How Much Universal Healthcare Would Cost in the US

Published November 11, 2015 by Ryan Whitacker

I recently published an article calculating the costs of all Bernie Sander’s proposals. The article spread quickly on Facebook and Reddit; the most common disagreement (by far) could be paraphrased, “Universal Healthcare wouldn’t cost anything.” Fiscal conservatives, of course, are quick to deride the idea that you can get something for nothing. There’s a lot of confusion in both conservative and liberal camps, so let’s see if we can set the record straight for both groups.

Jump to the conclusion

Government Spending Must Increase

First, let’s clarify something right off the bat: you cannot move from a system where 47% of healthcare costs are paid for by the government to a single-payer universal healthcare system without increasing government spending. The US cannot move from last place in OECD countries for public (government) spending on healthcare as a percentage of GDP without increasing the portion paid by the government.

Healthcare costs - USA vs. Switzerland vs. Netherlands

Countries don’t usually end up with 100% coverage from the government, partly because governments don’t usually pay for things like plastic surgery and other elective procedures that aren’t medically necessary. Except in countries where the government is the only provider of healthcare (rather than the only insurer) you see 80-85% of healthcare costs paid by the government. (Norway: 85%, UK: 84%, Sweden: 81%, Iceland: 80%, Switzerland: 66%)

When you look at healthcare costs from a government spending perspective there is no scenario where you increase US public spending by 59% (from 47% to 80%) for free. Some liberal supporters of UHC (universal healthcare) seem to have misunderstood what “savings” means and believe taxes (or payments to the government) wouldn’t even need to increase, and that’s simply not true.

With that ob(li)vious objection out of the way, let’s dig into the real interesting data question: Would universal healthcare save Americans money overall?

Analysis of Total Healthcare Costs

Opponents of UHC should be careful not to use the conclusion above as a straw-man argument against supporters who actually knows what they’re talking about. What most well-informed proponents of UHC mean when they say it “wouldn’t cost Americans anything” is that the private plus public costs would decrease. If the money Americans pay to private companies went to the government instead, would that be sufficient to cover universal healthcare costs?

The lazy approach would be to compare the US to other systems and subtract the difference. That would give you huge savings overall for private + public spending, but it would ignore the reality of differences in costs and potential increases in demand.

Part 1: estimating costs

US healthcare is expensive. The first big reason is that our doctors and surgeons are paid very well. The data below comes from Harvard economist David Cutler.

Average doctor earnings by country

I haven’t heard any politicians arguing that we should cut doctor’s salaries; it would be a mistake to assume the US system would mirror a system like the United Kingdom where doctors make half as much. Not surprisingly, the Netherlands (which pays doctors more) comes in second in total healthcare spending as a percentage of GDP. Switzerland is second in overall spending. Total costs and doctor pay align well in correlation with a few exceptions like Norway and Sweden.


The US far outpaces every country in both raw spending and spending as a percentage of GDP. The US is nowhere near first place for life expectancy (15th out of 34) or quality of care despite all that spending. We broke down OECD key statistics and found 3 major reasons American healthcare costs are so much higher.

  • Americans spend 160% more on drugs vs. the Netherlands, and about 36% more than the next-highest-spending countries (Japan and Canada). That’s partly because Americans want a pill for everything, but mostly because the US has generous patent protections for drug makers allowing them to charge high prices.
  • The US has a big obesity problem, ranking first in the world. 28.7% of Americans are obese versus 11.8% of people in the Netherlands. (Source above) Obesity costs extra because of weight-related costs like diabetes. Americans over 18 are about 45% more likely to have diabetes. (USA diabetes source, Netherlands diabetes source).
  • Hospital costs in the US are the highest in the world at $10,300 per stay. Hospital costs make up 16% of total healthcare costs. In the Netherlands the average hospital stay costs about $4,100. These are primarily administrative costs which include legal staff and expenses (arguably because Americans like to sue each other). US hospitals also have higher-than-average   profit margins.

Obviously we’re not going to get thinner by passing a new law, so we’ll have to keep the costs of an obese and more diabetic population. I also don’t see US politicians talking about patent reform for drugs, and the pharmacy lobby in the US is huge. It would again be a mistake to assume patent reform immediately as a part of universal healthcare.

