The Health Insurance Death Spiral Is Here
ACA premiums are spiraling toward $5,600/month for a family of four while CrowdHealth members saw prices drop 6–8% this year. Marty Bent and Andy Schoonover on the principal-agent rot inside the system and why the exit already exists.

↓ Jump to the video and timestamps
I've been texting Andy Schoonover for months telling him I think this is the year. Not the year CrowdHealth becomes interesting. The year it becomes obvious. Open enrollment notices are landing in inboxes right now and people are genuinely shocked by what they're seeing. Videos of people reading their renewal letters in their cars, barely able to get the words out.
That's not a fringe phenomenon. That's the system breaking in real time.
I've been a CrowdHealth member for four years. My family of five is on it. We paid for two births through the platform. When I left Great American Mining to go all-in on TFTC and Ten31, I hit COBRA at $1,800 a month, did it for two months, and got off as fast as I could. That was the forcing function. I haven't looked back.
And now, watching ACA premiums spiral toward $5,600 a month for a family of four while CrowdHealth prices actually dropped 6–8% this year, the opportunity cost of staying in the traditional system has never been lower.
Andy came back on the show to walk through exactly why the death spiral is real, why the system was designed this way, and why the exit already exists. This is the healthcare equivalent of the fiat money breakdown. Same principal-agent rot. Same administrative bloat extracting value from productive people.
Same moment where the alternative goes from fringe to obviously correct.
The receipts, if you want them before you decide: we compiled every month CrowdHealth has published into one place — what the community actually funded, how long it took, and what got refused.
Key takeaways
- The ACA premium spiral is not a bug. The law's Medical Loss Ratio rule caps insurer profit at roughly 15% of revenue, which means the only way to grow profit in dollar terms is to grow the premium base. The buyer and the seller of healthcare both have an incentive for prices to go up. That's the designed outcome.
- Healthy people are doing the math and walking. $24,000 in annual premiums plus a $10,000 deductible means $34,000 out of pocket before insurance covers a cent. For a healthy family, the probability of needing that much care in a year is less than 1%. The exit is the rational move.
- 80% of hospital bills contain errors or fraudulent upcoding, according to CrowdHealth's review of 25,000+ member bills. A $50,000 bill that should be $12,000–$15,000 at 125–150% of Medicare rates is not exceptional. It's the norm. Most people have no idea and no ability to fight it.
- CrowdHealth prices dropped 6–8% this year while ACA premiums rose roughly 18%. Those two lines are going in opposite directions. Scale makes CrowdHealth more deflationary, not less.
- The best doctors are opting out. Cash-only and direct primary care practices are growing because physicians are fed up with insurers dictating care. Dr. Mary Talley Bowden just went DPC. Doctors and dentists are among the earliest small-business Bitcoin adopters for a reason.
- This is the same move as Bitcoin. Opting out of traditional health insurance and opting out of fiat are structurally the same decision. You exit a corrupt principal-agent system and rebuild the relationship directly: patient to doctor, person to money.
The death spiral in real numbers
The highest premium Andy's team has seen so far: $5,600 a month for a family of four. That's not a typo.
I pulled up a tweet on the show from a sole-proprietor trucker, the only employee of his own business, staring at a renewal notice for himself and five dependents: roughly $3,300 a month. His business eats the premium, which means $2,600 is coming out of his actual paycheck on top of that. Andy's response was immediate: that family on CrowdHealth would be about $750 a month.
There was another screenshot on my screen, Blue Cross Blue Shield, $3,500 a month, deductibles ranging from $8,600 to $12,000. Andy did the math out loud. You're paying $24,000 a year in premiums plus a $10,000 deductible. That's $34,000 before insurance pays a single dollar.
For a healthy family, the probability of hitting that number in any given year is somewhere around half a percent.
The ACA premiums are rising by Andy's estimate at roughly 18% while CrowdHealth's prices dropped 6–8% this year. Those two lines are moving in opposite directions and the gap is accelerating.
