Dani Rodrick has a paper on a old idea that I've seen popping up here and there recently.

I provide evidence that undervaluation of the currency (a high real ex- change rate) stimulates economic growth. This is true particularly for devel- oping countries. There is also some evidence that the operative channel is the size of the tradable sector (especially industry). These findings suggest that tradable goods suffer disproportionately from the government or market fail- ures that keep poor countries from converging towards higher-income levels. I present two categories of explanations as to why this may be so, focusing on (a) institutional weaknesses, and (b) product-market failures. A formal model elucidates the linkages between the level of the real exchange rate and the rate of economic growth.

Before people jump on the "devaluation leads to growth" theme, read carefully. Why do countries that have undervalued currencies grow more rapidly? The answer is that undervaluation compensates for the bad government policies. His tests confirm that undervaluation has no real growth benefit in countries with decent policies. (He also talks about various nebulous "market failures" being a possible culprit but admits there is little real evidence on this front.)

I understand that Rodrick is just doing positive analysis here and isn't necessarily advocating a policy of undervaluation. At most, he suggests undervaluation as a second best. The first best soluation being to fix the government (and supposed market) failures. But I worry about two things. First, it would be bad if policy makers began to think of undervaluation as a substitute for real economic reform. Second, there are unintended consequences to undervaluation--namely the threat of monetary instability and high inflation down he road which is a known problem for growth in its own right.

To me this looks like another example of a miracle drug that is so potentially toxic that we should put it on the shelf and label it too dangerous to take.