• Social determinants of health and Kenneth Arrow

    Commenting on Adrianna’s post, Sean Parnell drew our attention to the second paragraph of the seminal paper by Kenneth Arrow:

    It should be noted that the subject is the medical-care industry, not health. The causal factors in health are many, and the provision of medical care is only one. Particularly at low levels of income, other commodities such as nutrition, shelter, clothing, and sanitation may be much more significant. It is the complex of services that center about the physician, private and group practice, hospitals, and public health, which I propose to discuss.

    If Arrow’s paper is not the most cited one in health economics, it surely must be close. (Google Scholar counts nearly 5,700 citations in the scholarly literature alone.) Indeed, the paper is often pointed to as the founding document of the discipline. It is, therefore, significant that Arrow was explicit in not addressing social determinants of health, while at the same time noting that they may be more important for low income individuals than aspects of the provision of health care or, I will add, the financing thereof.

    Adrianna is right that we wonks have fetishized health care financing to the detriment of health care provision and, even more so, other areas of social policy likely more determinative for health of poor Americans. Surely there are–and if not, there should be–politically viable proposals that would or could more effectively address the health of poor Americans than a tweak or two of the Affordable Care Act. In your view, what are they?

    Comments open until one week from the date of this post.


    Comments closed
    • In my view the solution is to write insurance contracts so they are strongly incentivized to improve health.
      See the third letter here: http://www.nejm.org/doi/full/10.1056/NEJMc1306867?query=TOC

      • As long as the majority of Americans gets employer sponsored insurance and the average American changes jobs every 5 years, it will be hard to create incentives for insurance companies to invest in the long term health of their patients.

        Individual insurance has always been a stop-gap abandoned the minute someone gets employer-sponsored. As it stands now, the ACA doesn’t do much to change that.

        Short-termism is a feature of the structure of our insurance system.

    • Two major programs trying to identify such policies:

      Community Transformation Grants (under the Prevention and Public Health Fund, part of the ACA) really get at the policy environment around public health/chronic disease prevention and control. Many of our community’s CTG-funded activities are around the food environment for low-income families, for example. (http://www.cdc.gov/nccdphp/dch/programs/communitytransformation/)

      RWJF Roadmaps grants are even more focused on social determinants and seek to fund initiatives around social and economic policies to improve health. (http://www.rwjf.org/en/research-publications/find-rwjf-research/2013/12/stories-from-community-grants.html)

    • There’s a paper I found somewhere (perhaps on this blog?) showing that even relatively small cash transfers to poor pregnant women substantially improved several standard metrics for the health of the newborns (probability of low birth-weight fell, for example). Moreover, since most papers of this type typically ignore the much larger inter-generational benefits of these types of outcomes (I have in mind Currie and Moretti: http://www.nber.org/papers/w9360), that suggests pure redistribution of income could possibly be one of most effective ways to improve public health.

    • Education has a causal link with health. Diet is hugely important so sugar/fat taxes (with rebates) would work. On the comparatively more important mental health side UPenn has been doing a lot of good work on teaching mental health hygiene to kids. Those programs could be expanded in schools. Subsidizing driving less could have a big impact on how much people exercise, esp. in the long run as people move to denser areas.

      Each of these would have orders of magnitude bigger impacts on health per dollar spent compared to health insurance expansion.

      • There was a release yesterday from a research center in Richmond, VA about the link between education and health: http://bit.ly/19eJP9T There is a brief and a short video about how the impact of education on health has increased over the past 2-3 decades. Research-wise, there is nothing here that most readers of this blog don’t already know, but it is pretty creative and contains a lot of detail on the topic. The director of the center that produced this project was the chair of the IOM panel that produced the “American Health Disadvantage” report that Austin cited in an earlier post.