What's driving the ACA spike isn't just medical cost inflation. Andy's explanation of the mechanics is the important part. The enhanced subsidies from the American Rescue Plan that masked premium increases for roughly the past four years are expiring. The actuaries at these insurance plans are now assuming healthy people will exit, which means the remaining pool skews sicker, which means rates have to climb to cover a population that actually uses care.
That assumption becomes self-fulfilling. Healthy people see the new number, do the math, and leave. Which makes the pool sicker. Which drives rates higher. That's the death spiral.
Why the ACA made everything worse
I pulled up UnitedHealthcare's max stock chart on tape. Pre-ACA, it was trading below $100. At its peak, it was near $300. Obviously it's down badly this year, but the chart tells the story.
The Affordable Care Act was supposed to keep insurers from gouging customers. What it actually did was hand insurers a federally structured profit guarantee and a captive market.
The mechanism Andy walked through is the Medical Loss Ratio rule. Insurers can only keep roughly 15% of premium revenue as profit. That sounds like a ceiling. In practice, it functions as a floor with an escalator. If you're capped at 15% of revenue and you want to grow your profits from $150 to $165 per member, you raise the premium from $1,000 to $1,100. The insurer benefits when premiums go up. The hospital benefits when prices go up.
So you have the buyer and the seller of healthcare both wanting the price to rise, with a regulatory structure that ensures they get it. You don't need a PhD in economics to see where that ends.
The Medicare Advantage piece is arguably worse. What sounds like privatization is actually a government-funded program administered by for-profit companies, which means you get the worst of both worlds. UnitedHealthcare is alleged to have coded members as diabetic at five to six times the national average to receive higher government payments for sicker patients.
Andy says "allegedly" and I'll maintain that qualifier here because it involves an active DOJ investigation. But the dynamic he's describing is the same one running through the whole system: every incentive points toward gaming the reimbursement, and the people who are supposed to be your agent are actually working against you.
Marty: "UnitedHealthcare has all the incentive to create that moat, that regulatory moat."
The $5,000 bag of salt water
Andy was at Bitcoin Commons about a year and a half ago. Thought he was having a stroke. Parker Lewis was right outside the office. Will Cole or Parker called 911. Andy went to the hospital, they put him on an IV he didn't need, and he sat on it for 12 hours.
One thousand milliliters of saline solution. You can buy that on Amazon for around $8. The hospital billed him $5,000.
That's not a rounding error. That's the system working as designed.
Based on CrowdHealth's review of more than 25,000 member bills, Andy puts the error or fraudulent upcoding rate at 80%. The other 10%, he says, you can still make a very good argument on. ER billing codes run from one to five, with five being "critical or life-threatening." By CrowdHealth's internal data, more than 80% of the ER bills they see come in coded at level four or five.
Andy has seen a broken finger coded as critical. The hospitals run those codes because they're maximizing what they can extract from insurers. And the insurers don't care enough to push back hard, because higher bills mean higher premiums, which means higher profits for them too.
CrowdHealth's AI negotiation platform is built on knowing what a given hospital has actually charged in the past and what they've actually settled for. Targeting 125–150% of Medicare as a fair price benchmark, they can take a $50,000 bill to something genuinely manageable. Most people going in alone get nowhere near that because they don't know the number they should be fighting toward.
I mentioned on tape that someone fed a hospital bill to Claude and got it reduced by around 85% just by having the AI identify double and triple charges. Andy said CrowdHealth does the same thing with its own HIPAA-compliant platform, but pairs it with the historical data on that specific hospital's billing and negotiation behavior. The combination is where you get the real number.
The exit already exists
The objection I hear most from people who are curious about CrowdHealth is some version of: "What happens when something really bad happens?"
Here's what happened last summer in Montana. A guy was fishing with a .44 revolver in his holster. He caught a fish, leaned down, the gun fell out and hit a rock. The gun discharged. The bullet went into his calf, out the back of his calf, into his thigh, out the top of his thigh, into his chest and out the back. Destroyed his gallbladder. Destroyed his intestines. His wife tied her shirt around his thigh because he thought he'd hit the femoral artery.
She ran to the top of a hill, called 911, and a helicopter pulled him out of the stream. He went into a coma at the hospital for two to three weeks. He had to drive home from Montana to California because they wouldn't clear him to fly. He got sepsis on the way home. The bill came in at almost a million dollars. The CrowdHealth community funded it without a problem.
