D-squared Digest -- FOR bigger pies and shorter hours and AGAINST more or less everything else: Well, I promised myself I'd finish this before the sequel appeared in the shops, and the conclusion has been made, shall we say, somewhat easier by the fact that the burden of my conclusion - that there is something terribly, horribly wrong with the state of modern economics - has become somewhat of an open door to push against. I swear that my notes for this review (begun in 2003!) contain the draft passage:
"When future generations ask the economics profession 'What were you doing while the great bubble built up ahead of the Second Great Depression?', and we have to reply 'Lots and lots of quirky little working papers about sumo wrestling and speed-dating', it is going to be really, really, fucking embarrassing"
And we did, and it was; thank God nobody told the truth to HM The Queen, or the high brows of the economics profession might be decorating a series of pikestaffs outside Traitors' Gate.
The basic problem with the Freakonomics era was that the profession abandoned the study of production, consumption and exchange. I don't wholly agree with Lord Skidelsky, but he is right - economics is the study of the economy, it's not the study of "rational choice" or "behaviour" in the abstract, and the fact that econometricians have invented a huge part of the toolkit of modern statistics doesn't mean that anything you can estimate using an econometrics package is thereby "economics".
We stopped doing economics and started doing awful amateur-hour sociology, basically, because we believed that all the major problems had been solved, that some form of dynamic general equilibrium was all that there was to be said about the economy considered as a system, and that the only interesting things to do were growth theory and finance. It is no coincidence that Freakonomics began in Chicago; for a guy like Levitt who doesn't possess the engineering-maths to be a finance theorist or the empirical skills to do endogenous growth, there was literally nothing to do.
The sociology of academia in the USA also played its part, as James Heckman spotted at the time. Because of the unenviable economics of the academic labour market in American universities, graduate students were encouraged to finish their PhDs according to a specific schedule, to write dissertations that were capable of being turned into journal articles in a specific way, and to follow fashion in citation-gathering. Heckman was tearing his hair out over this, obviously, as this made it more or less economically unviable to carry out the kind of economic work that he does (and did) - careful, time-consuming, incremental, often abstruse but always relevant to the very big questions of the economy.
And so we ended up with Freakonomics, the disciplinary equivalent of the battery chicken. The subject matter became more and more cutesy and trivial, methodological corner-cutting in "natural experiments" became the norm, and the idea that there could actually be a subject of macroeconomics became almost quaint. The Freakonomics blog still tries to preserve itself against irrelevance by declaring the financial crisis to have been merely a failure of macroeconomics (thus presumably leaving Freakonomics to carry the field?), but it's desperately unconvincing, particularly given that they end up citing a fairly badly misconceived critique of the whole discipline in order to do so.
So, where to now? Well, I don't think it's 100% bad news. I am hoping that the instrumental variables/natural experiment movement will die on the vine, simply because the interest in the profession has now swung back to the big, important macro aggregates, and "quirky" datasets aren't considered as interesting as they used to be. Also, I profoundly hope that Charles Manski partial identification approach catches on, as it seems to me from half way through his book that this is clearly the correct way to address the kind of problems that natural experiment/instrumental variable literature does so badly. (The guts of the partial identification approach is that one models the data nonparametrically in order to put some very wide bounds on a quantity of interest, then start making parametric assumptions in order to narrow those bounds, checking all the way along what assumptions you need to make in order to get results. It's an approach that has safeguards intrinsically built into it against the central problem of Freakonomics; the tendency to say that "whichever way you cut it, X", when you mean "whichever way I cut it, X". My basic issue with the whole Levitt approach is that Freakonomists have this horrible habit of presenting their results as being much more authoritative than they actually are, often in policy contexts where this overselling is almost guaranteed to have baleful implications.