    • If we’re talking about health specifically, driving smoking rates towards zero would be an enormous public-health boon, especially for the poor, who have a higher prevalence of smoking (see CDC here: http://www.cdc.gov/mmwr/preview/mmwrhtml/mm6144a2.htm ). There are a number of interventions to do so (sin taxes, awareness campaigns, cessation outreach, packaging regulations, etc.) Probably the single best thing we can do, with a host of literature to back it up.

      Beyond that, my sense is that addressing other environmental factors would be my next bet (a la “hotspotting”), but I don’t know of any great research to back that up — it’s a hunch, on the same level as my hunch that ACOs are a good way to reform payment models.

      Finally, my experiences around Federally Qualified Health Centers (expanded in the ACA, so I’m cheating) suggest that poor patients seem to do better in those models. That’s anecdotal, but I’d appreciate any pointers to good research on their quality/cost.

    • Much is made of the fragmentation in health care delivery and the need to make it less so, but fragmentation in health care finance is totally ignored. Or even encouraged. Financing health care differently for different groups (the poor, seniors, children, veterans, the gainfully employed, et al.) guarantees different delivery for the different groups, as each vies for preferential treatment. Unless and until everybody is subject to the same system of health care finance, large gaps will continue to exist in health care delivery. Of course, single-payer comes to mind, but it’s not the only alternative. There’s the suggestion that private insurance can be used to modify behavior, as by charging higher premiums for smoking (part of ACA) or for being over-weight (not part of ACA), but the suggestion is aimed at those in the individual insurance market not the group market – the fat, sedentary guy in accounting pays the same for his group insurance as the lean, athletic gal in sales. This isn’t to dismiss techniques for behavior modification, but to suggest that everybody must be subject to the same rules or gaps in health care delivery will continue to grow.

    • Well, to address the specifics of the research in Adriana’s post concerning the link between end-of-month food budgets and hypoglycemia, there’s two possible things that might help the situation:

      1. Change the funds distribution schedule from monthly to weekly or bi-weekly for food stamps. I’d think having 2 days at the end of every 14-day cycle where food is stretched vs. a 4-day period at the end of every month would at least limit some of the cumulative impact of a lack of appropriate food.

      2. Better food preparation/shopping training for people on welfare. I won’t recount the tales of food stamp recipients buying filet mignon, I’l just say there are relatively inexpensive foods that can be made from scratch that are much cheaper than prepared foods, allowing food budgets to last longer. So pancakes made from scratch > pancake mix > frozen Eggo waffles.

      There’s some larger issues as well that might help. For example, allowing WalMart to open stores in urban areas where they’re currently locked out. Being able to buy milk at $4 a gallon vs. $6 a gallon, or egs at $2/dozen instead of $3/dozen, would obviously help with the whole ‘food desert’ issue and allow low-income families to access lower-cost food.

      Just about anything that drives up the cost of food artificially could be looked at in similar lights – tariffs, ethanol mandates, energy restrictions, GMO labeling, etc. That’s not to say these policies don’t have their own merits, but I think one of the negatives that has to be acknowledged is that they can drive up food costs, which is obviously going to have an impact on how well the poor are able to get access to affordable food.

      • Of course, if we were really worried about “the poor” not having enough to eat at the end of the month, we could also just give them money, so they would be less poor.

        Rather than thinking about how we can “encourage” poor people — who are already over-burdened with a variety of stressors, tend to be chronically short of time, etc. — to make their own pancakes from scratch, we could just give them enough money that this isn’t an issue.

        Rather than trying to figure out ways to make food as cheap as possible, we could just give the poor enough money to buy more expensive food, and let food costs reflect the real costs of reasonable, sustainable production.

        There really isn’t a good reason to let the poor in this country — let alone the children of the poor — suffer from food insecurity. Yes, as a relatively well paid academic with a flexible schedule, I have the time and energy to make my own pancakes from scratch, make good use of dried beans in my dinners, etc. It’s fun. But to think that people working multiple part-time minimum-wage jobs while trying to raise kids in a country without any commitment to affordable child care (or even sane school schedules) should also be responsible for finding the lowest cost, even if most inconvenient, food options? That strikes me as misguided.