Andy also mentioned a member currently at MD Anderson dealing with a serious cancer diagnosis. They're being covered.
CrowdHealth has now processed more than 25,000 bills. They're at 15,000 members and Andy is projecting somewhere between 20,000 and 25,000 by year end. That's a big enough pool to fund the catastrophic events. The community mechanics work because the model is a flat subscription, not a premium-based structure.
CrowdHealth's revenue doesn't go up when your bills go up. They have every incentive to negotiate your bill down and get it paid. That alignment is something the insurance model is structurally incapable of offering.
My family is at roughly $695 a month for CrowdHealth plus direct primary care, all-in around $900–$950 a month for a family of five. Andy pointed out that you can submit up to $300 of your DPC cost per month to the crowd and it's fully eligible for funding. We go above and beyond with the DPC because we like having it, not because we have to.
And every month, the difference between CrowdHealth's maximum ask and what they actually ask gets automatically stacked into your Fold account in bitcoin. Last month, for an individual member, the max ask was $140, the actual ask was $85, and the $55 difference went straight into bitcoin.
I have the Fold integration turned on. I'll open the app to check my debit card rewards balance and notice my other bitcoin balance ticked up. CrowdHealth has been passively stacking sats for me for years.
Opting out is the same move as Bitcoin
Andy said it directly and I agree with every word: the bitcoiners are opting out of the fiat system, and everyone else is starting to opt out of the healthcare system. It's the same move, made for the same reason.
The social signal when you tell a doctor's office you're uninsured is identical to what it felt like to tell someone you use Bitcoin five years ago. They look at you like you're either poor or crazy. Then the price drops by 50% on the spot.
The front desk demeanor changes completely. It's fish-and-water stuff: they are so deep inside the insurance paradigm that hearing a cash patient is genuinely disorienting.
Andy has an "UNINSURED" T-shirt. He wore it through DFW and got the same looks you'd have gotten in a Bitcoin shirt in 2019. I think that trajectory plays out the same way.
What I find most compelling is where this goes from here. Andy is in conversations with major hospital systems about accepting Bitcoin. He won't name them, and neither will I, but the interest is real. CrowdHealth is building a database of Bitcoin-accepting doctors.
The circularity he's describing, bitcoiners paying doctors in Bitcoin, doctors who want hard money accepting it, the whole relationship disintermediated from the insurer and the hospital system billing department, that's not a fantasy. It's early and building.
The healthcare spend in the U.S. is $5.3 trillion a year, according to CMS figures. Making even a small dent in that with Bitcoin payments would be significant. The doctors and dentists who are moving to cash-only practices are finding Bitcoin naturally.
I know this from conversations with Parker Lewis and the team at Zaprite, who see doctors and dentists among the earliest small-business Bitcoin adopters. I was paying my primary care doctor in Austin in Bitcoin. The pipeline is real.
Andy made one other point I keep coming back to. He estimates something like 3–4 million people would go entrepreneurial if not for employer-based health insurance shackling them to their jobs. That's his estimate, not a published figure, but I believe the shape of it.
If you asked people who thought about leaving their job but didn't for the reason, health insurance would be the number one answer or top three. My COBRA bill was $1,800 a month. I did it for two months, then switched. That was the moment I went all-in on TFTC and Ten31. The health insurance cage is real and a lot of people are stuck in it.
If you work for a small employer, Andy's suggestion is worth taking seriously. Your employer is likely paying around $25,000 a year to cover your family under a traditional plan. CrowdHealth for the same family runs roughly $7,000.
Go ask your employer to cut the group plan, give you a $10,000–$12,000 salary bump, and let you source your own coverage. Both parties save money. The math works and plenty of CrowdHealth members have already done this.
What to do right now
The code is TFTC at joincrowdhealth.com/tftc. You get $99 a month for the first three months. Try it.
If you don't like it after three months, go buy fiat health insurance. But I'd bet you won't.
The referral program is also worth knowing about: $250 paid in Bitcoin for every member you bring in. Andy says dozens of members are now making more in referral fees per month than they pay for their membership. You can have healthcare and make money on it. That's not a bad position.
