Wednesday, August 31, 2011

Rodrik on the future of economic convergence

Dani Rodrik from Harvard has just published a really interesting paper on "The Future of Economic Convergence", where he argues--against very optimistic expectations of other pundits--that emerging markets are not likely to grow as fast as expected and--fully in line with what I have been arguing before--that predictions of high future growth rates for emerging markets "are largely extrapolations from the recent past and they overlook serious structural constraints"

Some useful quotes from the paper (a long list):

“Yet I find much of the optimism regarding the prospects for rapid convergence misplaced. In practice most of the convergence potential is likely to go to waste – just as it has since the world economy first got divided into a rich North and a poor South. As the empirical literature on growth has documented, convergence is anything but automatic. It is conditional on specific policies and institutional arrangements that have proved hard to identify and implement. Indeed, the recipes seem to vary from context to context. The experience of highly successful Asian countries is difficult to transplant in other settings”.

“It is true that the policy and institutional setting has improved across the developing world – at least as judged by conventional criteria. Developing countries have opened up to the world economy, place greater emphasis on macroeconomic stability, and are for the most part better governed. These changes have led many observers to think ―this time will be different.‖ My reading of the evidence is that these are improvements that serve mainly to enhance these economies‘ resilience to shocks and help avert crises, which often interrupted economic progress in the past. They do not necessarily stimulate ongoing economic dynamism and growth.”

“So generalized, rapid convergence is possible in principle, but unlikely in practice. Our baseline scenario has to be one in which high growth remains episodic. Sustained convergence is likely to remain restricted to a relatively small number of countries.”

“Easterly‘s bottom line is that empirical evidence gives little reason to have confidence that moderate changes in policies will yield systematic or sizable growth effects. Another way of putting the same result is to repeat the point made above: avoiding truly awful policies can prevent a country from turning into an economic basket case, but ―good‖ policies of the conventional type do not reliably generate high growth”

“In other words, the economy may be a mixture of activities that are already on the escalator up and activities that are going nowhere. Economies that grow rapidly are those that are able to push their resources into the escalator sectors. And those that grow in a sustained fashion are those that can accomplish this on an ongoing basis.”

“In other words, once a country begins to export something, it travels up the value chain in that product regardless of domestic policies or institutions.

“So convergence can be easy if an economy is able to push its resources (labor in particular) into the ―convergence sectors‖ – the industries on the automatic escalator up.”

“Why then do the conventional policies of macroeconomic stability, liberalization and openness not do the trick? After all, their objective is precisely to ensure that markets can work better and generate the requisite incentives. As a practical matter, however, creating well-functioning market economies requires considerably more than tinkering with specific policy instruments. It is a process that involves deeper institutional transformation measured in decades rather than years. Laws and regulations can be rewritten quickly, but that is not by and large where a nation‘s institutions reside. The rules of the game that we call ―institutions‖ are cognitive constructs that shape expectations about how other people behave (North 1990, Pistor 2000). These expectations are difficult to modify and replace, short of wars, occupation, revolutions, or other cataclysmic events. Furthermore, as long as the beneficiaries of the established order remain politically strong, they can easily circumvent reforms that undercut their privileges. As Daron Acemoglu and James Robinson have emphasized in their various writings, sustainable economic growth ultimately requires political change (Acemoglu and Robinson, forthcoming).”

“Consequently, structural change can remain too slow even when markets are liberalized, opened up, and made to work ―better‖ in the conventional manner. Growth requires remedies targeted at these ―special‖ sectors rather than general policies.”

“Of all methods of subsidizing modern tradables, perhaps the most effective is currency undervaluation. Growth-promoting structural change is greatly assisted by a highly competitive real exchange rate. In Rodrik (2008b) I show that there is a systematic and robust association between undervaluation and economic growth, a relationship that seems to work through undervaluation‘s positive effects on industrialization.”

“In fact, it is rather difficult to identify instances of nontraditional export successes in Latin America and Asia that did not involve government support at some stage”

“Currency undervaluation is often preferred for its non-selective nature, but that is actually a big problem in this context: undervaluation ends up subsidizing a lot of activities – traditional commodity exports, in particular -- that do not need to be subsidized while also unnecessarily taxing imports across the board.”

Here is also Rodrik's op-ed in FT on the same subject.

No comments: