What Can the US Learn from Other Countries' Health Care Systems?

Even after the Patient Protection and Affordable Care Act (PPACA) of 2010, and in part, because of it, health care remains a major issue of public policy in the United States. It is central to ideologically charged discussions of fairness, the role of government, and even the budget, since the cost of health care is the single largest driver of the federal deficit. In confronting this complex and sensitive issue, it seems only reasonable that we ask what we can learn from the experience of other countries. As the first in an occasional series, this post will look at broad international comparisons of health care systems. Subsequent posts will examine what can be learned from individual countries.


To some on the right, the answer to what can be learned from other countries' health care experience is simple: Nothing! In a recent survey by the Harvard School of Public Health, 68 percent of respondents who identified themselves as Republicans maintained that the United States already has the best health care system in the world. (So did 32 percent of Democrats). There is a widespread view among conservatives that foreign health care systems are socialized medicine. Socialized medicine is bad. Don't go there. Keep the American way.

To some on the left, the answer is equally simple: Providing universal health care is just a matter of political will. Once the commitment is made to treat access to health care as a basic human right, everything else, including better health outcomes at a lower total cost, follows automatically.

Sorry, folks, those easy answers are wrong.

To those on the right: Socialized medicine vs. the American way is catchy political rhetoric, but it has little descriptive value. The role of government varies widely among countries that offer universal health care—so widely that it is meaningless to lump them all together as "socialized medicine." In some countries the government pays the bills, in others, private insurers, and in still others, as in the United States, there is some of each. The same variety is found in other areas of government involvement, including pricing, choice of physicians, choice of treatments, and choice of drugs. What is more, even before passage of the PPACA, US government spending on health care, as a percentage of GDP, was already greater than total expenditure, public and private, in many countries with universal coverage.

To those on the left: Many countries do combine wider access to care with lower total costs and better outcomes, but none of those systems satisfies everyone. All countries face pressures from rising costs. All systems face tough decisions regarding treatments that are costly, but produce small or unproven benefits. All systems struggle with potential conflicts of economic interest between providers and patients. Furthermore, each society poses unique health care challenges. It is unlikely that any other country's health care system, if transplanted intact to the United States, would replicate the results in its home country.

So, does the United States have the best health care system in the world, or not?

There are certainly some good things to point to. The United States does have some of the best medical technology, best medical research, and best-trained doctors in the world. Most Americans, even those critical of the health care system as a whole, report that they are satisfied with the care they personally receive. Yet international comparisons consistently give the US system poor marks overall.

Perhaps the most widely cited study is one put out by the World Health Organization in 2000. Its headline rankings placed the United States 37th in terms of overall performance among 190 countries surveyed, just behind Costa Rica and just ahead of Slovenia and Cuba. France was first and Myanmar last.

To be sure, the WHO ranking has its critics. It is based on a weighted average that combines measures of population health, distribution of health care services, fairness of the financial burden of providing them, and efficiency. The resulting index is subject to several paradoxes: Because of its emphasis on equality, a country with uniformly bad health care could rise in rank relative to one with good health care for the poor and excellent care for the rich. Because it gives a substantial weight to efficiency, a country with excellent care obtained at great expense could rank lower than one with mediocre outcomes attained very cheaply. A country with good health habits in areas like diet and tobacco use could be ranked more highly than one with worse habits, but better treatment for the resulting medical conditions. A 2008 study from the Cato Institute provides a detailed discussion of these and other problems of the WHO rankings.

Although the WHO survey has its methodological problems, many people would still agree, in principle, that a health care system can't be called the best unless it provides good care to all at a reasonable cost. Others, though, would prefer a less subjective measure that focuses on the effectiveness of care. That is the approach taken in a study by Ellen Nolte and C. Martin McKee, both of the London School of Hygiene and Tropical Medicine. Nolte and McKee focus on amenable mortality, that is, deaths among people under 75 that could be prevented by timely and effective medical treatment, for example, deaths from infections, treatable cancers, diabetes, and heart disease. Their study covers the United States, fourteen European countries, Canada, Australia, New Zealand and Japan. Data for 1997-98 and 2002-03 were compared to capture rates of improvement as well as levels of mortality.

