Is percentage of GDP the right way to measure health care costs?


Generally when people question whether or not our central planners are allocating the right amount of money for health care, the debate is cast in terms of percentage of GDP (given that government will spend 67 percent of health care dollars in the Obamacare era (source) it seems fair to say that the level of health care spending is now almost entirely determined by bureaucrats in Washington, D.C.). Given how capital-intensive medicine has become, however, I’m wondering if this is the right measure.

The central planners are currently spending 18 percent of our GDP on health care and they are on track to increase this spending to 25 percent. As I noted in my health care reform proposal back in 2009, the Mexicans spend 6 percent of their GDP on health care (World Bank) . So in theory we will be spending only 4X as much as the Mexicans. But measured in dollars, we will actually be spending more like 20X (since our per-person GDP is about 5X larger, according to the World Bank).

If the only cost to health care were labor maybe it would make sense to use the percentage of GDP as a comparison. But health care involves a lot of stuff that is traded on the world market and costs about the same everywhere. There is cement to build hospitals. There are pills that come from factories in Israel and India. There are machines that go “bing” in the treatment rooms. There are LCD televisions in the waiting room.

So when people say that we spend double what the U.K. spends on health care right now (18 percent of GDP versus 9 percent), despite the fact that the U.K. manages to cover all of its residents without burying them in billing and insurance paperwork, that’s understating the difference. GDP in the UK is only about 75 percent of US GDP. A comparison in terms of spending power for buying stuff other than health care would show that we are spending closer to 2.7X what the UK spends.

What do folks think? Is it more accurate to assess our government’s competence in delivering health care with percentage of GDP consumed or absolute dollars spent?

13 thoughts on “Is percentage of GDP the right way to measure health care costs?

  1. I think percentage of GDP is a decent way to compare even if the best comparisons come are between those countries of similar development stages. As a country gets “richer” it will consume more healthcare in absolute terms as access is gained to new, expensive treatments before they have reached economies of scale (pharma) or been iteratively optimised (surgery and other interventions). But the % of GDP should, ideally, hold roughly steady if there is appropriate control on access and costs. This does not take into account that as healthcare keeps delivering longer life expectancies we have to treat people for more things over a longer period, they are more likely to require expensive interventions and there longer length in retirement means a greater time not contributing productively to GDP.

  2. Percent of GDP per capita would be better. That’s evidently what you’re referring to in the 4th paragraph (GDP/capita is about 75% in UK what it is in US).

    It corresponds to percent of income spent. It seems reasonable to spend a certain fraction of my income on medicine.

    An alternative is absolute spending per capita. But rich people spend more in on houses, cars, and everything else in life — so that doesn’t seem right.

    Fraction of GDP isn’t right because medicine is needed by people, but GDP/pop. varies by a factor of 500 among nations.

  3. What you want is expenditure per capita in inflation adjusted dollars, adjusted for purchasing power parity for international benchmarking purposes.

  4. I only recently spoke to someone about the health care system in the UK. I was not impressed. So until you are comparing apples to apples I am not sure how any comparing works out. What a particular patient gets in the UK is. Dependent on their local council some pay for IVF and some don’t. All patients unable to pay are “paying” the same but getting different things.

  5. Let see, if we pay a 20% premium to private companies as legally required, limited, for health care. it seems the employee share seems to be paying the overhead, at least where I work.

    Why can’t we get a more equitable approach like down under where everyone pays the same, and the salaries are set by industry to allow people to pay, live, etc.

    We spend more on military than the other top city tires, combined. What should be the measure for this.

  6. Phil,

    you make a compelling argument for the raise of a National Health Service in the US of A — funded by taxpayer money and covering every US citizen. You describe a situation where keeping insurance companies involved in the health providing business can only balloon the costs in a way that is utterly indefensible.

  7. Federico: I didn’t propose a government-run health service in . I know that it is common in most other countries (England, Mexico, etc.). However, there is a risk that it wouldn’t be cost-effective here due to the fact that government workers are paid so much more than private-sector equivalents (see ). What if the USNHS workers could retire at age 50 with a $200,000+ per year pension, like a California fire department or police department desk worker? Or retire at age 40 like a New York City police officer or U.S. military enlisted person? Calculating the true cost of providing the service would become impossible. Politicians would paper over the cost by postulating absurd future real returns in a pension fund. Whereas if universal health care is provided with a fixed-price voucher to be given to an HMO, the pension risk (if any) and the cost-escalation risk is passed on to private shareholders.

