I was in Sydney for a fascinating conference on Evidence, Science And Public Policy. It was worth the trip just to hear John Worrall on evidence-based medicine point out this paper on remote retroactive intercessory prayer [1]. Assuming, as appears to be the case, that the study was totally legit (no data mining etc), the obvious question for me was why anyone would think it worthwhile (ex ante) to test this out.

But that’s not the subject of this post.

In discussion at the conference, some reference was made to Rawls, Cohen and incentives, the subject of quite a bit of discussion at Crooked Timber. Rawls and Cohen (and those of us following them) spend a lot of time on the question of whether it is just/desirable to adopt policies that increase the income of the well-off relative to that of the poor, assuming that the result is a Pareto-improvement (everyone is better off, even if the rich gain more). That’s an important question if you want to think about an ideal social order, or to clarify concepts of justice, but there’s a big risk (evident in some of the discussion at the conference) of sliding into the assumption that this question is politically relevant right now. To put it another way, this question is politically relevant only if you accept something very clsoe the Laffer Hypothesis[2], that a reduction in tax rates will, under current circumstances, produce an increase in revenue. Chris Bertram made much the same point in relation to the banking crisis a while back. The term “trickle-down economics” describes the general form of the claim.

There’s very little reason to believe the Laffer hypothesis or equivalent claims about the banks. The reason tax rates aren’t higher and bankers are getting bailed out on hugely generous terms isn’t because Rawlsians have outvoted Cohenites behind the veil of ignorance, or even because lots of economists believe the Laffer hypothesis. It’s because the rich and powerful are, well, rich and powerful. Not only can they promote ideas, however dubious, that serve their cause, they can bring powerful force to bear against any government or political movement that threatens their interest. All of this is obvious enough, but after thirty years in which any mention of these facts has been shouted down as the “politics of envy” or “class hatred”, it may be necessary to restate the obvious.

Again, that’s not a reason not to talk about whether Pareto improvements are (necessarily) desirable and just. I only want to remind everyone to mention, from time to time, that we’ve got a long way to go before we need to worry about this in practice. At the moment, the relevant version of the question is how the left can regain some of the ground lost over the last thirty years, now that the trickle-down theory has failed so spectacularly and obviously.

fn1. Paywalled, sorry. The authors of the study took records of people who had been treated for blood infections some years previously, randomly assigned them to two groups and got a volunteer to read a brief, non-specific prayer for the recovery of the test group, while holding the list of names. It turned out that the prayed-for group had had significantly better outcomes. John Worrall presented it as an data point against believing in randomised trials as the gold standard of evidence based medicine. As I said, my main puzzle is, assuming that the results were just a 100-1 fluke, why the authors thought of doing this in the first place, given the low probability of a publishable result.

fn2. Laffer didn’t invent the curve to which his name is commonly attached, but he can reasonably claim responsibility for the hypothesis that the US in the 1980s was on the declining part of the curve.