Nolte and McKee's findings, shown in the following chart, do not support the claim that the United States has the world's best health care system. In 1997-98, the United States ranked 14 out of the 19 countries surveyed. Over the next five years, all countries improved their amenable mortality rates. For the most part, countries that had performed poorly in the earlier ranking made the largest gains. The United States was the exception. Although its performance was well below average to start with, its improvement over the next five years was the smallest of any country. As a result, its ranking fell to dead last by 2002-03.

Let's turn now to the issue of cost. While the WHO rankings can be criticized for paying too much attention to cost, the amenable mortality approach can be criticized for paying too little. It could well be that some countries have worse health care outcomes because their spending priorities lie elsewhere. For example, the United States has poor passenger rail service and lots of risky dams, but that is mainly because we have decided not to spend much in those areas.

Unfortunately, the same is not true of health care. The US health care system performs poorly not because we spend too little on it, but despite the fact that we spend more on it than anyone else. The next chart shows some data. France, which scored first in both the WHO rankings and the amenable mortality study, spends only two-thirds as much as the United States. Japan, second in the amenable mortality ranking, and Italy, second in the WHO ranking, spend barely more than half as much.

The fact that so much is spent without achieving better results suggests to some that the US health care system is uniquely inefficient. A paper by economists Alan Garber and Jonathan Skinner points to two possible kinds of inefficiency that could explain the results. One is productive inefficiency, which would occur if more capital and labor were required to provide a given service, say implanting a stent or administering a course of antibiotics. The other is allocative efficiency, which would occur if each given medical procedure were efficiently produced, but too much was spent on procedures that provided little benefit compared with their cost.

Comparing the US system to those of other countries, Garber and Skinner's study finds some evidence for both kinds of inefficiency. Higher administrative costs than in other countries suggests productive inefficiency. A tendency toward rapid adoption of expensive new therapies in advance of evidence that they produce significantly improved outcomes suggests allocative inefficiency. Their study also suggests that even when health care resources are used efficiently, their relative prices in the United States are higher than elsewhere.

What, then, is the bottom line? What can we take away from broad, multinational comparisions of the US health care system with those of other countries?

First, we learn that there really is a problem. True, the question, "Whose health care system is best?" is not sharply defined. By playing around with the weights given to coverage, efficiency, medical effectiveness, and fairness, we can get widely different rankings. However, no matter how deftly the data is massaged, it is hard to find empirical support for the claim that the United States has the world's best health care system. That means there must be something we can learn by studying the experience of others.

Second, although broad, multinational comparisions confirm that there is a problem, they suggest few specific solutions. They reveal no magic bullet, no single factor that determines whether a country's health care system works well or poorly. Making the US health care system work better will require multiple, well-coordinated reforms across a broad front, including many areas that the PPACA does not even attempt to come to grips with. Future posts in this series will examine specific strengths and weaknesses of other health care systems with an eye to generating recommendations for improving our own.

Follow this link to view or download a brief slideshow with additional data on international health-care comparisons.

How Chronic Budget Optimism Helped Dig The Hole We Are In

The budget for fiscal year 2012, just published by the White House, presents an optimistic prognosis for US fiscal health. Like all budgets, it looks ahead not just one, but several years. The budget deficit, expected to be 10.9 percent of GDP in 2011, is projected to fall to 7 percent in FY 2012 itself (October 2011 through September 2012), then to 4.6 percent in 2013 and 3.6 percent in 2014. By 2018, the budget is supposed to show a small primary surplus (surplus before interest expense), something essential if the debt-to-GDP ratio is to be stabilized.

Some of the deficit decrease is to come from spending cuts and measures to enhance revenues, but most of it comes from assumed improvements in the economy. Real GDP growth, which is expected to be 2.6 percent this year, is assumed to rise to 3.6 in FY2012, and then to 4.4, and 4.3 percent in the next two. At the same time, according to assumptions, the unemployment rate is supposed to fall steadily from 9.6 percent in 2011, to 8.6, 7.5, and 6.6 percent for 2012 through 2014. The numbers are not brilliant, but compared to the recent past, they don't look bad.

