China: Does 8% Growth Cause Less Satisfaction?

China's economy grew at extraordinary annual rates of 8% or more, on a per capita basis, in the two decades from 1990 to 2009. Using the old "rule of 72" that is sometimes taught to approximate the effect of growth rates, take 72, divide by the annual growth rate, and it will tell you (roughly) how many years it takes for the original quantity to double. So at an 8% growth rate, China's per capita GDP doubles in 9 years, and quadruples in 18 years. In the two decades from 1990-2009, average per person GDP in China has quadrupled, at least. The number of Chinese living below the international poverty line of $1.25 in consumption per day fell by 662 million from the early 1980s up to 2008, according to World Bank estimates.

But survey researchers as people in China (and all over the world): "All things considered, how satisfied are you with your life as a whole these  days? Please use this card to help with your answer:
       1 “dissatisfied” 2 3 4 5 6 7 8 9 10 “satisfied”."
These researchers find that people in China are not, on average, more satisfied in 2009 than in 1990. How can this be?

This finding is an example of the "Easterlin paradox." Back in 1974, Richard Easterlin wrote a paper called A. (1974) "Does Economic Growth Improve the Human Lot?" which appeared in a conference volume (Nations and Households in Economic Growth: Essays in Honor of Moses Abramovitz, edited by Paul A. David and Melvin W. Reder). The paper is available here.  Easterlin found that in a given society, those with more income tended to report higher happiness or satisfaction than those with less income. However, he also found that the average level of happiness or satisfaction on a 10-point scale didn't seem to rise over time as an economy grew: for example, in the U.S. economy between 1946 and 1970. He argued: "The increase in output itself makes for an escalation in human aspirations, and thus negates the expected positive impact on welfare."

But can this effect hold true even when the standard of living is rising as dramatically as in China?  Easterlin, still going strong at USC, looks at the data with co-authors Robson Morgan, Malgorzata Switek, and Fei Wang in "China's life satisfaction, 1990-2010," just published in the Proceedings of the National Academy of Sciences.  Here is a (slightly messy) graph showing survey results from six different surveys of satisfaction or happiness in China.: the World Values Survey, a couple of Gallup surveys, and surveys by Pew, Asiabarometer, and Horizon. The surveys use different scales: 1-10, 0-10, 1-4, 1-5, so the vertical axes of the graph are a mess. But remember, this is a time frame when per capita GDP more than quadrupled! It's hard to look at this data and see a huge upward movement.

Easterlin and co-authors summarize the patterns this way: "According to the surveys that we analyzed, life satisfaction in the Chinese population declined from 1990 to around 2000–2005 and then turned upward, forming a U-shaped pattern for the period as a whole (Fig. 1). Although a precise comparison over the full study period is not possible, there appears to be no increase and perhaps some overall decline in life satisfaction. A downward tilt along with the U-shape is evident in the WVS, the series with the longest time span."

Indeed, Easterlin and co-authors point out that the happiness trend may be biased upward, because of this time there was a large rise in the “floating population" of "persons living in places other than where they are officially registered) in urban areas." This group tends to have lower life satisfaction.  " Between 1990 and 2010, the floating population rose substantially, from perhaps 7% to 33% of the total urban population ...  If the floating population is not as well covered in the life satisfaction surveys as their urban-born counterparts, then this negative impact is understated, and thus the full period trend is biased upward."

Why has satisfaction not flourished in China with the rise in GDP growth? Surely one reason is what some call the "aspirational treadmill:" the more you have, the more you want. But the reason emphasized by Easterlin's group is that "the high 1990 level of life satisfaction in China was consistent with the low unemployment rate and extensive social safety net prevailing at that time. Urban workers were essentially guaranteed life-time positions and associated benefits, including subsidized food, housing, health care, child care, and pensions, as well as jobs for grown
children ..." However, urban unemployment in China rose sharply from about 1990 into the early 2000s, but has fallen some since the mid-2000s. In addition, "Although incomes have increased for all income groups, China’s transition has been marked by a sharp increase in income inequality. This increasing income inequality is related to the growing urban–rural disparity in income, increased income differences in both urban and rural areas, and the significant increase of unemployment in urban areas associated with restructuring ..."

Intriguingly, the rise in income inequality in China is mirrored by greater inequality in reported life satisfaction. "In its transition, China has shifted from one of the most egalitarian countries in terms of distribution of life satisfaction to one of the least egalitarian. Life satisfaction has declined markedly in the lowest-income and least-educated segments of the population, while rising somewhat in the upper SES [socioeconomic status] stratum." For example, here's a graph that divides the population into thirds by income level. The figure shows what share of the population gave an answer from 7-10 on the World Values Survey satisfaction data. Notice that in 1990, all three income groups are clustered together. By 2007, they have separated out, with the highest income group remaining at about the same level, and the other groups declining in reported satisfaction--despite the fact that the incomes for all groups are much higher.

Taken to an extreme, the Easterlin paradox and these results from China might seem to suggest that economic growth is a waste of time. After all, economic growth doesn't seem to be making people more satisfied! But Easterlin would not make this argument, and it doesn't quite fit the survey results.
When answering on a scale of 1-10 or 1-5, most people will make a choice thinking about the present. They don't answer by thinking:  "Wow, I'm sure glad that I wasn't born a Roman slave 2000 years ago, and compared to that, I'm a 10 in satisfaction." Nor do they think: "Wow, compared to people who live 100 years from now, I'm living a short and deprived life, so I'm a 1 in satisfaction."

When people in China answered a "satisfaction" survey in 1990, they were not all that far removed in time from a period of brutal repression, and so it's not shocking to me that many in the lower and middle part of the income distribution told surveyers (who after all might have a government connection) that they were really quite satisfied. It's not just the economy that has grown in China since 1990; it's also the willingness and ability of many ordinary people to express dissatisfaction or discontent. I suspect that not many people in China would view their 1990 standard of living as similar or preferable to their current standard of living.


People do seem to answer satisfaction questions with some perspective on the rest of the world. For example, in a Spring 2008 article in my own Journal of Economic Perspectives, Angus Deaton presents evidence that if you look across the countries of the world in 2003, the level of satisfaction seems to rise steadily each time per capita GDP doubles.

It would be unwise to use survey data at different points in time, measured on scales that only offer the same limited range of choices, to argue that people do not receive greater satisfaction or happiness from economic growth. If I was choosing a 1970 standard of living in 1970, I might give it a similar numerical satisfaction score that I would give a 2012 standard of living in 2012--but that doesn't mean I would be equally happy in 2012 with a 1970 standard of living!

However, the results of the satisfaction surveys do highlight that when people are asked about their satisfaction, they take many factors into account along with average income levels: health, education, personal and political freedom, economic security, risk of unemployment, inequality, and others.When China allowed a greater degree of economic and political freedom, it unleashed an extraordinary rate of economic growth, but it also made created a public space for many other potential reasons for dissatisfaction. A record of past economic growth, even when exceptionally rapid, doesn't trump present concerns in people's minds--nor should it.

McWages Around the World

It's hard to compare wages in different countries, because the details of the job differ. A typical job in a manufacturing facility, for example, is a rather different experience in China, Germany, Michigan, or Brazil. But for about a decade, Orley Ashenfelter has been looking at one set of jobs that are extremely similar across countries--jobs at McDonald's restaurants. He discussed this research and a broader agenda of "Comparing Real Wage Rates" across countries in his Presidential Address last January to the American Economic Association meetings in Chicago. The talk has now been published in the April 2012 issue of the American Economic Review, which will be available to many academics through their library subscription. But the talk is also freely available to the public here as Working Paper #570 from the Princeton's Industrial Relations Section. 

How do we know that food preparation jobs at McDonald's are similar? Here's Ashenfelter:  

"There is a reason that McDonald’s products are similar.  These restaurants operate with a standardized protocol for employee work. Food ingredients are delivered to the restaurants and stored in coolers and freezers. The ingredients and food preparation system are specifically designed to differ very little from place to place. Although the skills necessary to handle contracts with suppliers or to manage and select employees may differ among restaurants, the basic food preparation work in each restaurant is highly standardized. Operations are monitored using the 600-page Operations and Training Manual, which covers every aspect of food preparation and includes precise time tables as well as color photographs. ... As a result of the standardization of both the product and the workers’ tasks, international comparisons of wages of McDonald’s crew members are free of interpretation problems stemming from differences in skill content or compensating wage differentials."

Ashenfelter has built up McWages data from about 60 countries. Here is a table of comparisons. The first column shows the hourly wage of a crew member at McDonald's, expressed in U.S. dollars (using the then-current exchange rate). The second column is the wage relative to the U.S. wage level, where the U.S. wage is 1.00. The third column is the price of a Big Mac in that country, again converted to U.S. dollars. And the fourth column is the McWage divided by the price of a Big Mac--as a rough-and-ready way of measuring the buying power of the wage.

Ashenfelter sums up this data, and I will put the last line in boldface type: "There are three obvious, dramatic conclusions that it is easy to draw from the comparison of wage rates in Table 3.  First, the developed countries, including the US, Canada, Japan, and Western Europe have quite similar wage rates, whether measured in dollars or in BMPH.   In these countries a worker earned between 2 and 3 Big Macs per hour of work, and with the exception of Western Europe with its highly regulated wage structure, earned around $7 an hour.  A second conclusion is that the vast majority of workers, including those in India, China, Latin America, and the Middle East earned about 10% as much as the workers in developed countries, although the BMPH comparison increases this ratio to about 15%, as would any purchasing-power-price adjustment.   Finally, workers in Russia, Eastern Europe, and South Africa face wage rates about 25 to 35% of those in the developed countries, although again the BMPH comparison increases this ratio somewhat.  In sum, the data in Table 3 provide transparent and credible evidence that workers doing the same tasks and producing the same output using identical technologies are paid vastly different wage rates."

In passing, it's interesting to note that McWage jobs pay so much more in western Europe than in the U.S., Canada and Japan. But let's pursue the highlighted theme: How can the same job with the same output and the same technology pay more in one country than in another? One part of the answer, of course, is that you can't hire someone in India or Sough Africa to make you a burger and fries for lunch. But at a deeper level, the higher McWages in high-income countries is not about the skill or human capital in those countries, but instead reflects that the entire economy is operating at a higher productivity level.
 
Here is an illustrative figure. The horizontal axis shows the "McWage ratio": that is, the U.S. McWage is equal to 1.00, and the McWages in all other countries are expressed in proportion. The vertical axis is "Hourly Output Ratio." This is measuring output per hour worked in the economy, again with the U.S. level set equal to 1.00, and the output per hour worked in all other countries expressed in proportion. The straight line at a 45-degree angle plots the points in which a country with, say, a McWage at 20% of the U.S. level also has output per hour worked at 20% of the U.S. level, a country with a McWage at 50% of the U.S. level also has output per hour worked at 50% of the U.S. level, and so on. 

The key lesson of the figure is that the differences in McWages across countries line up with the overall productivity differences across countries. The main exceptions, in the upper right-hand part of the diagram, are countries where the McWage is above U.S. levels but output-per-hour for the economy as a whole is below U.S. levels: New Zealand, Japan, Italy, Germany. These are countries with minimum wage laws that push up the McWage. 

Ashenfelter emphasizes in his remarks how real wages can be used to assess and compare the living standards of workers. I would add that these measures show that the most important factor determining wages for most of us is not our personal skills and human capital, or our effort and initiative, but whether we are using those skills and human capital in the context of a a high-productivity or a low-productivity economy.

Ignorance as Asset and Strategic Outcome

The February 2012 issue of Economy and Society is a special issue focused on a theme of "Strategic unknowns: towards a sociology of ignorance." The opening essay with this title, by Linsey McGoey, is freely available here. Many academics will have access to the rest of the issue through their library subscriptions.

The central theme of the issue is that ambiguity and ignorance are not just the absence of knowledge, waiting to be illuminated by facts and disclosure. Instead, ambiguity and ignorance are in certain situations the preferred strategic outcome. McGoey writes (citations omitted): "Ignorance is knowledge: that is the starting premise and impetus of the following collection of papers. Together, they contribute to a small but growing literature which explores how different forms of strategic ignorance and social unknowing help both to maintain and to disrupt social and political orders, allowing both governors and the governed to deny awareness of things it is not in their interest to acknowledge ..."

Many of the examples are sociological in nature, but others are based in economic and policy situations. For example, consider a number of situations that have to do with a policy response to risky situations: the risk that smoking causes cancer, the risk that growing carbon emissions will lead to climate change, the risk of future terrorist actions (and whether invading certain countries will increase or reduce those risks), and the risk of fluctuations in fluctuations in financial markets. McGoey writes:

"Within the game of predicting risk, one often wins regardless of whether risks materialize or not. If a predicted threat fails to emerge, the identification of the threat is credited for deterring it. If a predicted threat does emerge, authorities are commended for their foresight. If an unpredicted threat appears, authorities have a right to call for more resources to combat their own earlier ignorance. ‘The beauty of a futuristic vision, of course, is that it does not have to be true’, writes Kaushik Sunder Rajan (2006, p. 121) in a study of the way expectations surrounding new biotechnologies help to create funding opportunities and foster faith in the technology regardless of whether expectations prove true or not. In fact, expectations are often particularly fruitful when they fail to materialize, for more hope and hype are needed to remedy thwarted expectations. Attention to the resilience of risks the way that claims of risk often feed on their own inaccuracy helps to highlight the value of conditionality for those in political authority."

One of the essays in the volume, by William Davies and Linsey McGoey, applies this framework to thinking about the recent financial crisis. They point out that many financial professionals begin from the starting point that risk and uncertainty are huge problems, and thus one needs their high-priced help to address these issues. In this way, claims of ambiguity and ignorance are an asset for the finance industry. If the investment go well, then the financial professionals claim credit for steering successfully through these oceans of uncertainty. But when investments and decision go badly, as in the Great Recession, they claim absolution for their decisions by reiterating just how ambiguous and unclear the financial markets are, and how no one could have really known what was going to  happen. And somehow, this just proves that their expertise is more needed than ever. They write: "We examine the usefulness of the failure or refusal to act on warning signs, regardless of the motivations why. We look at the double value of ignorance: the ways that social silence surrounding unsettling facts enabled profitable activities to endure despite unease about their implications and, second, the way earlier silences are then harnessed and mobilized to absolve earlier inaction.

In another essay, Jacqueline Best applies these ideas in the context of the World Bank's "good governance agenda" and the IMF's "conditionality policy." She writes: "Both policies have been ambiguously defined throughout their history, enabling them to be interpreted and applied in different ways. This ambiguity has facilitated the gradual expansion of the scope of the policies. ... Actors at both the IMF and the World Bank were not only aware of the central role of ambiguity in their policies, but were also ambivalent about it.  ... Finally, although staff and directors at both institutions may have been ambivalent about the role of ambiguity in these policies, they ultimately ensured that ambiguities persisted and even proliferated."  Best also notes that ambiguity is hard to control, and can lead to unintended consequences. 

In yet another essay, Steve Rayner write about "Uncomfortable knowledge: the social construction of ignorance in science and environmental policy discourses." He writes: "My interest is therefore in how information is kept out rather than kept in and my approach is to treat ignorance as a necessary social achievement rather than a a simple background failure to acquire, store, and retrieve knowledge." Rayner writes: "An example of clumsy or incompletely theorized arrangements is the implicit consensus on US nuclear energy policy that emerged in the 1980s and persisted for the best part of three decades. Despite the complete absence of any Act of Congress or Presidential Order, it was implicitly accepted by government, industry, and environmental NGOs that the US would continue to support nuclear R&D while operating an informal moratorium on the addition of new nuclear generating capacity. All of the parties agreed to this, but for various reasons, all had a stake in not acknowledging the existence of an settlement."