    Merely because a government-run program works in some other country does not mean that a government-run program would work in the U.S. We have a different political system, a different amount of corruption (less cash bribery than in some other places but more cronyism and benefits to companies that make political contributions and/or lobby), different laws regarding the ability of government workers to unionize, different average government worker quality (partly due to a different education system and partly due to different alternatives in the private sector), etc.

    [Today’s New York Times has a story about a city in California where the median household income is just over $31,1000 per year and the average police officer is paid over $177,000 per year (including benefits such as pension, though that is probably understated).]

  8. I think it should be measured in terms of the lifestyles of doctors, administrators and insurance providers. For example, the mean square footage of all primary homes, second homes, third homes, vacation houses, and some correction to account for hotels and rentals. True this leaves out the opulence, but would nonetheless be very revealing.

  9. There is nothing intrinsically wrong with spending 18% of the US GDP in any given segment of the economy. We certainly spend a lot more in computers as percent of GDP than three decades ago. I’ll be willing to pay way more than 18% of my income in exchange for unlimited access to excellent medical care. Are Porsches more expensive than FIATS? Sure, but many people are willing to pay extra for a superior product. The problem is that Americans don’t have much to show for what they pay for medical care. They pay for the Porsche and get to drive a Yugo.

  10. A family we know has a teenage daughter who had terrible pain in her back. It turned out to be a (benign, thankfully) tumor growing next to her spine. The dad described the surgery where they were able to monitor the blade as they removed the tumor and whether it was getting too close to the nerves.

    I have no idea what the “cost” of the surgery was. But one of my daughters has a slight speech impediment from surgery on her jaw 8 years ago at the same hospital. So it’s pretty amazing to hear some of the progress that they’re making.

    I’m not sure how you can make comparisons just on price. That’s certainly the easiest thing to compare.

  11. > I didn’t propose a government-run health service […]
    > it wouldn’t be cost-effective here […] government workers are paid so much more

    It can be government-managed through private contractors. That’s the way the US designs and produces the weapons which ensure its military and political supremacy.

    There’s an important cause for US excessive healthcare costs you rarely mention: ambulance-chasing lawyers. Not only do they command big insurance premiums, they force doctors to leave no rock unturned, and to protect themselves with diagnostic procedures of debatable cost/benefit value, and needlessly intrusive preventive actions. Finally, once you get a sector in a wasteful mood, where you don’t think twice about ordering very expensive procedures just-in-case, you have opened the pork barrel, and many other parasites will find ways to feed off the healthcare industry’s careless attitude.

    Unfortunately, given how many congressmen are former lawyers, they’re unlikely to fight a trend which move healthcare money into lawyer pockets. It’s a shame: offering special prices to patients who “voluntarily” waive most of their litigation rights would fix many issues. Even simply forbidding “no gain, no fee” lawyer contracts would moderate many abuses. It’s done in some countries, at least France, for this very reason: you don’t want litigators with little or no skin in the game.

    Coming back to the initial question (GDP or USD?), I can’t even think of a good reason to count in GDP. As you said, only salaries ought to be indexed on GDP, and medical salaries are hardly what explains US costs. Moreover, with easier and cheaper transportations, imore procedures become sensible to get done abroad. For instance, it’s increasingly popular in Western Europe to go have major dental care performed in Eastern Europe. They know how to glue fake teeth in your jaw, and charge ten times less, for procedures which are generally not covered by universal insurance.

    Such medical tourism would have another beneficial outcome in the US, namely it would disarm ambulance chasers: good luck suing a Mexican doctor in Mexico, under Mexican law, with a US lawyer. There are some very high-tech forms of care which are only accessible in the US, but they’re the minority, and almost inexistent if you’re willing to go as far as Europe. The only remaining issues are time-sensitive emergencies (you won’t be flawn to Tijuana for a PT scan following a road accident in Vegas), and chronic conditions such as being old and impotent (although retiring to Belize wouldn’t be such a stupid strategy, if you don’t mind seeing your grand kids mostly through Skype).

  12. > There are LCD televisions in the waiting room.

    There might be in the US. There aren’t in the UK, Germany, India or Russia – those being all the countries where I have had personal dealings with the healthcare systems. And, fortunately for me, the healthcare facilities I visited in Russia and India were rather upscale by those countries’ standards.

    So that’s part of your cost difference right there.

Comments are closed.