        If you want the poor to make their own pancakes, I think the right policies to pursue are those that would give them enough money, via a high-enough paying job with short enough hours, that they can spend leisurely Saturday mornings making pancakes over the newspaper… Barring that, as a recommendation it just sounds sort of cruel.

        • I’ll second this.

          Universal preschool and raising the minimum wage would be a start. Economic stimulus and a more progressive tax system would be nice, too.

          The problem, of course, is that this would take a generation.

          The interesting thing, though, is that crime, teen pregnancy, and the like are way down from where they were 20 years ago. The people most in need of our help are actually doing a rather impressive job of fixing their own problems themselves. When I was young and radical, back in the day, these things were bad and getting worse. It was quite depressing. But there are many fewer things that need to be fixed now, so it should be easier. (Heck, even obesity is starting to fall, if only slightly.)

          So maybe it wouldn’t take a generation.

          Another comment: life expectancy in the US is said to be “falling behind” the rest of the industrialized world. While that is true, it’s also true that life expectancy is _not_ getting worse. We haven’t lost the gains we made through 1995. So it’s not quite as bleak as a lot of commentators are saying. We just need to get back on track.

    • As we live longer and make progress in treating episodic life-threatening disease, cancers are emerging as the major cause of death. I would view environmental triggers of cancers as one of these social determinants of health.

    • -I feel like I’ve been chiming in and rehashing the point that variations in the way that people take care of themselves in a given area (“social determinants of health”) has a far more decisive impact on their health and longevity than variations in the way that doctors and nurses care for them for roughly twice as long as the age of the known universe, to little or no effect so I’m very please to see not one, but two posts that at least acknowledge the possibility.

      To answer the question, imagine that instead of talking about health outcomes, we were discussing the extent to which modifications to payor regimes and delivery systems could remedy differentials in crime and violence between Mattapan, MA and Newton, MA, or East LA and Santa Barbara, CA.

      It should be obvious that essentially doctors and hospitals have essentially zero capacity to modulate the immensely complex network of causal factors that drive crime and violence in a given community. At best – they can treat the physical manifestations of violence when they roll into their doors. When a meth dealer rolls into the ER with multiple gunshot wounds, there’s not a whole lot that any medical institution can do to avert the long-term health and mortality consequences of the this individuals life-history or future.

      This isn’t much different than the situation confronting a physician when a morbidly obese, opiate addicted, long-term disability recipient with chronic depression and an 8th grade education rolls into the ER with complications arising from the fact that he hasn’t been taking his diabetes meds. The overlapping network of social and personal pathologies that got this guy to his present condition is *vastly* beyond anything that can be addressed by tweaking payor/delivery regimes.

      My decidedly un-palatable policy prescriptions would be as follows:

      -Grant full tuition vouchers to children in failing school districts.

      -Immediately legalize all drugs and expunge all non-violent drug offenses from criminal records.

      -Eliminate the minimum wage and significantly expand the scope/magnitude of the EIC.

      -Replace extended unemployment benefits with direct subsidies to cover the cost of moving to take or find employment.

      -Allow all health insurance companies to fully price the costs of obesity into individual policies.

      The total effect on the health and longevity of the American public would be minor, but nevertheless far more consequential than any payor/delivery centric model.

      • “It should be obvious that essentially doctors and
        hospitals have essentially zero capacity to modulate the immensely
        complex network of causal factors that drive crime and violence in
        a given community.” Seems you are forgetting the powerful role that
        doctors can play as advocates and leaders in the community for the
        sorts of larger changes that might mediate factors driving crime
        and violence.

    • In short, I think the evidence is good that if we want to improve our population-level health outcomes in the U.S., we must address income and wealth inequality. I still find the evidence that there is a strong causal connection between within-nation income / wealth inequality and health outcomes (as well as a host of other social problems) to be compelling.