One might add that many environmental laws and other regulatory policies are chock-full of ambiguous language, which gives regulators the ability to interpret these rules as tough-minded while also giving potential offenders the possibility of saying that they had no way of knowing the rules would be applied in this way. Rayner also offers a nicely provocative claim about tendencies to dismiss and deny in the context of warnings about climate change: "It seems odd that climate science has been held to a `platinum standard' of precision and reliability that goes well beyond anything that is normally required to make significant decisions in either the public or private sectors. Governments have recently gone to war based on much lower-quality intelligence than that which science offers us about climate change. Similar firms embark on product launches and mergers on the bases of much lower-quality information."  


Academic research of course often uses of a feigned ignorance to generate a greater persuasive effect. The title of a research paper is often written in the form of a question, and the theory and data are often presented as if the author was a Solomonic figure encountering this material for the first time, guided only by a disinterested pursuit of Truth (with a capital T). The implications for reputation of past past work, or its political implications, are shunted off to the side. Research would have less persuasive effect if it started off by saying, "I've been hammering on this same conclusion for 25 years now, and I find pretty much exactly the same result every time I look at any data set from any time or place--and by the way, this conclusion also supports the political outcomes I prefer."

One of many implications of thinking about ignorance and ambiguity as assets and as strategic behavior is that it highlights that many economic actors and policy-makers have strong incentives to promote both their own ignorance, and more broadly, the idea that ambiguity makes true knowledge impossible. Ignorance can be a power grab, and the basis for a job, and a get-out-of-jail-free card.

Why Does the U.S. Spend More on Health Care than Other Countries?

Everyone knows that the U.S. spends far more on health care than other countries, but do you know how much more? In 2009, the U.S. spent 17.4% of GDP on health care (using OECD data). The closest contenders are Netherlands (12% of GDP), France (11.8%), Germany (11.6%), Denmark (11.5%), and Canada (11.4%). The U.S. has higher per capita GDP than these countries, so the gap in absolute spending is even higher. In 2009, the U.S. spent $7,960 per person on health care, and the closest contenders were Switzerland ($5,144 per person) and Netherlands ($4,914).

When I hear people argue that the U.S. should follow the path of the UK health care system, I sometimes find myself thinking: "You mean that U.S. health care spending per person should be slashed by 56%, from $7,960 per person to $3,487 per person? Really?"

What accounts for these differences in health care spending across countries? David Squires assembles some of the evidence in "Explaining High Health Care Spending in the United States: An International Comparison of Supply, Utilization, Prices and Quality," a May 2012 "issue brief" written for the Commonwealth Fund. I ran across it here at Larry Willmore's Thought du Jour blog.   I'll also contrast and compare it with a paper by David M. Cutler and Dan P. Ly, "The (Paper)Work of Medicine: Understanding International Medical Costs," which appeared in the Spring 2011 issue of my own Journal of Economic Perspectives. For readability, footnotes and references to exhibits are omitted from the quotations below.

 Higher U.S. health care spending is not because Americans on average are notably less healthy.
As Squires sums up: "U.S. has smaller elderly population and fewer smokers, but higher obesity rates. ...  Higher rates of obesity undoubtedly inflate health spending; one study estimates the medical costs attributable to obesity in the U.S. reached almost 10 percent of all medical spending in 2008. However, the younger population and lower rates of smoking likely have an opposite effect, reducing U.S. health care spending relative to most other countries."

Higher U.S. health care spending is not because the U.S. has more doctors or hospital beds.
"There were 2.4 physicians per 1,000 population in the U.S. in 2009, fewer than in all other study countries except Japan. Likewise, patients had fewer doctor consultations in the U.S. (3.9 per capita)
than in any other country except Sweden. Hospital supply and use showed similar trends, with the U.S. having fewer hospital beds (2.7 per 1,000 population), shorter lengths of stay for acute care (5.4
days), and fewer discharges (131 per 1,000 population) than the OECD median ..."



Prices for brand-name drugs are much higher in the U.S., but generics are cheaper.
Squires writes: "[P]rices for the 30 most-commonly prescribed drugs are one-third higher than in Canada and Germany, and more than double the prices in Australia, France, Netherlands, New Zealand, and the U.K. Notably, prices for generic drugs are lower in the U.S. than in these other countries, whereas prices for brand-name drugs are much higher."

Cutler and Ly confirm this general pattern, but also put the potential cost savings in perspective: "However, because pharmaceuticals are only about 10 percent of U.S. healthcare spending, the overall amount that could be saved by moving to U.S. government monopsony purchasing of drugs
is relatively small—perhaps 20 to 30 percent of pharmaceutical spending, or 2 to 3 percent of total medical costs. These cost savings also would have to be weighed against the possibility of reduced incentives for investment and innovation in the pharmaceutical industry. The dollar amount of excess pharmaceutical payments in the United States is approximately the total amount of pharmaceutical company research and development (R&D)."


U.S. doctors are paid more, but they also live in an economy with a more unequal distribution of wages.
Squires writes: "U.S. primary care physicians generally receive higher fees for office visits and orthopedic physicians receive higher fees for hip replacements than in Australia, Canada, France, Germany, and the U.K. ... U.S. primary care doctors ($186,582) and particularly orthopedic doctors ($442,450) earned greater income than in the other five countries ..."