However, if the assumed improvements in the economy don't materialize, neither will the deficit reductions. Overly optimistic assumptions for future years may bring short-term political gains at the moment the budget message is delivered, but they spell long-term trouble. They give Congress an excuse for tax cuts and spending increases the country can't really afford, and they give the White House an excuse for signing off on them. Unfortunately, the experience of the recent past suggests that the Office of Management and Budget (OMB) has had a tendency to look at the world through rose-colored glasses.

That tendency toward chronic budget optimism is well documented in the OMB's own data base, which lets us compare the assumptions on which past budgets were based with the actual performance of the economy. Here is what we find if we compare actual data for 2003 through 2010 with budget assumptions:

The following chart shows that during 2003 through 2010, the OMB assumed economic growth in the budget year itself (labeled Year B in the chart) would be an average of 1.6 percentage points higher than it actually turned out to be. There were no cases in which budget-year growth was underestimated. The same is true for the year following the year being budgeted (Year B+1). For the second year after the one being budgeted (Year B+2), the upward bias of the growth estimate was 1.8 points, even when the one year that was underestimated is averaged in.

Take 2003 as an example. The budget proposal for FY 2003 was delivered in January 2002. That proposal assumed real GDP growth of 3.8 percent, but actual growth turned out to be only 2.5 percent. The error, +1.3 percentage points, is shown by the bar B for 2003. The previous year's budget had estimated 2003 growth at 3.2 percent, for an error of +0.7 points, shown by the bar B+1 for 2003, and so on.

A similar pattern of bias can be traced for the OMB's assumptions regarding unemployment, shown in the next chart. The unemployment rate for each current budget year (B) was underestimated by an average of 0.9 percentage points. For the year following the budget (B+1), the average underestimate was 1.4 points, and for the year after that (B+2) 1.5 points.

No one could expect the OMB's budget forecasts to be accurate every year. Economic forecasting is an inexact art. What we could hope for, though, would be unbiased forecasts that would miss on one side as often as the other, unlike those of the OMB. The chronic bias toward optimism in the budget assumptions for the years covered in the charts clearly contributed to the budget mess the United States finds itself in today. If more cautious economic assumptions had been used, the tax cuts and unfunded spending increases in those budgets would have been harder to justify, and  the national debt might well now be considerably lower.

What about the budget assumptions used by the Obama administration? Are they subject to the same optimistic bias as those of the Bush years? The simple answer is that we don't know yet. The first full fiscal year budgeted by the current administration was 2010, so the above charts each contain only one observation—the "B" observation for 2010—that compares assumptions with outcomes for the Obama OMB.

In preparing the FY 2010 budget, which was published soon after the 2009 inauguration, the Obama OMB overestimated 2010 GDP growth by 0.1 percentage point more than the Bush OMB had done in the previous two budgets, but the error was small in all cases. The Obama OMB assumed 2010 unemployment of 7.9 percent, a large underestimate compared with the actual 9.7 percent, but the Bush OMB had made even larger underestimates for 2010 in the preceding two budgets.

Although we can't yet compare actual with assumed values for FY 2012, the economic assumptions in the just-published budget may once again turn out to be overoptimistic. In a refreshing change from usual practice, however, this White House freely admits that its just-released budget proposal is not sufficient to put the nation on the path to fiscal health. "To get where we need to go, we're going to have to do more," the President said in a February 15th press conference.

He went on to address the concerns of those who were disappointed that the FY 2012 budget incorporates so few of the proposals contained in last year's bipartisan fiscal policy commission. Asked why the commission's report had been shelved, Obama denied that it had been. Despite his disagreements with some specific proposals, he insisted that it still provided a framework for discussion. Why are things taking so long? He compared those who are unhappy with the fact that the deficit has not yet been eliminated to those who were impatient that it took three whole weeks for demonstrators to oust Egyptian president Hosni Mubarak, after three decades in power.

