This post is the first in a series considering costs in healthcare. I plan to reflect on the roles that insurance companies, Medicare, healthcare providers, and any other factor that comes to mind play. This first post in the series will be about the role of healthcare insurance companies in increasing the cost of healthcare. All of my figures are drawn from the National Association of Insurance Commissioners.
To my eye the remarkable thing about the insurance component of healthcare costs is how consistent they’ve been since 2017. That’s true whether you’re looking at healthcare insurance company profit margins (net income divided by net earned premium):
| Year | Profit Margin (%) |
| 2010 | 3.9 |
| 2011 | 3.4 |
| 2012 | 2.7 |
| 2013 | 2.2 |
| 2014 | 1.1 |
| 2015 | 0.6 |
| 2016 | 1.1 |
| 2017 | 2.4 |
| 2018 | 3.2 |
| 2019 | 3.0 |
| 2020 | 3.8 |
| 2021 | 2.0 |
| 2022 | 2.4 |
| 2023 | 2.2 |
| 2024 | 0.8 |
or as a percentage of total healthcare spending:
| Year | Insurance Costs (% of Total Health Spending) |
| 2010 | 6.4 |
| 2011 | 6.5 |
| 2012 | 6.7 |
| 2013 | 6.8 |
| 2014 | 6.7 |
| 2015 | 6.8 |
| 2016 | 6.9 |
| 2017 | 6.7 |
| 2018 | 6.6 |
| 2019 | 6.8 |
| 2020 | 6.9 |
| 2021 | 6.7 |
| 2022 | 6.6 |
Profit margins soared during the reduction in utilization that took place but they have averaged around 3%. As a percentage of GDP insurance costs have varied from 6.4% to 6.9%. That’s so consistent my conclusion is that they are artifacts.
My tentative conclusion from that is that healthcare insurance costs are not driving cost increases; they’re following them.
Please don’t construe this as support for healthcare insurance companies. I don’t think that healthcare insurance should exist, at least not in its present form. Too many things presently covered are not insurable risks.
It’s largely because medicine (and education) requires educated professionals. The world is awash in things, manufactured goods. It therefore takes a lot of things to compensate educated professionals. Used to be you might look forward to getting a toaster for opening a bank account. Nowadays, the landfills are full of discarded things.
Meanwhile, Trump apparently thinks highly educated Americans will do piecework in agricultural fields.
The US spends about double the OECD average on a per-capita PPP basis. The fact that educated professionals are required to provide healthcare services doesn’t explain the spending discrepancy. The biggest factors are (at least the last time I researched this) higher costs driven mainly by provider & hospital compensation and prescription drug costs. Lowering costs will require slaughtering some sacred cows, which won’t happen, except maybe drug costs.
Thanks for working on this series. I’m looking forward to your posts. One concern I have is around fraud being built into the costs. For example, there is a massive fraud problem with the “free” ACA plans in Florida as described in the attached article. The stat that stood out to me is that 669,115 people in Miami-Dade County should be eligible for subsidies based on census data. 782,118 people have claimed incomes below the threshold. Here is the article:
https://www.city-journal.org/article/obamacare-subsidies-fraud-florida-insurers
If there are a lot of ghost accounts, the profit margin may not change, but the overall costs would increase.
The root cause is that the insurance is “free.” People get signed up with or without their knowledge and never use the insurance. The data show that 35% of enrollees never filed a claim in 2024. Many of these are phantom accounts. If it is not free, people don’t create phantom accounts.
https://paragoninstitute.org/private-health/ghostbusting-aca-fraud-millions-who-dont-use-their-health-insurance-expose-abuse-in-the-program/
The current fight over the government shutdown will directly impact this problem. Guess which party is on the side of fraud.
Andy:
I didn’t include pharmaceutical costs in my list of factors because although it’s clearly a contributing factor it can’t be causal from a mathematical standpoint. Pharmaceutical costs are 8-9%. I don’t believe it’s possible for 9% of the total to be responsible for overall costs rising. I could do a segment on it.
Right now I’m working on a chapter on Medicare and what I’m finding completely supports what I’ve written in the past. There’s no real accounting of reimbursements under ACA plans but the best guess seems to be that they amount to a pretty tiny fraction of total healthcare spending. Same argument as the one above about drug costs. That’s not to say that it’s not a problem. It’s just not the biggest problem.
Charlie Musick:
I’m skeptical but I’ll do a chapter on fraud. I suspect the problem is much more insidious than that.
Andy: The fact that educated professionals are required to provide healthcare services doesn’t explain the spending discrepancy.