Cutler and Ly confirm: "The average U.S. specialist physician earns $230,000 annually—
78 percent above the average in other countries ... . Primary care physicians earn less (they earn $161,000 on average), but the same percentage more than their peers in other countries. ... If we reduced all physician incomes in the United States to match the international ratio of physicians’ incomes to per capita GDP, U.S. healthcare spending would be lower by roughly 2 percent.However, these seemingly high salaries for U.S. physicians appear less high in the context of the broader income distribution." Cutler and Ly go on to point out that high-compensation workers in the U.S. economy earn more than their international counterparts in just about every profession--after all, that's part of what it means to say that the U.S. has a less equal distribution of income.

Some medical device technologies like scanning are more widely used in the U.S; some like hip replacements are not.
"In 2009, the U.S., along with Germany, performed the most knee replacements (213 per 100,000
population) among the study countries, and 75 percent more knee replacements than the OECD median (122 per 100,000 population). However, the U.S. performed barely more hip replacements than the OECD median, and significantly less than several of the other study countries ..."

"Relative to the other study countries where data were available, there were an above-average
number of magnetic resonance imaging (MRI) machines (25.9 per million population), computed
tomography (CT) scanners (34.3 per million), positron emission tomography (PET) scanners (3.1 per million), and mammographs (40.2 per million) in the U.S. in 2009. Utilization of imaging was also highest in the U.S., with 91.2 MRI exams and 227.9 CT exams per 1,000 population. MRI and CT devices were most prevalent in Japan, though no utilization data were available for that country. ... [T]he U.S. commercial average diagnostic imaging fees ($1,080 for an MRI and $510 for a CT exam) are far higher than what is charged in almost all of the other countries ..."

The U.S. does a relatively poor job of managing chronic disease.
Squires writes: "[Consider] rates of potentially preventable mortality due to asthma (for those between ages 5 and 39) and lower-extremity amputations due to diabetes per 100,000 population. On both measures, the U.S. had among the highest rates, suggesting a failure to effectively manage these chronic conditions that make up an increasing share of the disease burden."

Many chronic diseases share the general property that if they are well-managed every single day, with a combination of drugs, lifestyle, and certain kinds of monitoring of physical conditions, it is possible to reduce the need for enormously costly episodes of hospitalization. As the Centers for Disease Control puts it: "Chronic diseases—such as heart disease, cancer, and diabetes—are the leading causes of death and disability in the United States. Chronic diseases account for 70% of all deaths in the U.S., which is 1.7 million each year. These diseases also cause major limitations in daily living for almost 1 out of 10 Americans ...."

Prices for hospital stays are substantially higher in the U.S.
Squires points out: "[H]ospital stays in the U.S. were far more expensive than in the other study countries, exceeding $18,000 per discharge compared with less than $10,000 in Sweden, Australia, New Zealand, France, and Germany." And remember, these higher costs per hospital stay happen even though the stays themselves are on average shorter in the U.S.

The tougher question is to what extent these higher costs per hospital stay reflect a larger quantity of concentrated and effective high-tech care being provided, and to what extent its just a matter of higher prices. The evidence here is mixed. It does appear that for some conditions, Americans receive more hospital care. Cutler and Ly write:  Americans also receive more-intensive care than do Canadians. While the population-adjusted hospital admission rates are about the same in the two countries, additional procedures are provided to those with the same diagnosis in the United States. For example, people with a heart attack in the United States are twice as likely to receive bypass surgery or angioplasty than are similar people in Canada." When it comes to cancer survival rates, Squires points out: "The U.S. had the highest survival rates among the study countries for breast cancer (89%) and, along with Norway, for colorectal cancer (65%)."

On the other side, the more aggressive use of heart surgery in the U.S. as compared to Canada doesn't seem to mean better health outcomes; instead, it reflects the existence of more heart-surgery facilities. Cutler and Ly: "  On one side, the greater use of intensive therapies after a heart attack in the United States compared to Canada is not associated with improved mortality, though morbidity is more diffifult to determine. Similarly, a recent study concluded that there was no systematic difference in outcomes in favor of the United States over Canada; if anything, Canadians had better outcomes in most circumstances ... [T]he province of Ontario has 11 open-heart surgery facilities, while the state of Pennsylvania, with roughly the same population as Ontario, has more than five times the number of heart surgery facilities. California is three times larger in population but has 10 times the number of heart surgery facilities. Given this difference in the number of facilities, it is simply impossible for physicians in Ontario to perform as many open heart surgery operations as those in Pennsylvania or California."

Also, not all cancer survival rates are better in the U.S. Squires writes: "However, at 64 percent, the survival rate for cervical cancer in the U.S. was worse than the OECD median (66%), and well below the 78 percent survival rate in Norway—indicating significant room for improvement."

