Syndicate content

Add new comment

Submitted by Jeff Hammer on

Hmmmm… no big policy changes that worked. Let me see… Oh yeah. Korea did something in 1962. It might have been liberalizing interest rates, promoting savings and having producers face international prices. It might have been industrial policy and deliberate export promotion. I rather think the former but this is a period worth studying having raised the income of the country from dirt poor to quite rich in two generations, massively raising the incomes of (now 50) millions of people.
China in 1978 took a wild guess that agricultural supply curves sloped up (where they got that idea, I’ll never know). They were right. Rural incomes skyrocketed. This was followed by attempts to get the same results in industry. Was that an “experiment”? The first, not at all – it was theory (supply and demand) and common sense. The latter? In the sense that it was done in stages and they watched what happened, perhaps. No one recommends abandoning any empirical work. But no randomization, no careful application (by graduates students) of some cute gimmick. The result was hundreds of millions of people getting massively better off. Second order big policies included managing transport networks along the (one – so no chance of a control) coast and ensuring a reliable electricity grid. The contributions of those are eminently possible to study but certainly not by an RCT.
India in 1991 dropped tariff rates dramatically (politics or not – a long story) and allowed quite a few markets to operate much more freely. Just as policy analysts familiar with economic theory (even as long ago as that!) would have, and did, suggest. Income of hundreds of millions of people rose dramatically since. Twenty years later the fact that not all of the billion people fared so well is a real problem that is, also, eminently study-able or would be if the statistical office had kept more standard data (as in rich countries) so that things could be tracked, analyzed, theories developed and checked against data.
As with the rest of Southeast Asia. Lots of policies happened and either they worked or were irrelevant but, in any case, can be, and have been, studied with other kinds of data. These studies are never perfect but have yielded a large number of well-agreed upon conclusions along with some stubborn areas of disagreement that may be studied but will not yield to anything like RCT’s.
So, I would contend that all the policies inspired by the little experiments that ever were and ever will be conducted will never help a small fraction of the many hundreds of millions of people that important policy changes have done, of which the above were just a few examples. Of course, they will not be studied by our best economists because of what is and is not acceptable in the profession.