Scott Sumner  

Ezra Klein on the Singapore health care miracle

PRINT
Reply to Adam Ozimek... Keep Your Eye on the Prize...

Ezra Klein has a good essay on the Singapore health care system. He starts off with the conservative case for the Singapore system, quoting the AEI:

What's the reason for Singapore's success? It's not government spending. The state, using taxes, funds only about one-fourth of Singapore's total health costs. Individuals and their employers pay for the rest. In fact, the latest figures show that Singapore's government spends only $381 (all dollars in this article are U.S.) per capita on health--or one-seventh what the U.S. government spends.

Singapore's system requires individuals to take responsibility for their own health, and for much of their own spending on medical care.


Most of the essay is spent partially (but not completely) debunking the conservative view. For instance:

According to the World Bank, in 2014 Singapore spent $2,752 per person on health care. America spent $9,403. Given this, it's worth asking a few questions about what Singapore's model really has to teach the US.

Are Singaporeans really more exposed to health costs than Americans? The basic argument for the Singaporean system is that Singaporeans, through Medisave and the deductibles in Medishield, pay more of the cost of their care, and so hold costs down. Americans, by contrast, have their care paid for by insurers and employers and the government, and so they have little incentive to act like shoppers and push back on prices. But is that actually true?

I doubt it. The chasm in total spending is the first problem. Health care prices are so much lower in Singapore that Singaporeans would have to pay for three times more of their care to feel as much total expense as Americans do. Given the growing size of deductibles and copays in the US, I doubt that's true now, if it ever was. (It's worth noting that, on average, Singaporeans are richer than Americans, so the issue here is not that we have more money to blow on health care.)


I think he's overstating the case here (I find that even many of my small health expenses are heavily subsidized) but there is some truth to what he says here. And I think this points to the necessary first step in health care reform---getting costs down. Indeed Klein's basic argument is that the Singapore system is actually a pretty effective health care regime, but it would be hard to implement in America because the cost of health care is so much higher here.

But this leads to what I see as the one major blind spot in Klein's article:

Singapore's system is probably better designed in terms of how consumers spend their own money. But the lower overall prices make them much less exposed to health costs than both patients and employers inside the American system -- which suggests to me that Americans have at least as much incentive as Singaporeans to try to use their power as consumers to cut costs.

The fact that that hasn't worked is, I think, a reason to believe we've gotten the lesson of Singapore's health system backward. Singapore heavily regulates both the pricing and provision of medical care to keep costs low (as do all other developed countries) and then, working off that baseline of low costs, has Singaporeans pay out of pocket in order to keep them mindful of how much they're spending.


Unless Klein doesn't consider America to be a "developed country", the statement is flat out wrong. In America the government heavily regulates both the pricing and provision of health care to keep costs high. In America:

1. There are massive subsidies to employer provided health insurance. This leads to enormous consumption. My personal lifetime health care consumption has been at least doubled by various subsidies (including tax breaks), and it has not improved my health one iota.

2. There are huge regulatory barriers to the efficient provision of health care, at virtually every level of the system. It's too difficult to become a doctor. Nurses are prevented from doing too many things. There are insurance mandates. There are restrictions against patients signing contracts making it more difficult to sue for malpractice. There are restrictions on importing foreign drugs, even foreign generics. There are barriers to the immigration of foreign doctors and nurses. I could go on and on.

My dream policy would start with massive deregulation, as well as the elimination of all tax deductions for health care, to get costs as low as possible. Then add mandatory health savings accounts to get costs even lower.

Klein also overlooks the fact that the argument he uses against conservative supporters of the Singapore system applies equally well against progressive supporters of the European approach. He argues that since health care in America is so much more expensive than in Singapore, we probably pay almost as much out of pocket as they do, despite paying a lower percentage.

But that's also true of socialized medicine. Thus our government spends about 8% of GDP on health care (out of a total of roughly 18% of GDP), whereas European governments spend about 8% of GDP on health care out of a total of 9% to 10%. So we are already spending roughly as much as in Europe. If a single payer regime were actually feasible in America, then our government should already be able to provide universal coverage out of the already existing health care programs. We should to be able to put all 325 million Americans on Medicare/Medicaid/VA, etc., without spending a dime more. Obviously that won't work.