The opportunity cost of staying in the traditional system has never been lower than it is right now. With ACA premiums at $3,500–$5,600 a month for a family, the math has already made the decision for most healthy people. The question is just whether they know the exit exists.
It does. We've been using it for four years.
About Andy Schoonover
Andy Schoonover is the founder and CEO of CrowdHealth, a community-funded alternative to health insurance he started building in 2021. Before CrowdHealth, he had extensive experience in healthcare and business operations. He is based in Austin, Texas, and is a Bitcoin holder who describes roughly 80% of his liquid assets as held in Bitcoin. He has appeared on TFTC multiple times to discuss the ACA premium spiral, hospital billing fraud, and CrowdHealth's model for disintermediating the insurance industry from the doctor-patient relationship.
Sources mentioned
- CMS National Health Expenditure Accounts: official US health spending totals: $5.3 trillion in 2024, 18% of GDP
- KFF Health Benefits Survey: annual data on employer health insurance costs; context for small-employer premium figures
- KFF ACA Marketplace Premium Data: premium trend and subsidy threshold data Andy references
- CrowdHealth on Trustpilot: member reviews Andy cites on tape; verify current counts before treating specific numbers as settled
- Fold App: the Bitcoin rewards integration Andy describes for the automatic sats-stacking mechanic
- KFF: how much and why ACA Marketplace premiums are rising in 2026: confirms the 18% median proposed increase for 2026 and attributes it partly to expiring enhanced premium tax credits
- CMS: Medical Loss Ratio: the rule requiring insurers to spend at least 80-85% of premium dollars on care, which is the structural argument in this episode
- CMS Medicare fee schedules: the Medicare rate benchmark CrowdHealth prices against
- MedPAC report to Congress on Medicare payment policy: the independent commission's analysis of Medicare Advantage payment and coding intensity
- HHS Office of Inspector General reports: federal audits of improper billing and upcoding across Medicare and Medicaid
- Peterson-KFF Health System Tracker: comparative data on what private insurers pay relative to Medicare
- HealthCare.gov: COBRA continuation coverage: why COBRA costs what it does once the employer subsidy disappears
- Health Affairs: the unintended consequences of the ACA's Medical Loss Ratio requirement: peer-reviewed analysis arguing the MLR cap turns insurers into cost-plus businesses with no reason to hold premiums down, which is exactly the mechanism Andy describes
- 45 CFR 158.210: minimum medical loss ratio: the regulation itself, setting the 85% floor in the large group market and 80% in the individual and small group markets
- KFF: how much more than Medicare do private insurers pay?: a review of 19 studies on the gap between private and Medicare payment rates
- HHS OIG: coding trends of Medicare evaluation and management services: the federal audit behind the ER-level argument: billing of the top-level code 99285 rose from 27% to 48% of visits in ten years while every lower level shrank
- HHS OIG: improper payments for E/M services cost Medicare billions: 55% of evaluation and management claims were incorrectly coded or undocumented, 26% of them upcoded
- CMS Comprehensive Error Rate Testing (CERT): CMS's own measurement of improper Medicare payments, $28.83 billion in FY2025
- KFF 2025 Employer Health Benefits Survey: family coverage now averages $26,993 a year, with workers contributing $6,850
- KFF: what we know about 2026 ACA Marketplace enrollment, premiums and deductibles: marketplace deductibles hit a record $3,786 for 2026 and enrollment is projected to fall by millions
- 29 U.S.C. 1162: COBRA continuation premium cap: the statute capping COBRA premiums at 102% of the full plan cost, which is why leaving a job costs what it does
Watch the conversation
Timestamps
- 0:07 - Bitcoin and fiat currencies
- 0:36 - Open enrollment and the death spiral
- 1:35 - CrowdHealth growth and ACA premium records
- 2:29 - Real renewal notices: $3,300 and $3,500/month
- 4:03 - Why the ACA failed and who exits first
- 5:36 - No government plan exists; burn it down
- 7:46 - Ozempic subsidies and the obesity cost driver
- 10:43 - Healthy people pay less; CrowdHealth incentive model
- 22:00 - Hospital billing fraud: the $5,000 saline bag
