Let me be clear, Pedro and Renato dont need me to defend them. But the thing is: I get really pissed off when an inferior intelect like Oreiro goes around shouting idiotic economic ideas in a unpolite manner. And that's why I am back after a protracted slumber: to teach Oreiro and his claque a bit of economics. Arguably, a positive externality from my intervention is to convince society at large and maybe even Oreiro himself that he is not as smart as he thinks he is.
Ok, enough of bullshitting...let me go now to the point that really matters:
Oreiro enlists Rodrik and others to his argument, but even if Dani is a smart guy, he is wrong on this issue of Exchange rates and growth -- and have been told so already. Pay attention now Oreiro, 'cause it is not that hard to grasp this.
Developing economies growth rates are positevely correlated with savings as a share of their GDP. That much is true. But why? There are 2 possible explanations:
(i) they are still converging and capital markets are imperfect (thus domestic savings matter for capital accumulation);
(ii) there is a "frontier agency-problem" meaning that foreigners fear bringing capital and know-how and new technology to a country with shaky property rights. Because of this imperfection, they will be more willing to invest in a country if jointly with natives from the private sector (the frontier risks are smaller in this case since expropriating the local guy is politically harder). That in turn is more likely if domestic savings are larger -- otherwise there will be few co-investments.
Now, it is also true that the real exchange rate is more depreciated in countries that save more. This is because a weaker demand (higher saving) for internally produced non-tradables puts downward pressure on non-tradeable goods' prices (Pnt) but do not affect prices of tradables (Pt) -- which are determined internationally. Then, of course, Pt/Pnt is higher the higher the saving rate.
Finally, if a cross-country regression does not control for savings (as most dont), the correlation between growth and Exchange rate will be spurious -- omitted savings positevely correlates with both. Aghion and some friends ran this growth and exchange rate regressions controlling for savings...surprise surprise...the beta on the exchange rate loses significance.