So our current bloated health care spending levels means that America is not able to choose either the Singapore or the European system, we are stuck with a far worse, system---the American health care system.

But if we insist on dreaming the impossible dream, then why not shoot for the best? And that's clearly Singapore, not Europe. And the first step toward getting there is cutting costs, by doing things like replacing the tax deductibility of health insurance with a lump sum tax credit.

Screen Shot 2017-04-26 at 9.59.38 AM.png


Comments and Sharing






COMMENTS (24 to date)
RPLong writes:
And the first step toward getting there is cutting costs, by doing things like replacing the tax deductibility of health insurance with a lump sum tax credit.

I'm curious as to why you think tax policy is the first step toward getting a better health care system. There are many ways to improve the system, and a good number of them would have widespread political support among the American population and would make tangible positive impacts on the system. What makes tax policy the first one to choose?

Maniel writes:

@RPLong

I'm curious as to why you think tax policy is the first step toward getting a better health care system.
I’m with Scott on this one for three reasons. First, if you want less of something, tax it; if you want more, subsidize it (got milk?). By making employer-provided health insurance tax deductible (Your employer gets a tax break on its portion of Social Security and Medicare taxes), government has encouraged employers to use employee-pre-tax money to purchase insurance in the employee’s favor. This advantage is not available to that part of the population not covered by employer plans. Using government think, this was the rationale for the ACA: if you have one bad idea in place, rather than undo it, devise another, even worse idea.

Second, the more I spend on health care – including insurance, contact-lens fluid, massage therapy, etc. – the more I can deduct from my income tax. (Got contact lens fluid?).

And third, Scott did say “first;” so did you. By the way, he said “the first step toward getting there is cutting costs” and tax policy is a major driver. Getting government out of the subsidy business is necessary to begin moderating government-driven health-care inflation. He mentions much more of course, but trying to get government to undo bad policy is pretty much a fool’s errand; so, better to exercise, eat healthy, and drive safely.

Seth writes:
The basic argument for the Singaporean system is that Singaporeans, through Medisave and the deductibles in Medishield, pay more of the cost of their care, and so hold costs down.

I doubt it. The chasm in total spending is the first problem. Health care prices are so much lower in Singapore that Singaporeans would have to pay for three times more of their care to feel as much total expense as Americans do.

What am I'm missing? This seems like circular reasoning.


ZC writes:

"My personal lifetime health care consumption has been at least doubled by various subsidies (including tax breaks), and it has not improved my health one iota."

Why do you keep seeking out (and spending money on) health care that doesn't improve your health?

Todd Kreider writes:

Singapore is a wealthy, authoritarian city-state.

* American auto fatalities are three times higher than in Singapore along with a much higher non- fatal accident rate, which puts a strain on health care costs.

* The U.S. is far more violent than Singapore with a homicide rate that is 13 times higher. There are also far more violent, non-fatal crimes in the U.S., which puts another strain on health care costs.

* Drug addiction is a far greater problem in the U.S. than in Singapore, adding to a third strain on health care costs.

* Singaporeans are far healthier than Americans with 3% who are obese compared to 35% of Americans, which adds a fourth major strain on health care costs.

Klein makes a reference to this toward the end of the article but without the details that show how stark the differences are. Through Matt Yglesias he writes: “If you imagine America with no guns, less booze, much less drugs, and radically less driving, our public health outcomes would soar.”

But at the same time health care costs would plummet.


RPLong writes:

Maniel, thanks for the reply. You mention valid reasons why we might think about adjusting tax policy, but unless I'm not understanding your argument, you haven't provided any reasons why this is the best first step instead of other options that have widespread political support, such as allowing customers to import cheaper medications from abroad.

Changes to the FDA approval process don't even require political will, but only a decision on the part of the Executive Branch to uphold the law in a different way. For example, it would be easy to streamline the approval process for bio-similars even further than it already has been. It could be done in a day, would save the industry millions, would save customers millions, would increase the supply of medication, and would be a politically popular move. Why is tax policy a more logical starting point? That's my question.

Scott Sumner writes:

RPLong, Because it's the biggest single problem, by far. Under our current tax system, the government pays roughly 40% of the cost of many private medical expenses. That encourages massive overuse of health services.

