Income inequality follow-up point

Offline convo with a friend about the inequality measurement issues that I discussed in my last two posts (concerning the Auten colloquium) has prompted some further reflection on the relevance of healthcare.

As noted in the previous post, U.S. income distribution looks more even than it otherwise would if one includes, at the bottom, the cash value of healthcare that poor people get from sources such as Medicaid and (if they are old enough) Medicare. This also reduces the percentage of the pie that one measures as sticking to the top. So it makes things look more even than otherwise.

In the Auten-Splinter measure, this is a key reason why things look less unequal than they otherwise would. US healthcare costs have been exploding relative to GDP over the last couple of decades, this does reflect technology-driven improvements in care, and the bottom 50% of the distribution are getting some of this care.

Piketty, Saez, and Zucman spin this same data point differently, arguing that a huge proportion of any income gains at the bottom are being swallowed up by rising healthcare costs. This also is unmistakably true; the question is what to make of it.

Auten and Splinter, rightly in terms of their purposes, view the question as one of whether poor people are getting full cash value. Arguably they are, given how important healthcare can be to one's quality and length of life, and especially if one forces on them the hypothetical calculation of whether they'd swap it for cash (and whether it would be rational for them to do so) in the absence of free emergency care when things get truly dire.

I certainly wouldn't favor eliminating Medicaid and Medicare for the poor and handing them the cash instead (in the amount of how much the services cost), even if it were politically stable to continue doing so. But this admittedly reflects not just the value I see it as having to them, but also the altruistic externality to others (such as myself) from their not being denied vital medical care even if they'd rationally spend the money on something else given the other great deficits in what good things in life they can afford to buy.

But now let's add another point to the mix.US healthcare is by far the costliest in the world, pre capita and relative to GDP, and does not provide better results than other economically advanced countries get at a far lower cost. From the example of other countries, it's plausible that we ought to be able to get healthcare that is just as good (judged by results) for half the costs. I don't mean, of course, that it's realistic to think that we can get there from where we are now, but rather that, had the system evolved in a different fashion over decades, that's where we might be. It's also plausible that rent seeking is an important part of the story of why we are where we are, rather than there.

Now let's imagine a world where the U.S. is like that, hence people in the bottom 50% are getting healthcare of the same value for half the present cost. Suddenly, all else equal, their incomes have substantially declined (as measured by Auten-Splinter and anyone else who values the healthcare at cost), so, at least before we make any other conforming changes, the society looks far more unequal in the statistical measures.

Next step, of course, is that someone is being paid less, so distribution higher up will change as well. Plus, we then have to run the thought experiment, what else happens instead? E.g., is GDP simply lower since healthcare costs less, although by hypothesis everyone is just as well off as before except that some people have been paid less by other people? Consistency suggests trying to think this through a bit further, e.g., in terms of other output that occurs instead, but by now we're not only in the realm of speculative fantasy but far beyond what distributional measures can reasonably give us.

Multiple bottom lines are possible off this speculative chain, but here's one. In a distributional measure that (perhaps reasonably) views healthcare as worth its market cost to poor people, the fact that it is more expensive due to all the defects in how our system has evolved causes poor people, in a sense, to look spuriously better-off than they would, all else equal, in the scenario where we didn't get on a different healthcare path than peer countries. On the other hand, despite its high cost, we are in fact still giving this care to poor people (to the extent that, as per the measures, we actually are). To some extent it all comes down to how you want to think about it, apart from the undeniable point that these measures really can't tell us everything that we would like to know about inequality.