"My goal here," the President said, "is to actually solve the problem.  It’s not to get a good headline on the first day.  My goal is, is that a year from now or two years from now, people look back and say, you know what, we actually started making progress on this issue."

Let's hope that, working with Congress, he can realize that goal.

Follow this link to view or download a short slideshow on the FY 2012 budget and past budget optimism.

A Policy Dilemma: Budget Deficit vs. Infrastructure Deficit

As the federal budget season moves into full swing, infrastructure is not only on the table, but in the center of the table. The Obama administration budget, which would cut some areas of spending and freeze others, calls for more infrastructure spending, including high-speed rail, wireless Internet, and modernization of the electric grid. Across the aisle, House Republican leaders, vowing to "leave no stone unturned and allowing no agency or program to be held sacred," envision infrastructure cuts, including Amtrak, EPA grants for municipal clean water, and other programs. Some Republicans want to outdo the leadership and cap federal spending at 20 percent of GDP, something that would require even more drastic infrastructure cuts.

Who is right? Is infrastructure spending an essential investment in our future or a morass of waste and boondoggles? Where can we safely prune the infrastructure budget, and where can we not?


A good place to start is to ask why we are concerned with the budget deficit in the first place. The cliché is that we do not want to be the first generation to leave our children a national balance sheet with a thinner margin between assets and liabilities than we inherited from our parents.
The trouble is, the federal budget deficit is not the only thing that shapes the national balance sheet. There is also an infrastructure deficit--the difference between what the country invests each year in new bridges, sewers, and power lines and the rate at which the old ones fall apart. If investment in infrastructure exceeds depreciation, the country is that much richer at the end of the year. If depreciation exceeds investment, it is poorer, as surely as if the Treasury sells bonds and uses the proceeds for the most shortsighted spending programs you can think of.

If you have any doubt that the infrastructure deficit is real, try taking a look at the Report Card for America's Infrastructure published periodically by the American Society of Civil Engineers (ASCE). The Report Card assigns grades of "A" through "F" to various infrastructure categories. In the latest report, no area rates higher than a "C+." Roads, aviation, and transit system all declined in score from the previous report, which was issued in 2005. Dams, schools, drinking water, and wastewater stagnated at grades of D or lower. Just one category, energy, improved, from a D to a D+.

Consider dams, for example. There are more than 85,000 dams in the United States with an average age over 50 years. Some 4,000 dams are rated as deficient, including 1,819 high hazard dams. As the following chart shows, for every deficient high hazard potential dam repaired in recent years, two more were declared deficient.

The same story repeats itself in one category of infrastructure after another. The Report Card estimates 5-year infrastructure spending needs at $2.2 trillion. Actual spending is expected to be less than half of that. The 2009 fiscal stimulus package, the American Recovery and Reinvestment Act (ARRA), included $72 billion for infrastructure upgrades, but that was enough to cover only six percent of the 5-year infrastructure deficit estimated by the ASCE.

These realities suggest why we have to be very careful when cutting infrastructure spending. Cuts to essential repairs and upgrades will decrease the federal budget deficit only at the cost of increasing the infrastructure deficit. The trade-off is especially unfavorable when deferred maintenance leads to costly catastrophic failures. And realistically, just eliminating the infrastructure deficit isn't enough. Future economic growth will require an infrastructure surplus, so that wireless communication networks and renewable energy grids can be built at the same time needed repairs are made to aging sewers and bridges.

Still, although cut, cut, cut is not the right approach to infrastructure, spend, spend, spend isn't the answer, either. The problem is, not all infrastructure projects are created equal. Spending on roads, sewers, and parks has a reputation for pork-barreling and corruption that is all too often deserved. If your community is anything like mine, you probably can't drive to the store without "benefiting" from some infrastructure project that has no visible purpose other than generating revenue for some politically connected contractor. Is there any way to separate the wheat from the chaff?