- 27:00 - The MLR rule and the UnitedHealthcare stock chart
- 36:00 - The Montana gunshot story and catastrophic coverage
- 48:00 - Bitcoin integration, Fold stacking, and hospital Bitcoin conversations
- 57:00 - Doctors opting out; Dr. Mary Talley Bowden goes DPC
- 1:04:00 - Employment cages, COBRA, and the entrepreneur opportunity
- 1:09:00 - How to switch: the TFTC code, referrals, and the employer conversation
Sponsors
- CrowdHealth: the community-funded alternative to health insurance my family has used for four years, covering two births and a family of five at roughly $695/month. Code TFTC gets you $99/month for the first three months. joincrowdhealth.com/tftc
- Bitkey: Bitcoin self-custody hardware that embeds natively into a two-of-three multisig. One key on the device, one on your phone, one with Block in the cloud. The easiest zero-to-one step for anyone who has Bitcoin on an exchange and hasn't moved it off. Code TFTC20 for 20% off. bitkey.world
- SLNT: Faraday gear for your hardware, cards, and phone. Patented technology, special operations approved, free shipping. Code TFTC for 15% off at slnt.com/tftc
- Obscura VPN: built by a bitcoiner for bitcoiners, designed so it cannot log. Pay in Bitcoin over Lightning for a fully private setup. Code TFTC for 25% off. obscura.net
- Salt of the Earth: electrolyte drink mix and creatine packets, pink lemonade, glass bottle. Code TFTC for 15% off. drinksote.com
Frequently Asked Questions
The death spiral happens when healthy people exit the insurance pool because premiums become too expensive to justify. That leaves a sicker remaining population, which forces actuaries to raise rates further, which pushes more healthy people out. The ACA masked this dynamic for several years through enhanced subsidies. As those subsidies expire, the cycle is accelerating fast, with premiums rising roughly 18% while the insured population skews increasingly toward high-cost patients.
A family on CrowdHealth pays roughly $700–$750 a month. That same family on the ACA marketplace is now looking at $3,000–$5,600 a month in premiums, plus deductibles of $8,600–$12,000 before insurance pays anything. When you add premiums plus the deductible, a family on ACA can be $34,000 out of pocket before their coverage activates. The CrowdHealth number is all-in from the start.
Yes. Andy described a member in Montana who accidentally shot himself with a .44, sustained wounds to his calf, thigh, chest, and internal organs, spent weeks in a coma, and got sepsis on the way home. The bill came in at almost a million dollars. The community funded it. CrowdHealth also currently has a member at MD Anderson dealing with a serious cancer diagnosis who is being covered. With 15,000 members and more than 25,000 bills processed, the community has demonstrated it can handle catastrophic events.
If you work for a small employer, the math often works in both of your favors. A small employer typically pays around $25,000 a year to provide family health insurance under a traditional plan. CrowdHealth for the same family runs roughly $7,000. You can approach your employer, ask them to drop the group plan, give you a $10,000–$12,000 salary increase, and let you source your own coverage through CrowdHealth. Both sides come out ahead. This won't work at every large company, but for small businesses the conversation is worth having.
Based on CrowdHealth's review of more than 25,000 member bills, Andy puts the rate of clear errors or fraudulent upcoding at 80%. Another 10% of bills have charges that a well-informed negotiator can still successfully contest. ER bills are particularly egregious: more than 80% of the ER bills CrowdHealth processes come in coded at the highest severity levels, including broken fingers classified as critical or life-threatening. These numbers reflect CrowdHealth's internal processing data, not an independent published study.
Each month, if the community is doing well and CrowdHealth asks for less than the maximum contribution, the difference is automatically sent to your linked Fold account and converted to Bitcoin. The maximum individual ask is $140 per month; last month the actual ask was $85, so $55 went straight to Fold in sats. CrowdHealth also has a growing database of Bitcoin-accepting doctors and is in early conversations with major hospital systems about Bitcoin payment acceptance. The referral fee of $250 per new member is paid in Bitcoin.