Administrative costs of health care are much higher in the U.S.
Squires doesn't mention this point, but it is a main emphasis for Cutler and Ly. They write:

"[T]the U.S. healthcare system is in great need of administrative simplification. There are few other areas of the U.S. economy where waste is so apparent and the possibility of savings is so tangible. ... Perhaps the most troubling difference between the U.S. and Canadian healthcare systems is the differential amount spent on administration. For every office-based physician in the United States, there are 2.2 administrative workers. That exceeds the number of nurses, clinical assistants, and technical staff put together. One large physician group in the United States estimates that it spends 12 percent of revenue collected just collecting revenue. Canada, by contrast, has only half as many administrative workers per office-based physician.  The situation is no better in hospitals. In the United States, there are 1.5 administrative personnel per hospital bed, compared to 1.1 in Canada. Duke University Hospital, for example, has 900 hospital beds and 1,300 billing clerks. On top of this are the administrative workers in health insurance. Health insurance administration is 12 percent of premiums in the United States and less than half that in Canada.

"International comparisons of medical care occupations are difficult, but they suggest that the United States has more administrative personnel than other countries do. ... [T]he United States has 25 percent more healthcare administrators than the United Kingdom, 165 percent more than the Netherlands, and 215 percent more than Germany. The number of clerks of all forms (including data entry clerks) is much higher in the United States as well."

"What are all these administrative personnel doing? ... One part is credentialing—receiving permission to practice medicine in a particular hospital or for a particular health plan. The average physician submits 18 credentialing applications annually—each insurer, hospital, ambulatory surgery facility, and the like, requires a different one—consuming 70 minutes of staff time and 11 minutes of physician time per application. Verifying eligibility for services is also costly. Insurance information must be verified for 20 to 30 patients daily, including three or four patients for whom verification must be sought orally. Because people change insurance plans frequently and the cost-sharing they are charged varies with plan and with past utilization (for example, how much of the deductible have they spent?), the determination of what to charge a patient is especially difficult. ... Finally, significant time is spent on billing and payment collection. On average, about three claims are denied per physician per week and need to be rebilled. ...  Three-quarters of denied bills are ultimately paid, but the administrative cost of securing the payment is very high. Provider groups in the United States employ 770 full-time equivalent workers per $1 billion collected, compared to an average in other U.S. industries of about 100. By all indications, the administrative burden is rising over time as insurance policies have become more complex, while the technology of administration has not kept pace."

Conclusion

The question of why the U.S. spends more than 50% more per person on health care than the next highest countries (Switzerland and Netherlands), and more than double per person what many other countries spend, may never have a simple answer. Still, the main ingredients of an answer are becoming more clear. The U.S. spends vastly more on hospitalization and acute care, with a substantial share of that going to high-tech procedures like surgery and imaging. The U.S. does a poor job of managing chronic conditions, which then lead to episodes of costly hospitalization. The U.S. also seems to spend vastly more on administration and paperwork, with much of that related to credentialing, documenting, and billing--which is again a particular important issue in hospitals. Any honest effort to come to grips with high and rising U.S. health care costs will have to tackle these factors head-on. 

Occupational Licensing and Low-Income Jobs

Pretty much everything I know about the economics of occupational licensing I learned from Morris Kleiner, a colleague from the days when I was based at the Humphrey School at the University of Minnesota. Morrie lays out many of the issues here in a Fall 2000 article in my own Journal of Economic Perspectives, as well as in his  2006 book, Licensing Occupations: Ensuring Quality or Restricting Competition?

He points out that nearly one-third of the U.S. labor force works in jobs where some form of government license is a requirement. Some of the largest occupations that require licenses include teachers, nurses, engineers, accountants, and lawyers.  Occupational licensing poses a potential tradeoff: on one side, requiring licenses offers a promise of a reliably high quality of service; on the other side, requiring licenses is a barrier to entry that tends to reduce the quantity of jobs in that occupation but increase the wage. Kleiner and others investigate this subject by looking at differences in licensing requirements for a certain occupation across states, and searching for evidence of wage and quality differences. A typical finding is that the wage differences are readily perceptible, but the quality differences are not. Licensing is distinguishable from certification: with certification, you are free to hire someone who doesn't possess the certification if you like, but with licensing, hiring someone without the license is illegal. As an example, travel agents and mechanics are often certified, but they are typically not licensed.

Dick M. Carpenter II, Ph.D., Lisa Knepper, Angela C. Erickson and John K. Ross focus on documenting differences between states in 102 of the job categories counted by the Bureau of Labor Statistics that requires a license in at least one state and that pay below-average wages. They report the results in License to Work: A National Study of Burdens from Occupational Licensing, a report from the Institute for Justice. They make the case that many of these occupational rule are more about limiting competition than about quality of service in an indirect way: they point out that licensing rules about fees, training, exams, minimum age, and minimum schooling vary enormously across states, with no particular evidence that reliability or safety are worse in states with lesser or no licensing requirements. The report goes into state-by-state and occupation-by-occupation detail, but here are some summary comments: 



"The need to license any number of the occupations in this sample defies common sense. A short list would include interior designers, shampooers, florists, upholsterers, home entertainment installers, funeral attendants, auctioneers and interpreters for the deaf. Most of these occupations are licensed in just a handful of states; interpreters are licensed in only 16 states, while auctioneers are licensed in 33. If, as licensure proponents often claim, a license is required to protect the public health and safety, one would expect more consistency. For example, only five states require licenses for shampooers, but it is highly unlikely that conditions in those five states are any different ..."


"Quite literally, EMTs [emergency medical technicians] hold lives in their hands, yet 66 other occupations have greater average licensure burdens than EMTs. This includes interior designers, barbers and cosmetologists, manicurists and a host of contractor designations. By way of perspective, the average cosmetologist spends 372 days in training; the average EMT a mere 33."