It certainly doesn’t explain all the discrepancy, but it does have a significant impact on dollars per capita spent. However, that the USA has about the worst system is certainly a factor, one which distorts the economy in all sorts of ways.
Country, GDP per capita (thousands of dollars)
USA, 85
UK, 53
EU, 43
American workers are in very high demand: Americans don’t make smartphones, they design them. Making smartphones in America would be cost prohibitive. Unlike smartphones, providing medical care can’t easily be offshored. Increasing the supply of medical professionals through importation can help. Increasing the supply of other workers can provide economic support for the more skilled professionals.
That doesn’t mean there is no other way to rein in costs, but that requires a more rational system of providing benefits. The three pillars of provision are affordability, universality, and quality. You can’t have perfection in all three, but there can be reasonable tradeoffs, which could lower costs.
Of course, there is the prospect of a technological revolution.
My healthcare insurer for most of the past twenty-five years announced that it was quitting the insurance business by the end of the year (except for Medicare Advantage). The reason given:
“The organization said its provider-owned subsidiaries [Health Alliance of Illinois and FirstCarolina Care of North Carolina] have struggled to remain “financially viable” due to a changing landscape for health insurance — including medical inflation as well as increases in prescription drug costs and those experiencing chronic medical conditions.”
https://www.nprillinois.org/health-harvest/2025-02-26/health-alliance-insurance-plans-to-end-citing-financial-challenges
I have no particular insight other than it sounds like they couldn’t charge more to pass on these costs to enough willing customers. As of now, there is competition in the healthcare insurance industry on price.
The eye popper for me is the 3% net margins. Pretty paltry. As reference, (and we measure EBITDA margins) we look for 10% companies and try to drive them to 20%. Using Drews magic eyeball for capex, interest and taxes id wager the insurance companies are making 10% EBITDA margins. Not wildly rich.
You might think of researching that topic. I’ve posted on it here. Yes, making smartphones in China is cheaper. iPhones retail at $799 to $1,199. Making them here would cost less then $100 more. Cost isn’t the reason they aren’t built here. Access to the Chinese market is the reason they’re built in China. If U. S. manufacture of the iPhone were a requirement to sell them here, Apple would make them here. They’d retail for a little more, Apple would make less profit on each phone, but it wouldn’t reduce sales much if at all. Cost-conscious Americans already buy something other than Apple. That’s been true for 40 years.
“I didn’t include pharmaceutical costs in my list of factors because although it’s clearly a contributing factor it can’t be causal from a mathematical standpoint. Pharmaceutical costs are 8-9%. I don’t believe it’s possible for 9% of the total to be responsible for overall costs rising. I could do a segment on it.”
My view is that the high costs are a combination of things and drug prices are a non-trivial part of that.
Zachriel –
I agree with a lot of that, which is why I think real cost reduction will be difficult. Reducing costs is going to impact someone’s income.
And generally, the incentives in the system promote higher costs. We have a public that wants unlimited, timely, quality health care that someone else pays for, and then to try to provide that, we have a complicated system where providers and the people who pay them agree on the costs.
I’ve told a story before about my daughter who got what turned out to be a mild concussion while snowboarding. We went to the closest clinic, signed in, signed all the usual paperwork, and then a nurse or someone came out and told us there wasn’t much they could do since they didn’t have the righ diagnostic equipment and told us to go to the ER instead. They never looked at my daughter or anything – didn’t even take BP or temperature. I found out that they charged well over $1000 (I forget the exact figure, it’s been several years) to our insurance (Tricare), plus an almost $100 copay to us. That much $$ for zero service except to be told to go somewhere else. So I submitted a fraud report with Tricare and they came back a couple weeks later and said it was legitimate.
I have similar stories from various surgeries I’ve had or family has had when, long after it’s done, I see the bill and what was charged and what was paid. Often some really crazy shit in there. My sister with dementia, now deceased, broke her hip a few years ago and her insurance and Medicare were charged many thousands for various occupational and other types of rehab visits in the hospital which were never performed because she was unable to perform them (dementia). When I complained and brought this up, I was again told there was no problem, that was all normal.
I don’t think most people really understand how ugly the sausage making is in the system and how stuff that looks like fraud is considered completely legitimate.
The City journal article, basically a Manhattan Institute guy, is very much a partisan publication with a POV. As such their numbers could be correct but I would really like to see other numbers, especially how much did Medicaid spending increase during that period. At least on a national basis we know that Medicaid spending increased at a slower rate than that of private insurers. (See link)
At any rate fraud is generally more of a problem with public insurance though some of that is just based on how it’s defined. If the paperwork is not done correctly then it’s counted as fraud. So when looked at recently all of the long term care for veterans in the VA system was called fraudulent. So its possible no one in the VA system needs long term care and its all fraud. Or maybe the paperwork is messed up.
