My friend Alan Nelson recently posted a link on facebook to the following article: http://www.nytimes.com/2014/03/23/business/economic-view-when-the-scientist-is-also-a-philosopher.html with an appropriately snarky note that the author, N. Gregory Mankiw (the Chair of the econ department at Harvard, natch), seemed to be arguing that the only changes to the status quo permissible are those that are verifiably Perato efficient improvements.  An obvious corollary is that, since every reasonably substantive and complex policy change will have winners and losers, we should never change policies at all.



The author is right about one thing: economics, no matter how much some practitioners might like to pretend otherwise, is wrapped up in philosophical assumptions.  These are too often held implicitly, and too rarely made explicit.  (One of my favorite books is Varoufakis' "Foundations of Economics: A Beginner's Companion"  http://yanisvaroufakis.eu/books/foundations-of-economics-a-beginners-companion/ simply because it does, I think, a really nice job trying to make explicit some of those usually implicit assumptions.)

But Mankiw's suggestion, that "economists should be sure to apply the principle 'first, do no harm'," is deeply problematic, and obviously wrong-headed.  It treats current policy as if it were handed down from on-high, or, worse, as if it were the result of fair bargaining on the part of those people now engaging with it. 

(Is it merely a coincidence that he is arguing for a "do no harm" approach at a time when organized labor has been effectively destroyed, the tax system made effectively regressive, the minimum wage eroded by inflation, and regulations on industry etc so severely weakened?  Because of course, a "do no harm" approach adopted at an earlier period would have resulted in a very different economic world, wouldn't it?  You can't avoid picking winners and losers.  But notice the winners Mankiw implicitly picks!)

To take an example from his article, health care insurance reform, most of the people suffering under the previous "system" (really a hodge-podge of partially overlapping non-systematic bits) of funding health care in the U.S. had no say in its creation, and many of those who were "losers" in the previous system suffered very badly under the policies that had grown up around healthcare delivery and payment in the U.S.  It is fairly obvious that there was no way to get from the system we had to a system in which the current losers weren't suffering as badly without "harming" someone.   But the argument that we ought therefore have done nothing, that tens of thousands of unnecessary deaths per year was simply unavoidable, because to change the "system" would involve, by necessity, making some people worse off, is surely not the right answer to reach in this case.  But I can see no way to read Mankiw's article except as recommending precisely that.  

 Whether one thinks that the ACA was, all things considered, a good response to the real problems with the old "system" notwithstanding, any change that addressed the substantial problems that existed before would create winners and losers.  A policy that demands that all changes produce  only winners works only to entrench the current winners, and dooms the current losers.  When the losers are suffering in ways that are morally unacceptable, such a policy cannot be right one.

The "philosophy" (and politics) that get imported into economic theory need not be limited to naive utilitarianism, nor is it.  We need to admit that certain "economic" questions require us to not only make guesses about possible economic outcomes (narrowly construed) but about other important values as well.  One such value, that I at least hold dear, is that an honest day's work should earn an honest wage, which I, at least, interpret to be one that respects basic human dignity.  Another reflection of these values is that a "system" in which  many people in one of the richest countries in the world die every year because they lack access to reasonably affordable healthcare is morally repugnant.  And neither a naive utilitarian approach, nor especially a pathetic half-assed defense of the status quo, captures the importance of these values worth a damn.

If this is the best economics professors at Harvard can do to integrate important values into economics, something has gone terribly wrong.

(Edited to remove some able-ist language!  Thanks to John P for calling me on it.)

Posted in , , , , ,

4 responses to “Bad Economics/Economists: Right Diagnosis, Wrong Prescription”

  1. Christian Marks Avatar

    “An obvious corollary is that, since every reasonably substantive and complex policy change will have winners and losers, we should never change policies at all.” This doesn’t follow. The Pareto-efficient frontier need not be a single point. There could be room for a range of policies, some or all of which may be “unfair.” One needs a mathematical notion of economic justice–for me this is a prohibition against one group systematically winning asymmetric zero-sum games against another group. Economists and game theorists could formulate class differences in terms of repeated asymmetric zero-sum games (and take advantage of the putative “value neutral” language of contemporary game theory). However, they seem to have little to say on the subject, though asymmetric zero-sum games go back to von Neumann and Morgenstern’s Theory of Games and Economic Behavior. There is no question that your concerns can be formulated in terms of mathematical economics. I believe this undertaking is necessary.

    Like

  2. M Lister Avatar

    There’s a lot that’s bad in that piece (not least the almost certain dishonesty in it) but it’s also worth noting that very few economists (and I’m almost certain this would apply to Mankiw, too) are especially insistent on the Pareto standard, but rather use the less-demanding Kaldor-Hicks notion of efficiency, where what matters is that the “winners” could pay off the “losers” and still come out ahead, even though no actual changes are expected to be made. That’s the standard used in almost all cost-benefit analysis and is really the standard in mainstream economic analysis. It still suffers from the problems noted here, but I’d be shocked if this isn’t the standard Mankiw actually uses, when he’s not trying to make bad-faith arguments against improving the lot of the worst off.

    Like

  3. Jonathan Kaplan Avatar
    Jonathan Kaplan

    Christian — Addressing the second comment, yes, the failure to model that particular injustice is indeed a failure, but it isn’t the only failure. Addressing the first, if your criterion for “do no harm” is that no one in fact be in the slightest inconvenienced by your change in policy, I can see no way of substantively changing policy. Yes, there are more sophisticated interpretations available, but those are not what Mankiw endorses in this odd little article, are they?
    Lister — Yes, it would be shocking if Mankiw in fact endorsed a position as down-right stupid as the one he decided to publish in the NYT, wouldn’t it? And yet, there it is. In the NYT. Hardly the place to publish slap-dash musings, I would think…
    Look, I’m willing to cut people a lot of slack — Eric Schliesser has noted that I haven’t begun to fully account for Mankiw’s hypocrisy in this. But if you claim, as Mankiw clearly does (in print, in the NYT) that a) any policy change with harms anyone should be avoided, and b) the ACA was a bad policy, because it mildly inconvenienced some people while permitting millions of others reasonable access to health care (note: there are other reasons to think the ACA is problematic, but this is the argument Mankiw actually uses! Some people who liked their insurance got it canceled and had to get new insurance. That is the entirety of his argument against it, his entire evidence that it “does harm” and hence should be rejected), I think I am justified in attributing to you the position that no policy should be accepted if it in fact negatively influences anyone in anyway. And from that it pretty much follows that no policy changes of any substance are possible, potential Pareto efficiency notwithstanding.
    Yes, it is, at best, a very odd position for a “serious” economist to endorse. That is part of what makes arguing for it in a public forum so odd, isn’t it?

    Like

  4. M Lister Avatar

    Hi Jonathan,
    Yes- I agree. He does seem to be invoking a strong Pareto criteria for evaluating policy. My point was only to add one more reason to think he is being dishonest here by noting that that’s not at all the normal criteria you’d expect someone like Mankiw to use. That he busts out a standard he wouldn’t normally use when he doesn’t like the subject of evaluation just helps show that he’s not being honest.

    Like

Leave a comment