      The obvious ways to reduce income inequality are the tried and true methods of aggressive progressive taxation (including taxing capital gains at the same or higher rates than earned income), high minimum wages, strong protections for organized labor, strong social “safety-net” programs, etc. That these are reasonable policies to achieve lower levels of income/wealth inequality, and with that better population level health-outcomes, should not, it seems to me, be the slightest bit controversial — these are the policies that we know work to reduce income inequality, and that are strongly associated with both better population-level health outcomes, and with reductions in a variety of other social ills. There might, of course, be other reasons for resisting those policy recommendations. But again, I think we have good evidence that they would achieve the stated ends.

      I think the evidence is good that reforming our health-care delivery system would be of real value, but that that value will generally speaking not translate into meaningfully better population-level health outcomes without addressing the social determinants of health. And again, I think we have more than sufficient evidence about what works and what doesn’t.

      • Generally when people talk about reducing income inequality, they point to the post-WWII era as the ‘right’ level of income distribution/economic growth/prosperity. For example, from the Center for Budget & Policy Priorities:

        A Guide to Statistics on Historical Trends in Income Inequality

        “The broad facts of income inequality over the past six decades are easily summarized:

        The years from the end of World War II into the 1970s were ones of substantial economic growth and broadly shared prosperity.

        Incomes grew rapidly and at roughly the same rate up and down the income ladder, roughly doubling in inflation-adjusted terms between the late 1940s and early 1970s.

        The income gap between those high up the income ladder and those on the middle and lower rungs — while substantial — did not change much during this period.

        Beginning in the 1970s, economic growth slowed and the income gap widened…”

        Assuming a return to the period running roughly from 1945 – 1970 is the goal, my question would be, are you ready to bomb our nearest competitors into oblivion, and tell roughly 2/3 of the American population that their opportunities to freely trade their goods and services will be eliminated or at least not protected in law? Because between WWII, segregation, and de-facto legal sex discrimination were also major components of the economic system of that period.

        Probably strayed into the territory that caused Austin and Aaron to end general comments, so I’ll stop now. If anyone REALLY wants to discuss this, there’s always my e-mail sean [a] impactpolicymanagement.com

        • Nah, you’re not near the line. Moreover, halting comments by default was more of a volume than a quality issue.

        • Sean, I don’t think the issue is whether we can return to the growth rates of the post-war period. The issue is how the growth we do have is distributed.

          As you say, “incomes grew rapidly and at roughly the same rate up and down the income ladder . . . between the late 1940s and early 1970s.” Since then, nearly all income growth has accrued only to those at the top of the ladder, despite the fact that much of the reason for the growth is higher productivity from those at the lower rungs.

          It seems to me that this is an issue that lends itself directly to public policy solutions, starting, as Jonathan suggests, with reforms to the tax code.

        • I don’t think there is any good reason to believe that discrimination (on the basis of race, sex, etc) and income / wealth inequality have anything to do with each other. There are countries that beat us now with respect to income equality, and which do better on health-outcomes, that have not maintained their income distributions by embracing discrimination (but in many cases have stronger and better enforced laws against e.g. gender discrimination than do we). Also note that income distributions were similarly skewed in the early 1900s, when there was of course massive discrimination. So I just don’t think there is any real evidence that discrimination on the basis of sex or race was necessary for the income distribution in the U.S. during the period in question…

          The claim re: the necessity of bombing our competitors would likewise seem to run afoul of the fact of economic growth; the issue is not that we didn’t, as a country, make enough money since the mid- 1970s, nor that workers in the U.S. did not become more productive, etc., but rather that profits become increasingly unequally distributed.

          Finally, if by restricting “opportunities to freely trade their goods and services” you mean that, for example, people will not have the legal power to undermine e.g. union-negotiated contracts by agreeing to work for less than the negotiated rate (or undermine laws regarding working hours by agreeing to work more than the legal maximum, etc.), then yes, we would have to accept that. Labor is usefully conceived as a collective action problem, and protecting reasonable wages and working conditions has, historically, required collective action in order to succeed.