ZC, In some cases I thought it might help, but only afterwards did I find out it did not. In other cases (such as daily wear contact lens), it was for comfort and cosmetic reasons, unrelated to health.

Todd, The issues you raise affect life expectancy (outcomes.) They do not necessarily affect health care costs. The amount of medical care consumed by a guy murdered at age 24 or in a fatal car accident at age 30 is often far less than if he had lived to 87.

James M writes:

@Todd Kreider

Excellent points. Ezra and Matt mention in their article and podcast that Singapore has a 15% car ownership rate, no gun ownership, and that they also execute for drug offenses.

Steve J writes:

Scott claiming that a less healthy society is cheaper to care for than a more healthy society does not make sense at first glance. Society will put more effort into saving an unhealthy 65 year old compared to an unhealthy 85 year old. At some point people say they've lived long enough. That doesn't happen as often in the US.

I'm with RPLong on what the first step should be.

Todd Kreider writes:

Scott, you replied:

Todd, The issues you raise affect life expectancy (outcomes.) They do not necessarily affect health care costs. The amount of medical care consumed by a guy murdered at age 24 or in a fatal car accident at age 30 is often far less than if he had lived to 87.

Of course they have a major impact on costs. It costs a lot to send the many survivors of violence and auto accident to the ER, then often for a hospital stay and at times costly rehabilitation.

Obesity significantly increases the risk of diabetes, heart disease, Alzheimer's disease and doubles the risk of getting several cancers. Those are also very expensive to treat and often for many years.

ZC writes:

"My personal lifetime health care consumption has been at least doubled by various subsidies (including tax breaks), and it has not improved my health one iota."

But it has improved your health. You've benefited greatly from vaccines (even smart economists apparently overlook the benefits of disease avoided/eradicated).

And you're the beneficiary of corrected vision -- you obviously take it for granted, as all of us do so many advances in our modern, rich lives.

And, you benefit greatly from the preserved longevity and contributions of so many in this country, people who have survived horrible cancers or other maladies that were a death sentence years or only a few decades ago, but are now treatable with people returning to their productive lives.

Heath care in the US has myriad problems, but to state that your health hasn't been improved one iota is painting with far to broad a brush.

AlanG writes:

I'm sorry but this is a rather silly post. Singapore has a lot of price controls built into its healthcare system just as a lot of other countries do. TR Reid covered a lot of this in his book, "The Healing of America: A Global Quest of Better, Cheaper, and Fairer Healthcare." Even though the book came out in 2009, many of the points are just as relevant today. There is just no political will to do what needs to be done. It's the same for tax reform.

Doing away with the employer tax exemption for health insurance will be even more important as it appears that the individual exemption will be wiped out if the current tax "reform" package goes through. Gary Cohn said today that the only deductions for individuals will be home mortgage interest and charitable contributions. All you freelancers better start saving up money to pay the full amount of your health insurance.

Jamie writes:

People also forget that Singapore and other small countries can keep down costs on drugs because they have monopsony bargaining and can pay MC. If the US did the same, we'd destroy medical innovation. There is a huge free rider problem.

Additionally Ezra misses the fine point that while Americans pay large out of pocket costs relative to Singapore, the out of pocket costs do not vary with the cost of the MRI.There is no incentive to drive twenty minutes to get the MRI that costs half as much. That is even more of an issue than choosing whether or not to get the MRI in the first place.

Also the cash only clinics are a glimpse into how efficient we can get the supply side of the system without regards to increases in innovation and entrepreneurs. The question is how can we shape policy to transition to the world instead of accept status, mediocrity and single payer like Ezra offers.

I also found Ezra too easily dismissed private sector administration of catastrophic care. While a small percentage, you can't dismiss it and not ask the question why can't it be 100%?

Todd Kreider writes:

Two more points. Scott wrote in his reply to me:

The amount of medical care consumed by a guy murdered at age 24 or in a fatal car accident at age 30 is often far less than if he had lived to 87.

First, medical expenses can be high if the person lives a few days or couple of weeks in a hospital before dying.

Second, a 24 and 30 year old today is very unlikely to incur medical costs until 60. That is the year 2053 for the 24 year old and the year 2047 for the 30 year old.