The Netherlands is actually a good baseline to use, since private companies still exist as more regulated entities alongside the government system, and it’s not likely the US will ever shut down private companies. When we add in the costs of obesity and the drug costs listed above, we land at around $6,862 per person on average in the United States. That’s a total savings of around $1,851 per person; $600 billion overall, or about 21.2% of the total cost of healthcare.

That’s right in line with other estimates – arrived at in a totally different way by looking at private insurance companies directly. It’s satisfyingly close to economist Gerald Friedman’s 2008 estimate that universal healthcare would save $570 billion per year in total healthcare costs.

Final word: eliminating the middleman in insurance could save the USA about $600 billion on current care. (About 5% of GDP)

Part 2: Increased demand for healthcare

When you reduce the cost of something, demand tends to go up. One example is the Affordable Care Act (Obamacare). The ACA included limits on how much could be charged as a co-pay for routine checkups. When co-pays for check-ups fell we saw doctor visits rise. You can also look at out-of-pocket expenses for visiting doctors around the world and see that the cheaper it is to go to the doctor the more visits people make on average. Given that uninsured working-age people are about 40% more likely to die than their insured counterparts, it seems naive to assume that people without insurance are receiving the same level of care that insured people receive. People don’t “go to the doctor anyway.”

Rather than try to build complex economic models that try to guess at behavior, let’s just look at the behavior itself. On average people without insurance spend about half as much on healthcare. This means they’d spend about twice as much (199.6%) if they had insurance. As the first graph shows, average total spending is $8,713 per person, which is $6,970 if we assume 80% of costs covered. Given that uninsured Americans are already spending 50.1% of that, the increase in government spending would be $3,492 per newly-insured person. If the 32 million uninsured Americans were to receive insurance today it would cost the government about 111.8 billion. (80% of $139.7 billion in total new spending)

Put it Together

Total Spending

USA Total cost of healthcare today: $2.8 trillion

(-) Estimated cost with Single-payer universal healthcare: $2.2 trillion

= Savings of $600 billion on current care

(+) Increase to demand: $140 billion

= Total cost of US health costs under UHC: $2.3 trillion

Government Spending

Current government spending: $1.3 trillion

Government covering 80% of total 2.3 trillion costs: $1.9 trillion

Extra taxes required: $562 billion


Universal Health Care would save the people of the United States about $600 billion for the same level of care they’re receiving. We found it would require an additional $562 billion in taxation to cover the government spending, after savings and increases to demand. Is a single-payer universal system worth it? As always, that’s for you to decide – we’re just here to give you the data that helps you make important decisions.

9 thoughts on “How Much Universal Healthcare Would Cost in the US

  • “On average people without insurance spend about half as much on healthcare. This means they’d spend about twice as much (199.6%) if they had insurance.”

    Yea, about that:

    “…[T]he administrative savings from switching to a single-payer system would offset much of the increased demand for health care services. Later, the cap on the growth of the national health budget would hold the rate of growth of spending below the baseline.”

    The CBO also scored Sen. Paul Wellstone’s American Health and Security Act of 1993 in December 1993, finding that “by year five (and in subsequent years) the new system would cost less than baseline.”

    • Oh hey, good to see you again. I actually estimated that the administrative savings would far more than “offset” the cost of increased demand. But demand would still increase, and it’s part of the overall calculation. The “baseline” stat is CBO’s own baseline, so I can’t speak to what went into it.

  • Wow this data about how much universal healthcare would cost is great. Solid breakdown and great research Ryan, learned a lot here

  • Great analysis! I may have overlooked this, but what is the timeframe for this estimated savings and tax increase?

  • Perhaps I a missing a major part here but I am still blanking on how you arrived at your total savings ( ~1.8K per person). You layout three reasons why US healthcare is more expensive (drugs, obesity et. al, and hospital costs) and then you add the first two to arrive at your new total cost? Are you writing off all of the hospitalization costs?

    • I wasn’t too clear on that point, but yes, higher hospital costs (based on the number of hospital visits per capita) are included in the final number. I didn’t want to assume hospital reform so I kept the current US costs in there.

  • Really interesting analysis, thanks for putting it together. Has anyone tried to quantify what the economic impact would be for some of the secondary benefits of UHC? Reduced economic impacts from healthcare-related bankruptcy? Improvements in productivity & reductions in overall spending because people have the opportunity to get medical care earlier rather than waiting until a medical condition reaches a critical stage and they have to go to the ER etc?

Comments are closed.