          Again, if we want relatively more of the people who live in this country to live relatively long and healthy lives, I think policies of the sorts I’ve recommended are the right direction to look. Indeed, I am convinced that if we do not reduce income and wealth inequality, we will not make any meaningful improvements in our population-level health outcomes (no matter how much we improve our health-care system narrowly conceived). We might not, as a society, care enough about people living relatively long and healthy lives to take these steps (we might care more about other things). But that’s a different issue.

      • @Jonathan:

        I’m surprised that you find the evidence for a causal association between income inequality and health outcomes, rather than a correlation with magnitudes that vary widely depending upon the geographic area and the demographic composition of the area under consideration.

        What specific studies stand out as the strongest that conclusively demonstrate a specific causal association between the magnitude of variation in income/wealth and a particular health outcome or outcomes that can’t be explained more parsimoniously by variations in patterns of behavior that tend to co-vary with income? E.g not eating/drinking/smoking too much, prudent behavior, the capacity to delay gratification, impulse control, etc, etc.?


        • JayB:

          Due to the nature of the beast, there are of course no studies that “conclusively demonstrate a specific causal association” between income inequality and health-outcomes; the sorts of controlled manipulations necessary to “conclusively demonstrate” a causal relationship are simply not available for population health outcomes.

          Obviously, Marmot and Wilkinson have been pushing the psychosocial line the most aggressively. You may not buy their arguments for the importance of psychosocial influences, but they do provide some reasons for thinking that relative deprivation and social standing is part of the causal pathway.

          This is an oldish, but freely accessible, summary with reasonable cites: http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1120336/

          Of course, Wilkinson’s views on the importance of the psychosocial elements have been disputed, and there are no “knock down” arguments or studies that prove he is right.

          I find the fact that even for people with relatively high incomes, living in places with higher income inequality is associated with worse health-outcomes to be suggestive (and to speak against the kind of individual focus on absolute deprivation and individual activities you seem to point towards):


          That said, I was being a bit glib in referring to a causal connection between income inequality and health outcomes; obviously, income inequality per se doesn’t drive the observed differences. I am confident that it is *in principle* possible to have both high levels of income equality and reasonably good health outcomes (but I am equally confident that such a society, while in principle possible, is practically impossible); further, I think it is obviously possible to have reasonably low levels of income inequality and poor health outcomes.

          The suggestion that income inequality and health outcomes are causally related is I take it a somewhat sloppy shorthand for a series of claims like the following: the same social / political policies that result in high levels of income equality are terrible for population health, because of the straightforward impacts on relative material deprivation, etc; under most realistic conditions, high-levels of income equality themselves have a negative impact on health outcomes via both psychosocial mechanisms (stress, etc) and the undermining of the social systems that promote good health (social cohesion, etc.); those policies that have been associated with lower levels of income inequality tend as well to promote good health outcomes via ordinary material mechanisms; lower levels of income inequality itself also tends to be associated with social systems that promote reasonably good health outcomes via a reduction in the severity of social hierarchies, lower stress, etc.. All of these I take to be “ceteris paribus” claims.

    • It’s the money , stupid. (To paraphrase a famous quote and NOT to be mean)

      Being poor is hard work. Hard, hard exhausting work. Lack of money in this society is agonizing. every little decision becomes stressful and stress, as is well documented, does horrible things to our bodies if experienced over a long period of time. If you worry about eating, clothing, housing and other necessities on a daily basis, well, you suffer. Your body exhibits the suffering. The best health care in the world will only do so much to help that.

      The sad and horrifying thing is that there is plenty of money in America to ensure that everyone can eat properly, dress safely, and live in decent housing. We just allocate it wrongly and concentrate it in too few hands.