Most ailments will be gone by then (diabetes, cancer, heart disease and Alheimers) and whatever remains could be treated for peanuts 30 years into the future.

And the 24 year old will be 87 in 2080! Any shifts in health care/costs by then?

Gary Irwin writes:

In Singapore doctors and other health care professionals make between 1/2 and 1/4 on average what comparable professionals make in the United States and, add that to the nanny state elements, the harsh penalties for drugs etc, lower levels of cigarette smoking and I suspect that accounts for at least 50% in the difference in costs.

ZC writes:

@Scott Sumner

"The issues you raise affect life expectancy (outcomes.) They do not necessarily affect health care costs. The amount of medical care consumed by a guy murdered at age 24 or in a fatal car accident at age 30 is often far less than if he had lived to 87."

No, they don't just affect outcomes. Violence and motor vehicle accidents are responsible for tremendous healthcare expenditures. Even the fatalities can incur huge health care expenditures as we're really good at keeping even hopeless cases alive for hours, days, weeks, and longer.

Your dismissiveness of Todd's solid points that cultural factors in the US are huge drivers of high (and ineffective) healthcare expenditures lead me to believe you have little personal understanding of how heath care is organized, delivered, consumed, and paid for here in the US. Spend a weekend night in an busy urban ER where you can watch people who have never, and will never work a productive day in their lives collectively consume millions in health care expenditures as a result of bad behavioral decision on their part.

You're right that Singapore has successfully reigned in health care expenditures, but it's in no small part because the authoritarian regime harshly enforces their laws including prohibitions on drugs and firearms as well as strict limitations on the timing of alcohol sales and public consumption on alcohol (a drug responsible for billions in health care costs in the US annually). While such laws and enforcement undoubtedly result in decreased health expenditures, I'm not sure they'd ever fly here in the US. To ignore the impact of cultural differences on per capita health care expenditures is to be ignorant of the driver's of health costs.

ChrisA writes:

I think it is instructive to compare Singapore health costs with Hong Kong. https://www.bloomberg.com/news/articles/2014-09-18/singapore-beats-hong-kong-in-health-efficiency-southeast-asia. Hong Kong has almost exactly the same outcomes, at slightly lower overall cost (as percentage of GDP). However the HK medical system is like the NHS in UK, basically free to use. So I doubt it is the way that the health system in Singapore is funded that makes health care so cheap there, I would say that the underlying genetics and culture (i.e. lack of violent crime, etc etc) that makes the real difference. So importing that system to the US would mean replacing the population of the US with overseas chinese, not exactly a practical suggestion.

Justin Dailey writes:

I'd suggest converting the public side of U.S. expenditure into publicly funded health spending accounts.

The problem with current HSAs and high deductible health plans is that most big ticket health care spending is past the deductible and thereby unaffected. Plus, you're still stuck working through the insurance system, and the insurance company (or government) is the ultimate customer.

The U.S. spent $5,960 per person on health care in 2013 directly or indirectly through government (Medicaid, Medicare, tax deductibility of health insurance, VA hospitals, insurance for government employees). With Obamacare and spending growth, that number is surely higher now, but let's use the lower number. If the average person lives to 78, then $5,960/yr represents $464,880 in total spending. Why not just populate an account with that much money, adjusted for age and health expenditures by year (e.g. a new born would get $464,880, a thirty year old perhaps $405,000, a sixty year old maybe $255,000). This lets everyone form their own insurance company with friends and family (e.g. a 30 year old couple with newborn twins would have $1.74 million to spend on health care for the rest of their lives). People really would be the consumers in charge on all of their health care. You'd effectively create your own insurance company via your family and friends, who could share these resources. And this wouldn't cost any more than the existing system; indeed, it would cost less, and the government's maximum exposure is explicitly limited. Health care costs should fall significantly and outcomes should improve significantly if what we believe about market forces is true. As with Medicare for all, administration costs should be lower as most payment will occur directly between consumers and providers at stated prices rather than long back and forth negotiation and claims processing.

Balances would grow at GDP per capita growth, limiting the government's exposure to a fixed proportion of GDP in the long run. Any balances left in a person's account at death would be split between designated beneficiaries HSAs and returning to the government.

