Separating the micro from the macro

Dani Rodrik's blog has become a must-read for those interested in economic development. Recently, he's been writting quite a bit on the merits of industrial policy. Contrary to most orthodox economists, he argues that extensive market failures do justify intervention, at least in some cases (see here).

Now, industrial policy models come in many flavors. There's the East Asian development model, based on exports, and the old import substitution model (ISI)followed in Latin America and in India from the 1950's to the 1980's. Results were, obviously, very different.

But, as Rodrik points out, industrial policy falls within the realm of the micro. While it does have macro repercussions, it's absurd to blame it for events such as the Latin American debt crisis of the 1980's. After all, some countries with open economies, such as Chile, experienced severe crises at that time, while ISI-followers such as Colombia and India did not.

That makes The Economist's economics blog's screed against import substitution so bizarre. One can debate ISI's merits, but arguing that ISI caused the debt crisis because it hampered the recovery that came after (which is obvious) makes no sense.

This doesn't mean the two are completely separate. An argument can be made that by the 1970's the ISI model had run against diminishing returns, lowering growth. Facing acute political and social pressures, many Latin American governments worked to raise growth by massively (and recklessly) increasing public spending. This was possible because of the abundant liquidity in the 1970's, which enabled governments go get abundant, low-cost foreign currency loans. If Latin American nations had opened their economy earlier or developed more efficient ways to combat unrest, things may have turned out very differently.