"Licensure irrationalities are doubly evident in the inconsistencies by burden across states. Looking again at manicurists, while 10 states require four months or more of training, Alaska demands only about three days and Iowa about nine days. It seems unlikely that aspiring manicurists in Alabama (163 days) and Oregon (140 days) truly need so much more time in training. But manicurists are not alone. The education and experience requirements for animal trainers range from zero to almost 1,100 days, or three years. And for vegetation pesticide handlers, training obligations range from zero to 1,460 days, or four years, with fees up to $350. This high degree of variation is prevalent throughout
the occupations. Thirty-nine of them have differences of more than 1,000 days between the minimum and maximum number of days required for education and experience. And another 23 occupations have differences of more than 700 days."


"Finally, irrationalities are particularly notable when few states license an occupation but do so onerously. One clear example is interior design, the most difficult of the 102 occupations to enter, yet licensed in only three states and D.C. Another is social service assistants, the fourth most difficult occupation to enter. It requires nearly three-and-a-half years of training but is only licensed in six states and D.C. Dietetic technicians must spend 800 days in education and training, making for the eighth most burdensome requirements, but they are licensed in only three states. Home entertainment installers must have about eight months of training on average, but only in three states. The seven states that license tree trimmers require, on average, more than a year of training."



"The 102 occupational licenses studied require of aspiring workers, on average, $209 in fees, one exam and about nine months of education and training. ·· Thirty-five occupations require more
than a year of education and training, on average, and another 32 require three to nine months. At least one exam is required for 79 of the occupations. ... Particularly noteworthy is the percentage of low- and middle-income workers with less than a high school diploma—15.7 percent. As documented below, a number of the 102 occupations studied require the completion of at least 12th grade, a requirement that effectively bans a substantial number of people from those occupations."

"[S]even of the 102 occupations studied are licensed in all 50 states and the District of Columbia:
pest control applicator, vegetation pesticide handler, cosmetologist, EMT, truck driver, school bus driver and city bus driver. Another eight occupations are licensed in 40 to 50 states. Thus, the vast majority of these occupations are licensed in fewer than 40 states, and five are licensed in only
one state each: florist, forest worker, fire sprinkler system tester, conveyor operator and non-contractor pipelayer. On average, the occupations on this list are licensed in about 22 states."

My own guess is that the politics of passing state-level occupational licensing laws is driven by three factors: 1) lobbying by those who already work in the occupation to limit competition; 2) passing laws in response to wildly unrepresentative anecdotes of terrible or dangerous service; and 3) the tendency when setting standards to feel like more is better. But in a U.S. economy which is hurting for job creation, especially jobs for low-income workers, states should be seriously rethinking many of their occupational licensing rules. Many would be better-replaced with lower standards, certification rather than licenses, or even no licenses at all. 

Teen Pregnancy: What Causes What?

Here is a classic problem of cause and effect. Teenagers who give birth are more likely to be from households with lower income levels. Also, teenagers who give tend to end up later in life in households with lower income levels. But does the lower income level cause teens to be more likely to give birth? Or does giving birth cause as a teen cause that woman to be more likely to end up in a lower-income household? How can one untangle cause and effect? Melissa S. Kearney and Phillip B. Levine tackle these questions in "Why is the Teen Birth Rate in the United States So High and Why Does It Matter?" which appears in the Spring 2012 issue of my own Journal of Economic Perspectives. They have lots of interesting comments to make about variation in teen birthrates across states and countries. Here, I'll focus on their analysis of the cause and effect question, which surprised me and offers a nice example of  how economist try to disentangle these sorts of issues.

"Our reading of the totality of evidence leads us to conclude that being on a low economic trajectory in life leads many teenage girls to have children while they are young and unmarried and that poor outcomes seen later in life (relative to teens who do not have children) are simply the continuation of the original low economic trajectory. That is, teen childbearing is explained by the low economic trajectory but is not an additional cause of later difficulties in life. Surprisingly, teen birth itself
does not appear to have much direct economic consequence."


Conceptually, how would one tell whether giving birth as a teenager is a cause of lower future economic prospects? Just comparing life outcomes for teenage girls who give birth and those who don't will give you a correlation, but not causation.  "A comparison of the outcomes of women who did and who did not give birth as teens is inherently biased by selection effects: teenage girls who “select” into becoming pregnant and subsequently giving birth (as opposed to choosing abortion) are different in terms of their background characteristics and potential future outcomes than teenage girls who delay childbearing." The problem is made more difficult because some of the background characteristics may be measurable in the data (like family income level, or ethnicity, or if it's a single-parent family) but many other characteristics are not available in the data (like the personality traits of the teenage girl or the values lived by the family).

 In an ideal experiment, one might want a research design in which a random sample of teenagers becomes pregnant and gives birth, and then you could track the outcomes. Of course, randomized pregnancy is an impractical research design! But here are four approaches used by clever economists to disentangle this question of cause and effect. 


A within-family approach. Look at life outcomes for sisters who give birth at different ages. The result of this kind of study is "once background characteristics are controlled for, the differences are quite modest. Furthermore, even these modest differences likely overstate the costs of teen childbearing, since the sister who gives birth as a teen is likely to be “negatively” selected compared
to her sister who does not."