I'd also considering modifying the above to have somewhat smaller accounts, but have a few carve outs for extremely expensive conditions/situations and publicly funded ERs as there are limits to how markets can work in that setting in particular.

1. Local governments to fund ERs
2. Government catastrophic coverage for any condition that has cost a high amount over a lifetime (e.g. $250,000 or $350,000 threshold), which provides free additional coverage for that particular condition alone.
3. Subsidies for long term care (e.g. 95% of median home health care costs, 85% of median nursing home/assisted living).

Todd Kreider writes:

One more point that I haven't seen mentioned: As with Japanese, Singaporeans have been much more reluctant to seek help for mental issues. I know that this started to change in Japan in the early 2000s and may have in Singapore, too, but not sure.

My guess is that mental health costs are still much lower in Singapore than in the U.S.

Scott Sumner writes:

Everyone, I'm not at all convinced that factors like auto accidents and gunshot wounds are a big driver of US health care costs. I suspect the biggest expenditures are for treating the elderly. If someone can show me that most expenditures are not related to treating illnesses, I'll change my views.

Todd Kreider writes:

Scott,

First, asking to show that auto accidents and gunshot wounds (and stabs and beatings) and auto accidents are most of U.S. health care costs is a strawman -- it is clearly significant. You also left out mental care costs as did Ezra Klein and Yglasesis. (That is typical as younger people without mental health issues often forget they exist.)

A third of Americans are not just overweight but obese whereas only 3% of Singaporeans are. The science shows this has a major effect on getting cancer, diabetes, heart disease and dementia.

By coincidence, last week I read in the NY Times comments a doctor in NYC who wrote: {paraphrasing} "If you want to know what is driving up health care costs it is men in their 40s who come to the ER who are seriously obese with a drug and/or alcohol addiction."


Hazel Meade writes:

High deductibles and co-pays are a relatively recent and not very universal phenomenon. I don't know why Klein expects them to have already driven down costs. It took decades for the employer based system to degenerate to the point it was at in 2010. Most people are still covered by employer-based plans with low deductibles.
Give the market some time to correct.

Relatedly, since prices are negotiated via insurers, it may be a barrier to consumers shopping around. First, an employee cannot control terms of the employer-based plan his employer has selected. He generally has only 2-3 plan options, maybe even from the same insurer. So the negotiated prices are beyond his control.

Maybe the consumer might want to use a lower-cost provider since he has not met his deductible - how does he go about doing that? Negotiated prices are not publicly advertised. Insuerer websites do not list the negotiated prices easily via a user interface. The patient must call around and ask for that information, in order to identify the lowest cost in-network provider. Many providers won't even provide it over the phone. The patient must get it from the insurer. It's a lot of work. It's possible, but the system is not yet set up to make this easy. Maybe over time insurers will start offering utilities to make it easier for patenits on high-deductible policies to find price information, but that system isn't there yet. Limited competition in employer-based plans reduces the increntive to create it.

Again, give the market some time. Price signals via the employer based system are obscured. It's going to be slow. Getting rid of employer-based plans would help.

DD0000 writes:

I agree 120% with ChrisA. People often make these cross country comparisons without taking into account demographics.

You see people comparing the Nordics to the US and coming to the conclusion that a large state is ideal. However comparing the Nordics to a demographically similar country, like Switzerland, would get you the exact opposite conclusion.

If you want to see what healthcare system is the best, look at which gives best results *based on the country's demographics*. E.g. Which country outperforms its genetically similar neighbors?

Floccina writes:

1. If it weren't for the over 65's you might be able to have the Gov provide a health insurance plan for all for cheap but since Medicare covers a lot and they would need to be in the plan the AARP would kill it. Perhaps you start with the new program with people born after 1961.

2. Another idea: Because it is easier to move to a new state than to a new country one of the federal government's comparative advantages is raising tax revenue, but the states do most of the regulating of health care and that is a problem for socialized medicine, so you could have the feds send all the health care money to states and tell them that they must cover everyone with the money. The state governments could do plans that cover minimal care or add money and provide full care. In the minimal care states people could buy additional private coverage. This would motivate state level politicians to regulate with an eye on cost.

Comments for this entry have been closed
Return to top