already on the books from my last visit, set for mid-October. The soonest alternative slot was mid-November, four months out, for an existing patient moving an already-scheduled routine visit, not someone shopping for a first appointment. The scheduler told me that waiting even a few more weeks to make that call would likely have pushed it into 2027.
So the honest comparison isn’t fast American care against slow rationed care elsewhere. It’s slow American care, that you also pay the most for in the world, that you might also get denied through prior authorization on top of, against slower care elsewhere that’s at least free at the point of service. We didn’t dodge the wait other countries deal with openly. We added a price tag and a paperwork maze on top of one of our own.
There’s also a fair concern about transition cost and disruption. Moving over 160 million people off employer coverage is not a small lift, and it deserves more respect than “just do Medicare for All” tends to give it as a slogan. I don’t think that complexity is a reason to keep the system we have. I think it’s a reason to plan the transition seriously instead of pretending it’s simple. Complexity isn’t the same argument as impossibility, and I’m not aware of another wealthy democracy that made this kind of transition and now wants its old system back.
What I Think Healthcare Should Be
I don’t think this is really a debate about efficiency, or deductibles, or which prior authorization vendor is the least terrible. I think it’s a debate about whether a person’s access to care should depend on what they can afford to lose.
Right now it does. If you have money, healthcare is a service you purchase, calmly, on your own schedule, with your comfort built into the design. If you don’t, it’s a fight you have to win over and over: against a deductible, against a claims algorithm, against a network map, against a bill that shows up months after you thought the whole thing was handled. Both of those systems exist in the same country, often inside the same hospital, sometimes down the same hallway.
A market can ration bread. Nobody dies if they can’t afford a second loaf. Healthcare isn’t bread, and pretending it is hasn’t made it behave like bread. It’s just widened the gap between the two tiers every year I’ve been alive. I don’t think this fixes itself. I think it gets fixed when enough people say plainly that a system where your zip code, your job, and your bank balance decide whether you get treated isn’t healthcare. It’s a filter, and it’s been sorting us by money the whole time we called it something else.
Frequently Asked Questions
Why do health insurance companies require prior authorization? Insurers say prior authorization controls costs and prevents unnecessary care. In practice, KFF polling found 73% of the public and large majorities of physicians consider it a major barrier to timely treatment, and AMA survey data shows most doctors believe it delays care and sometimes causes serious harm.
What happened to ACA health insurance subsidies in 2026? Enhanced premium tax credits that had lowered ACA marketplace premiums since 2021 expired at the end of 2025. Average marketplace premiums more than doubled in 2026, and millions of enrollees are expected to lose coverage as a result.
How does US healthcare spending compare to other countries? The US spends roughly $14,800 per person on healthcare, far more than any other wealthy nation, yet has the lowest life expectancy among peer countries and lower healthcare utilization rates than many of them.
What does it mean that healthcare doesn’t fit a free market model? Free markets rely on buyers being able to compare options, walk away, and choose when to buy. Medical demand is triggered by unpredictable illness, and patients often lack the expertise to evaluate the care they’re being sold, which breaks the assumptions that let markets function efficiently.
Is concierge medicine only for wealthy patients? Some direct primary care models serve middle-income patients for a modest monthly fee. But the fastest-growing, most exclusive tier of concierge medicine, with retainers running into the tens of thousands of dollars a year, caters specifically to patients who can pay to skip the waits and denials everyone else deals with.