Another useful infrastructure report, this one from the Bipartisan Policy Center, tries to address that question. Although it focuses specifically on transportation infrastructure, it makes some common sense recommendations that are more widely applicable.

• Beware of putting new, borrowed money into existing distribution channels. Those channels tend to share out funds on political grounds rather than zeroing in on the most productive projects. It would be better not to spend at all than to spend without rational prioritization.
• At this stage of the recovery, the highest priority should go to projects that are both shovel-ready and consistent with long-term productivity standards. That is not an argument against budgeting funds now for planning and design of long-term projects like high-speed rail, in order to move them along to the shovel-ready stage. But funds for such projects will never become available if debt is piled on now to fund unproductive, short-term make-work projects.
• Be skeptical of the "jobs multiplier" rationale for infrastructure projects. Focus on the outputs from infrastructure spending, not the inputs.

Unfortunately, these principles are easier to state than they are to implement. At present, the government's budget process seems to be moving away from them, not toward them. Since the onset of the global economic crisis, far too much infrastructure money has been spent on an ad-hoc basis, with specific projects thrown in as "sweeteners" to get bills like TARP and the ARRA through Congress or added as earmarks to unrelated legislation. Congress seems completely to have abandoned the kind of orderly budgeting-authorization-appropriation process that is supposed to allow considered evaluation of individual spending proposals. Instead, we get omnibus spending bills and across-the-board freezes that spend and cut without any sense of priorities.

In short, dams and electric grids are not the only things in need of essential upgrades. President Obama is right that "we can't expect tomorrow's economy to take root using yesterday's infrastructure," and that goes for the infrastructure of budget policy rules, too.

Follow this link to view or download a short slideshow on infrastructure spending.

The Case Against the Mortgage Interest Deduction

As the US economy struggles to recover from recession and cope with a budget crisis, all past policies must be put on the table for review and revision. Even the sacred cows. Even the mortgage interest deduction.

A new report from the OECD, which deserves more attention than it has been getting, explains the role badly-designed housing policies played in triggering the recent economic crisis. As the report shows, housing policy varies greatly among developed economies. There are some areas where the United States scores well. For example, it has a relatively liberal regime of building and land use permits. As a result, the supply of housing responds more to rising prices than in other OECD countries. Also, with the exception of some urban areas like New York and San Francisco, the US rental housing market has a healthier balance between the rights of landlords and tenants. However, in the area of tax treatment of owner-occupied housing, the United States comes off poorly.


From an economic perspective, the goal should be equal tax treatment of housing and other forms of investment. Unequal tax treatment of housing encourages speculative behavior, increases price volatility, and crowds out more productive forms of investment, all to the detriment of growth and macroeconomic stability.

In principle, there would be two ways to level the playing field. One would be to tax the imputed rental value of owner-occupied housing (that is, the rent it would bring if put on the market), while allowing interest costs to be deducted as an expense, just as interest is deductible and rental income taxable for owners of rental housing. The alternative would be to tax mortgage interest but leave the imputed rental value of owner-occupied housing untaxed.

On purely theoretical grounds, the better approach would be that of taxing imputed rents while allowing deductions for interest and depreciation. However, this approach is hard to implement and has rarely been used. A survey of international housing tax policy notes that Sweden taxed imputed rents until 1991 and the UK until 1963, but that both countries have since abandoned the system. Italy still taxes imputed rents, but the marginal rate is lower than for other forms of income, and the tax basis is thought to understate actual rental values. Many countries, including the United States, have property taxes that vary according to value, but these are considered to be imperfect substitutes for taxes on imputed rents. Among other reasons, they are believed to seriously understate both the market value and imputed rents of housing.

The alternative would be to forgo both taxation of imputed rents and deductibility of mortgage interest. Several countries take that approach, including Australia, Canada, Germany, and Japan. Such a system does not level the playing field as well as a fully implemented system of taxing imputed rents, but is far more practical from an administrative point of view.