      We choose to do that and only we can change that.

    • @Sheldon.

      Yes – productivity has been increasing faster than wages. The question is – what is driving productivity growth. That’s not the question. The question is what is actually driving the productivity growth. People working harder or capital investments that enable them to produce more output with the same level of skill and effort.

      Your contention seems to be that all of most of the increased hourly output a result of unskilled and low-skilled workers is a consequence of them working more efficiently than they did 40 years ago. There’s an easy way to test this hypothesis – compare productivity gains in processes with the highest rates of capital investment over that interval – like semi-conductor manufacturing, to sectors with the lowest levels of capital investment, such as gardening, home-care, sandwich making, hairdesssing, etc and compare the aggregate productivity gains over time. If the productivity gains in processes with high and low capital investment have been the same, then most or all of the productivity gains over the period have indeed been a result of low-skilled people working more efficiently. If it’s not the case, then it’s clear that the productivity increases have been driven by capital investment that allows people with roughly the same skill level, making roughly the same effort, to produce more output per hour. If that’s true then, like it or not, the vast majority of the returns from productivity will go to the people who created or invested in the equipment that enabled the average worker to produce more.

      The guy driving the earth mover is 10,000X as productive as the guy with the shovel, and will command a premium over the guy with the shovel, but the vast majority of the returns to productivity in that scenario will go to the guy who owns the machine.

      • “If that’s true then, like it or not, the vast majority of the returns from productivity will go to the people who created or invested in the equipment that enabled the average worker to produce more.”

        That isn’t a natural fact about the world; it is a political decision.

        Under conditions of strong organized labor and high progressive taxation, the “vast majority” of the returns from increased productivity of this sort will *not* go to the people who created or invested in the equipment. (Though, if we wish to encourage investment that spurs growth, certainly *some* of it will be have to be so-allocated!)

        Under conditions that fail to protect the legal right to meaningful ownership of such equipment or the returns of investment (e.g. in failed states, in very corrupt states, etc.) it is unlikely that the possible returns of increased productivity will be realized, and if they are realized, their distribution will not necessarily follow lines of ownership as usually understood.

        It is only under some very particular social / political conditions that the rewards will be distributed as you suggest.

        The current distribution of income in the U.S. is not some natural fact that came out of nowhere and about which we can do nothing. It is the result of particular policy decisions that were made, policy decisions that, like many policy decisions, picked winners and losers.

        We *could* pass laws that made organized labor more powerful, and made it is easier for workers to organizer (we’ve in fact done exactly the opposite). We could have a high minimum wage and reasonable protections for workers (we are well behind the curve here). We could have a much more progressive tax system. Etc. We choose not to. This is a decision, not a natural fact about the world.

    • History proves that economies based on markets and competition thrive while those that are centrally directed don’t. In view of this, the answer to improved healthcare (and just about anything else in the economy) is to foster more genuinely competitive markets.

      Yet liberal wonks continue to suggest policies that either call for or imply greater control over the economy by a central authority, such as some made in response to this post. And the suggestions are made so casually.

      How is this support for central planning justified given the historical evidence? Exactly how does one go about ignoring such evidence?

      • “History proves that economies based on markets and competition thrive while those that are centrally directed don’t. In view of this, the answer to improved healthcare (and just about anything else in the economy) is to foster more genuinely competitive markets.”

        There is no link between the necessity of market-based approaches to economic success and the claim that more unregulated market-based approaches are better or necessary in every domain. Calls for increased regulation in labor, education, and healthcare are not calls for the central planning of the production of consumer goods, and it seems a little disingenuous to pretend otherwise.

        “History” does not prove that market-based approaches are always, or even often, superior, if this is understood to mean that everything should be distributed via “genuinely competitive markets.” Quite the opposite, in fact. There are many, many things that are best “distributed” in civilized societies via non-market mechanisms, for both practical and moral reasons.

