{"id":317147,"date":"2010-02-13T14:00:27","date_gmt":"2010-02-13T19:00:27","guid":{"rendered":"tag:business.theatlantic.com,2010:\/\/3.35923"},"modified":"2010-02-15T10:21:09","modified_gmt":"2010-02-15T15:21:09","slug":"firming-up-the-argument","status":"publish","type":"post","link":"https:\/\/mereja.media\/index\/317147","title":{"rendered":"Firming Up the Argument"},"content":{"rendered":"

For some reason, everyone I’ve seen who has so far responded to my piece on mortality and the lack of insurance has pulled exactly the same debater’s trick:  they restate my argument in maximalist form, and then proceed to really kick the hell out of the strawman they’ve created.  I mean, there’s hay and shreds of fabric on the ceiling of offices three floors above them.  Button eyes flying off so fast that several have achieved escape velocity. Crows shrieking in terror as far away as Fresno and Marseilles.  It’s a stunning display of . . . something.<\/p>\n

But I have not asserted that insuring the uninsured wouldn’t
\nsave anyone’s lives, so I don’t know why they’re making this argument
\nwhile linking to me<\/i>.  What I said is, the studies so far done
\noften cannot exclude the possibility that there is no effect–this is
\ntrue of one of the two studies that IOM\/Urban relied upon, and also of
\nthe largest observational study done to date, which found no effect. 
\nThat is not the same as saying there is<\/i> no effect.  Health
\ndata, like economic data, is very noisy.  Sometimes effects that we’re
\npretty sure exist just can’t be easily teased out of the data . . .
\nlike, oh, I dunno, the effectiveness of fiscal stimulus, say.<\/p>\n

What I am<\/i>
\nsaying is that we don’t know how big the effect is. Refuting me
\ninvolves, not saying that well, here’s another study showing some
\neffect, but rather, taking a stand and saying we do know how big the effect is<\/i>, or at minimum, that we can prove it’s probably at least 20,000 people a year, the figure I was discussing. <\/p>\n

Because
\nof course, size matters.  If you want to argue in favor of a national
\nhealth care system on the basis of improvements in mortality, then the
\nnumber really has to be quite large.  By 2019, the CBO expects the
\ngovernment to be spending just about $163 billion more the exchanges,
\nand the Medicaid\/S-Chip expansions.  (About $100 billion of that is to
\nbe offset by Medicare and other cuts–but I’m just trying to isolate
\nhow much we’re going to spend to expand coverage, since we could do the
\ncoverage expansion without the Medicare cuts, or the Medicare cuts
\nwithout the coverage expansion–and doing the latter would give us
\nmoney for other things, so there is an opportunity cost to using them
\nfor this).  <\/p>\n

If 1,000 people die a year, that means that we
\nwill be spending $163 million per life saved.  If the number is 5,000
\nwe would be spending $33 million.  In fact, you need the number of
\npeople dying from lack of insurance each year to be quite large–more
\nthan 20,000–to get the dollars-per-lives-saved within the ranges that
\nsay, the EPA or the NHTSA use for doing cost-benefit analyses on
\nregulations.<\/p>\n

Now, obviously, as I’ve also said repeatedly, there
\nare reasons beyond mortality that we might support this system. 
\nMortality is only one element, albeit an important one.<\/p>\n

But I
\nthink that journalistic hunger for “a number” has resulted in some very
\nrough numbers with a lot of weaknesses being adopted as a fact in the
\ndebates.  They’re not facts, they’re very rough guesses, and they
\nshouldn’t have been used as a selling point for this plan without at
\nleast some investigation of how reliable they were.  Moreover, I know
\nthat the people arguing with the study understand the problems, because
\nthey suddenly rediscovered them in regards to the data on deaths before
\nand after age 65.  A lot of people are arguing that we should ignore
\nthe aggregate data on Medicare mortality statistics in favor of Card’s
\npaper on the discontinuity between health outcomes for ER admits who
\nare just under 65, and those just over. <\/p>\n

I see the argument for
\nusing easier-to-measure subgroups in an attempt to isolate causality. 
\nBut here’s the thing:  you cannot say, well, aggregate data isn’t very
\ngood for capturing causality, and also cite the figures from the Urban
\nInstitute, or Himmelstein et. al. as if they had some meaning.  Those
\ndata are far worse than the Medicare data, because at least the
\nMedicare data gives you a natural experiment, and it doesn’t try to
\nisolate “the uninsured” on the basis of their insurance status on a
\nsingle day.<\/p>\n

It also rather misses the point of why I find the
\naggregate Medicare figures so striking, something I’ve been hammering
\nout in private emails for several days.  People under the age of 55
\naccount for a very, very small proportion of deaths in this country. 
\nIt’s after 60 that you’d expect to be getting the largest mortality
\nbenefit from expanding insurance coverage.  If switching people to
\ngovernment-run insurance at the age of 65 doesn’t produce any
\nmeasurable improvements in the mortality rate of a population with a
\nhigh mortality rate, then how big an effect could a national health
\ncare system for younger people have on mortality outcomes?  Would it
\neven register in national statistics?  Is this the right use of $163
\nbillion? I mean, we can say that a policy is a success if it saves even
\none life, but it is not actually possible to run a country this way.<\/p>\n

If
\nyou want to see the argument only in maximalist terms, I’d like to see
\nsome of the people who have pushed my argument to its maximalist
\nconclusion endorse some maximalist propositions of their own.  If
\nyou’ve used the 20,000 or 45,000 figures regularly, approvingly cited
\nthe Himmelstein et. al. bankruptcy data, and stated as a fact that the
\nhealth care bill will reduce deficits because the CBO said so, you
\nshould have no trouble signing on to the following propositions:<\/p>\n