Other tax policies can also affect the relative attractiveness of housing investment, including tax treatment of capital gains on housing and deductibility of local property taxes against national income taxes. One way to capture the total impact of the tax system on returns to housing investment is to look at the gap between market interest rates and the after-tax debt financing costs for housing. The greater the gap, the greater the pro-housing bias of the tax system. The following chart from the OECD report shows that the gap for the United States, while not the largest of all member countries, is considerable.

Aside from its effects on growth and economic stability, the mortgage interest deduction is open to criticism on distributional grounds. For reasons detailed in a recent study from the Urban Institute-Brookings Tax Policy Center, the great bulk of the benefits go to higher-income households. First, because it is a tax deduction, not a tax credit, mortgage interest relief benefits only those who itemize deductions on their personal income tax returns. Some 98 percent of tax units with incomes over $125,000 itemize, compared with just 23 percent with incomes of $40,000 or less.Second, because there is no cap on the deduction, owners of more expensive homes gain more than those with more modest homes. Third, a dollar of tax deduction is worth more to households in higher tax brackets than to those in lower tax brackets.

The Tax Policy Center study concludes that the mortgage interest deduction is worth $5,393 a year for tax units in the top 1 percent of the income distribution (average income $1,302,188) but only $215 per year to those in the middle 20 percent (average income $43,678). For households in the bottom 20 percent of the income distribution, the deduction has almost no value.

The final element of the case against the mortgage interest deduction is the potential contribution its elimination would make to resolving the US budget crisis. As I have argued in a previous post, tax reform that lowers marginal rates while broadening the tax base is by far the most growth-friendly path to fiscal consolidation. The Tax Policy Center study estimates that elimination would increase federal revenues by $108 billion in 2012, rising to $162 billion in 2019. Taking into account the likelihood that people would sell interest-earning assets to pay down mortgages would reduce the revenue figures slightly, to $91 billion in 2012 and 138 in 2019. By comparison, it would take a 15% across-the-board cut in non-defense discretionary spending to achieve the same deficit reduction.

On the other side, the principal argument in favor of the mortgage interest deduction is the claim that home ownership would otherwise be unaffordable for a large segment of the population. For two reasons, that argument does not hold water.

First, as explained above, the mortgage interest deduction has little value for lower- and middle-income families, those most likely to be on the borderline between rental and ownership. A subsidy worth $215 a year is not going to move many $40,000 families out of rental housing into a home of their own. Neither would a tax increase of $5,000 a year be likely to induce many millionaires to move out of their own homes into rentals.

Second, the impact on the affordability of housing is reduced to the extent that the interest deduction is capitalized into the market value of homes. Greater tax benefits cause buyers to bid up home prices. Conversely, elimination of those benefits would make home ownership less attractive, so home prices would fall. Capitalization of tax benefits does not fully offset the impact on home ownership because higher prices also encourage construction of new homes, but the effect is substantial.

The above discussion is framed in the context of immediate and full repeal of the mortgage interest deduction. Some would argue that such a cold-turkey approach is politically infeasible and would be too disruptive to the housing industry and family finance. For those who prefer a more gradual approach, at the cost of a more modest contribution to fiscal consolidation, there are many well-developed proposals for partial or phased repeal. One idea would be to switch from a deduction to a tax credit, which would be worth more to low-income families. At the same time, the value of the deduction to upper-income families could be capped in one way or another. Either of the above ideas could be phased in gradually over time, and structured to make a larger or smaller contribution to deficit reduction.

Will any of it happen? Some people think not. There is a widespread, although poorly substantiated, belief that home ownership makes people better citizens. (Skeptics think it is just as likely that the causation runs the other way, with responsible, hard-working citizens being more likely to own homes.) There is a more strongly substantiated tendency for wealthy individuals, who receive the bulk of the benefits of the interest deduction, to exercise disproportionate political influence through higher voting rates and bigger campaign contributions. For both reasons, the mortgage interest deduction is truly a sacred cow of American politics. It is a cow that should be led to slaughter.

Follow this link to download a brief slideshow with additional charts and data concerning the mortgage interest deduction.