        But, if we want to think about “history” note for example, that there has never been a successful economy that didn’t involve substantial government regulation and “interference” with a substantial portion of the possible markets. Never. Anyplace.

        On the topic of the post, it is worth noting that no country has ever achieved anything even resembling decent health-outcomes for its citizens without a robust regulatory framework, including strongly enforced laws regarding a wide variety of markets, including the labor market, financial markets, education, and yes, healthcare. No country. Ever. Anyplace. Look at the places with the best health outcomes. Are they the places with the least regulation, either of the healthcare market or more generally? No. They are not.

        Similarly, the countries that are doing well educating their citizens are not those that have left education up to market forces; the U.S. states with the best educational outcomes are not those with the most privatized school systems. There is no evidence that markets improve educational outcomes, and plenty of evidence that they do not.

        *That* markets are important to economic success (and, as Sen for example points out forcefully, to individual freedom more generally) is not doubted by many people.

        But the claim that because markets are important, the less-regulated the markets the better simply does not follow from the importance of markets.

        I think any reasonable view of “history,” as well as any reasonable cross-national comparisons, reveals that in fact both economic success and population well-being depend *critically* on the (strong and extensive) regulation of markets, often including the near-complete *exemption* of particular ranges of goods (education, access to a reasonable level of healthcare, access to the courts and the political realm more generally, etc) from ordinary markets and competitive forces entirely.

        There are, to the best of my knowledge, no examples of countries that have achieved reasonable levels of human flourishing via anything like ‘pure’ market-based approaches to education, to healthcare, or even to labor. None. Ever.

        Of course, neither are there examples of countries that have achieved reasonable levels of human flourishing without embracing some (extensive range of) market-based approaches to a variety of important domains. But that market-based approaches are necessary does not imply that they are sufficient, and “history” (and a glance at successful countries around the world!) would certainty seem to imply that they are not.

        If we want our health outcomes to look like the best health-outcomes in the world, it seems that we might think about adopting some of the policies of the countries that achieve those health-outcomes. In no cases are those policies that involve deregulating (or otherwise making more market-driven) health-care, education, and labor. Rather, the places doing better than us would seem, in general, to have left far *less* to the market in these domains, not more, and have more “central planning” and regulation, not less.

        • Thanks for the reply. Wonks like charts, especially those
          comparing U.S. and European healthcare. Maybe we need one showing
          the relative performance over the last 100 years of
          centrally-directed economies versus those based on markets and
          competition. But even without a chart, we know that all those
          “natural experiments” of the past point us in the direction we
          ought to go. Which is toward genuine markets, not away from them
          like we’re doing today. This does not mean that markets can’t be
          regulated – the idea is to be pointed in the right direction. It’s
          not a stretch to think that healthcare in the U.S. (which
          represents one-sixth of the world’s largest economy – hardly an
          insignificant amount of activity) is moving in precisely the wrong
          direction. It’s getting to the point, for example, where health
          insurers are nothing more than extensions of the central
          government. The idea that other countries are getting great results
          in healthcare and other areas of the economy through greater
          central planning and regulation is questionable. Again, the basic
          evidence suggests that these results would be even better with less
          centralization. Markets mobilize and coordinate fragmented and
          widely dispersed knowledge better than any small group of
          individuals, no matter how much expertise they think they have. And
          central planners cannot achieve the same efficiencies as those that
          result from competition. But if we’re happy with a relatively
          stagnant economy as we continue to centralize, okay, but we should
          understand that stagnation (check out Greece and Spain with their
          26% unemployment rates, 50% for young people – such a bright
          future) will be our end point.

    • “Maybe we need one showing the relative performance over the last 100 years of centrally-directed economies versus those based on markets and competition.”

      But no one is suggesting that we move towards “centrally-directed economies” in general — none of the posts above that you identify as suggesting “liberal” solutions suggest that we turn control of the entire economy over to central planners. That is not the argument that is being made, and so the fact that it would be a very bad idea is simply irrelevant.