Miscarriages.  Of those teens who become pregnant, some will suffer miscarriages. Compare women who are similar in measured characteristics of family background, but some of whom gave birth as teenagers while others had a miscarriage. It turns out that their life outcomes look quite similar: that is, giving birth as a teenager doesn't appear to cause any additional decline in later life outcomes.

Age at first menstruation. Girls who menstruate earlier are at greater risk of becoming pregnant as teenagers. One can use a statistical approach to look at two groups of women who are similar in measured characteristics of family background, but where one group has a higher pregnancy rate because they began their menstrual cycle earlier. However, the life outcomes for these groups look quite similar; is not correlated with lower life outcomes: that is, a random chance of being more likely to give birth as a teenager (because of an earlier age of first menstruation) doesn't appear to cause any additional decline in later life outcomes.


 Propensity scores. Look at girls within a certain school, so that they live in more-or-less the same neighborhood. Using the available data, develop a "propensity score" that measures how likely a girl is to give birth as a teenager. Then compare the life outcomes for girls with similar propensity scores, some of whom gave birth and some of whom did not. There doesn't seem to be a difference in life outcomes, again suggesting that giving birth as a teenager doesn't much alter other life outcomes. 

Kearney and Levine sum up the evidence on cause and effect this way: "Taken as a whole, previous research has had considerable difficulty finding much evidence in support of the claim that teen childbearing has a causal impact on mothers and their children. Instead, at least a substantial majority of the observed correlation between teen childbearing and inferior outcomes is the result of underlying differences between those who give birth as a teen and those who do not."

Kearney and Levine also offer an unexpected (to me) perspective on policies to reduce teen pregnancy:

"Moreover, no silver bullet such as expanding access to contraception or abstinence education will solve this particular social problem. Our view is that teen childbearing is so high in the United States because of underlying social and economic problems. It reflects a decision among a set of girls to “drop-out” of the economic mainstream; they choose nonmarital motherhood at a young age instead of investing in their own economic progress because they feel they have little chance of advancement. This thesis suggests that to address teen childbearing in America will require addressing some difficult social problems: in particular, the perceived and actual lack of economic opportunity among those at the bottom of the economic ladder."

The statement about teenage girls "choosing" nonmarital motherhood should be understood not as a claim that all pregnant 15 year-olds carefully considered their life options and decided on pregnancy!  Instead, the economists' view of choice is that we all make groups of choices every day--say, choices about exercise and calories consumed--that make certain outcomes more likely. Decisions that are not well-considered, or that raise the risk of undesired side effects, still have a large ingredient of choice. For example, we typically view those who drive drunk as having made a "choice."

The cause-and-effect evidence here suggests that for many women who give birth as teenagers, their life outcomes like level of education achieved, income, employment, and chance of marriage are already so constrained that they are not made worse off by having a child as a teenager. Encouragement about contraception or abstinence can help reduce teen pregnancy on the margin. But what many teen girls from low socioeconomic status backgrounds need is a reduced prospect of marginalization, and a greater chance for personal and economic advancement.

On the Job for 100 Issues of JEP

I was hired by Joseph Stiglitz 26 years ago to start a new economics journal, the Journal of Economic Perspectives. It took us a year from the starting line to mailing our first issue in the mail, but the Spring 2012 issue, now available on-line, is the 100th issue. Like all issues of JEP back to 1994, it is freely available to all, courtesy of the American Economic Association. The first three articles are about the journal: one by current editor David Autor on the effect of the journal within the economics profession, one by Joe Stiglitz remembering the early years and commenting on how the journal has evolved, and one by me called "From the Desk of the Managing Editor."

Here are the two opening paragraphs and the closing paragraph of my essay:

"Editing isn’t “teaching” and it isn’t “research,” so in the holy trinity of academic responsibilities it is apparently bunched with faculty committees, student advising, and talks to the local Kiwanis club as part of “service.” Yet for many economists, editing seems to loom larger in their professional lives. After all, EconLit indexes more than 750 academic journals of economics, which require an ever-shifting group of editors, co-editors, and advisory boards to function. Roughly one-third of the books in the annotated listings at the back of each issue of the Journal of Economic Literature are edited volumes.

Editors are gatekeepers, and editors are road-blocks—or perhaps these are essentially the same task. Editors shape “the literature,” both what and who is included and how it is presented. I’ve come to believe that “editing” is no more susceptible to a compact single defifi nition than “manufacturing” or “services.” But here is one take on the enterprise of editing from someone who has been sitting in the Managing Editor’s chair for all 100 issues of the Journal of Economic Perspectives since before the first issue of the journal mailed in Summer 1987. ...

My job as Managing Editor of JEP has been a pride and a pleasure for these last 25 years. It’s consistently interesting work: after all, my job is to do close readings of the highly varied work of a succession of prominent economists who are trying to explain their thinking—and then to ask them questions until they explain it all to me! Editing an academic journal also offers the psychic frisson of leaving something behind: 100 issues and counting, to be precise. When I visit another college or university, I sometimes walk through the periodical stacks just to see JEP on the shelf. Running an academic journal for a long time offers a pleasing sense of place within the discipline of economics, spinning a web of personal contacts from the up-and-comers to the well-established in academic institutions around the world. Some of my friends refer to my job at the journal as “the guy who gets thanked” at the end of articles. There are worse epitaphs."