I'm Darshak Sanghavi and I Like Making Bad Arguments

So there's an irritating piece at Slate by Darshak Sanghavi about allocating health care in developing countries. The piece is annoyingly vague - it implies more than it says, so it's hard to articulate a critique that couldn't be dodged fairly easily.

Mainly, though, the piece criticizes health care economics by pointing out instances in which people have messed it up. Apparently economists use a concept called a QALY - a quality-adjusted life year. The idea is to spend scarce health care dollars efficiently by allocating them to the cheapest QALY available.

One might object that aggregate QALY is not really what we want to maximize. Sanghavi's criticism of QALY, though, is that in the past people have used inaccurate prices for various treatments, resulting in skewed QALY analysis (at times, he also implies that QALY is meaningless as a concept). Very well - but this seems to be an argument not against QALY, but against using bad data. If you've calculated the cost of a drug at $100, and it actually costs $50, that's going to mess things up under any allocation system I can imagine.

Sanghavi also uses a bunch of examples that are designed to evoke outrage. Should we really spend more on thumb-sucking treatments than on cystic fibrosis? Well, maybe not, but it's hard to see how we can avoid funding health care according to some system of priority. Sanghavi doesn't provide an alternative, so he takes potshots at the current system without pointing out any way to avoid its shortcomings. Moreover, Sanghavi claims that the system is "unfairly rigged, so the world's poor can never win." This isn't an argument about how much aid to give to developing countries, though - it's an argument about how to spend whatever money is available. Sanghavi himself rigs the thought experiment by introducing us to an African child who will die without an expensive heart operation. True, saving the kid's life might be more valuable than any amount of thumb-sucking prevention, but those aren't the only options. Plenty of other children will die of diseases that are much more cheaply prevented or treated. You can forget about the problems with calculating quality of life: we're talking about apples and apples, children dying of one thing or another. It's grossly irresponsible to find a few problems with the status quo and then reject the concept of using quantitative analysis to allocate health care. Lines like this are absurd: "With this [QALY] logic, Africans with AIDS and expensive heart troubles never will qualify for life-saving drugs or surgeries." But of course, under Sanghavi's logic, thousands of children who would have received cheap preventive care won't get it and will die. You can't figure out if a trade-off is worth it unless you know both sides - the costs and the benefits.

In fairness, sort of, Sanghavi doesn't (explicitly) argue for the abandonment of quantitative analysis - he never tells us how he would allocate the extremely scarce health care resources in poor countries. In reality, though, I think fairness demands that he take the problem seriously, and I don't think it's fair to pretend that sanctimony constitutes seriousness.

Stop Confusing Me

I really don't get this line from a NYTimes piece on a tourist attraction in Maine:

The observation tower, which opened in May and is the only one in the United States, provides an awe-inducing panorama of Maine’s mountains and coastline and a bird’s-eye view of the nearby village of Bucksport.

I really just can't figure it out. True, the observation tower is part of a bridge, so maybe that's unique? That's a very odd construction of the sentence though, and I can guaran-goddamn-tee you that it's not the only observation tower in the US. The title is also odd: "Sure, the Bridge is a Marvel, but How About the View?" As the piece makes clear, though, the view is excellent. To me, the title implies that someone, somewhere is questioning the view. Not so, to all appearances. Glad we cleared that up.

In any case, it does look like a cool bridge. If I were one of the engineers, I'd clip the NYTimes piece and save it, so that I could get it out every once in a while and say to myself, "What the fuck?"

It's Like Royal Dutch Shell

So I don't understand why arbitrageurs haven't closed the gap between futures on the Intrade and Iowa election markets. At the time of writing, Senator Clinton is trading at about 43 cents on Intrade and about 46 cents on the Iowa markets. Why can't someone just sell on Iowa and buy on Intrade? You could make a quick 3 cents on the dollar while taking no net position on the underlying bets. Both markets are real-money markets, and I can't discern any difference between the bets.

[UPDATE: I think there's an 8 cent gap on Obama getting the nomination. Anyway, these numbers are just illustrative - the real question is why any (significant) gap exists for any candidate.]