To Dave’s point, if you follow that chart back to earlier than 2010 you will find that it steadily increases, faster than inflation and faster than Medicare and Medicaid (at least on a per person basis). So what would say in that case? Doesnt matter actually. The insurance companies are still the ones that determine how much is paid and what they pay. There are cases when an individual provider has enough market power so they can demand higher rates. Someplace like a cancer center that offers unique treatments or has a national rep. For the rest of the world they just have to accept the rates of pay they are offered.
On the global topic, many people have written on why health care is so expensive. It mostly comes down to lots of individual reasons that add up to a large whole. Drug costs are part of it. Doc pay, but also nurses, techs and therapists. Medical devices, lab tests, X-rays, ultrasounds, etc. Admin costs (including billing costs) , profits (it’s not just the insurers making profits). It’s pretty easy to track where our money goes vs other countries and in nearly every case where we spend money our costs are higher. There are odd exceptions like where generic drugs cost less in the US than other places but those are exceptions.
Also, I think you too readily accept the 3% profit number touted by the insurance industry. (You do realize your quoted source is funded by that industry?) How do companies only making 3% profits manage to advertise so much? AFAICT that 3% number is based solely upon the amount of premiums paid vs what they claim as profits after paying covered bills. It does not include money made on the investments they hold on the money in their pool. As CHAT-GPT notes…
“Investment income: This is the earnings from investing the money that remains from premiums after paying claims and expenses.
The profit generated from these investments (interest, dividends, capital gains) is added to the underwriting profit.
Total profit: The total profit is the sum of the underwriting profit and the investment income.
A note on combined ratio
The combined ratio is a common metric used to evaluate an insurer’s underwriting profitability.
However, this ratio does not include investment income and therefore may not reflect the company’s true overall profitability.”
And it gets worse. For other companies you generally calculate profit using the ratio of what a company spends on its product vs what it takes in as total revenue. Insurance companies are different They dont actually spend that much of their own money to collect premiums and then pay out claims. In fact we know that due to medical loss ratios being limited to 15% that’s the most they can spend of their own money. So we are really looking at 3%/12% ie 25%, before you add in the investment income.
https://www.healthsystemtracker.org/chart-collection/u-s-spending-healthcare-changed-time/#Total%20national%20health%20expenditures,%20by%20source%20of%20funds,%201970%20and%202023
Steve
Andy: Often some really crazy shit in there.
Of course, you’re going to see some “crazy shit”, regardless of the system, so we should be careful with anecdotal information. (That doesn’t mean to ignore your experience.)
Andy: We have a public that wants unlimited, timely, quality health care that someone else pays for, and then to try to provide that, we have a complicated system where providers and the people who pay them agree on the costs.
The way to align the two factors; higher median income with systemic problems; is to look at percentage of GDP spent on medical care, where the US is much higher than countries in Europe. If pay were the only issue, then the percent of GDP spent on medical care should align between countries.
On the former, we do expect medicine in the US to cost more simply due to the higher overall pay. For instance, first-year resident doctors get paid about the median income in the US as well as in Europe, but the median income is higher in the US. On the other hand, specialist doctors in the US get paid much more proportionally than specialists in Europe. For instance, US surgeons get paid about 5 times the US median income, while UK surgeons get paid just 2½ times the UK median income. Are US surgeons really worth twice UK surgeons? One more point: Because Americans are generally richer, it also means many can afford and want more medical care, a demand which will also drive up costs.
But that alone doesn’t explain the difference in spending as a percentage of GDP.
The US system is very ad hoc, and this has distorting effects on the economy. For instance, for historical reasons medical insurance is tied to employment. But that makes changing jobs more difficult, creating friction in the labor pool. Furthermore, lack of universal coverage results in many people not getting proper medical care until a condition becomes urgent. Obamacare addressed some of these problems, but introduced a slew of new problems.
Andy: I don’t think most people really understand how ugly the sausage making is in the system and how stuff that looks like fraud is considered completely legitimate.
Sure. And some pork intestine (and other parts best left unmentioned) is legitimately ground into sausage, such as ‘enhanced’ payments to rural hospitals to help keep them open. But while legitimate, it shows the inherently kludgy nature of the US system. A simpler system would be easier to navigate and easier to control costs and fraud. But that would require significant changes that would be resisted by those for whom “Reducing costs is going to impact someone’s income.”