      I admitted above that well-functioning markets in a wide variety of arenas are of critical importance to both a functioning economy, and an important aspect of meaningful human freedom. I accept that as given.

      What I don’t accept is that the markets necessary for economic growth and human freedom include markets in e.g. reasonable access to healthcare or to education. Nor do I accept that the markets necessary for flourishing economies and people are incompatible with reasonably high marginal tax rates (including capital gains and estate taxes), a reasonably high minimum wage, or laws that generally support the rights of workers to organize and collectively bargain. And no, there is nothing in history that suggests otherwise.

      What is relevant to the argument is not that “centrally planned economies” don’t work well. Rather, it is that the evidence with respect to *health care* and *education* suggests strongly that these are areas for which *markets* to a first approximation simply don’t work.

      Again, you can cite NO examples of countries that have achieved anything like reasonable levels of population health and well-being that have embraced market-based solutions to health care and education, because there are none. None. Ever. Anyplace. Period. Doesn’t that simple fact mean anything to you?

      So don’t lecture on me on fact that a well-functioning market in e.g. cars or shirts or soybeans works a whole lot better than a centrally planned one — that isn’t the issue.

      Again, I point you towards the obvious fact that the places achieving the best health outcomes are not those with unregulated, competitive markets in healthcare. The places achieving the best educational outcomes are not those with markets in education. These are places where success seems to follow non-market approaches. (And please do note that not every non-market approach is “centrally planned,” either; consider for example that higher education was quite diverse, often local in all kinds of weird ways, and yes, quite innovative, back in CA, back in the day, when it was free…)

      The claim that a more or less pure market based solution *would* work in these areas, if it was embraced aggressively enough, is based on nothing but faith. And the basic nature of these arenas would seem to suggest that these are not realms where market-driven competition makes sense, for the very kinds of reasons that have been noted above (asymmetries of information, the non-optional nature of many key “purchases,” the gap between who decides on the course of action and who benefits, etc.).

      The more general so-called liberal suggestions above — that we work to reduce income inequalities through the standard mechanisms of progressive taxation, a high minimum wage, strong labor laws including protection for collective bargaining, a robust safety net, truly free universal high quality education — are not some Soviet-era communist pipe-dream incompatible with economic growth. They are the exact systems that have been associated with strong economic growth over and over again, in successful countries around the world. (Do they *always* work, in *every* permutation, under all circumstances? No. But that’s really not the question, is it? So please don’t lecture me about how badly Greece is doing.)

      If you can find a place with (some combination of) weaker worker protections than the U.S., a more market-driven educational system, a less regulated, more market-driven healthcare system, someplace with higher levels of income inequality, less social/economic mobility, a tax system that is *overall* less progressive, a weaker social safety net, and that actually achieves better health outcomes than we do, let’s see it. I’m pretty sure there isn’t one. But you know what? I can point to dozens of countries that do more or less the opposite in almost all of those realms, and beat us hands down on pretty much every measure of population health and well-being. And if you look at between-state variation in the U.S., you find the same basic pattern. So if that’s what we can about, maybe, just maybe, we should think about doing those things that the evidence would suggest actually work, rather than pointing towards the fact that in other, completely different realms, different things work.

      I want to stress again that this isn’t, in and of itself, a reason to adopt these policies. If one thinks that e.g. the freedom to contract freely with respect to one’s labor is of fundamental importance, one will likely be unmoved by the argument that it undermines collective bargaining and increases income inequality, even if increases in income inequality are associated with higher morbidity and mortality.

      But the “game” of this post was to come up with policy recommendations outside of the health-care system as normally conceived that would have the best chance of improving population-level health outcomes. I stand by my recommendations as the having the best chance of making a meaningful difference in our health outcomes, given what we know about the influences of the social determinants of health. I am equally confident that moving towards more market driven approaches to healthcare and to education, and weaker protections of workers would have the opposite effect.