« In which I contemplate a few of Max Sawicky's "Ten Boxes of Heterodoxy" | Main | Roll out the barrel »
July 14, 2007
Boudreaux vs. Rodrik
Don Boudreaux scores the first hit, and it's a beauty.
I suspect that if someone proposed to Dani Rodrik that he explore the wealth-creating potential of state-level protectionism, he would refuse. He would likely (and correctly) say that it's ridiculous on its face to suppose that such protectionism would make the people of Tennessee as a group wealthier over time. If my suspicion is correct, then to what would Rodrik himself attribute his out-of-hand dismissal of the notion that Tennessee tariffs might well make Tennesseeans richer? Would he realize to his chagrin that he is a benighted, faith-based non-scholar? Or would he instead understand that the case for an extensive, market-driven division of labor is so strong -- and that the political border that separates Tennessee from other states is so economically meaningless -- that it would be as pointless for a serious economist to explore the economic potential of Tennessee protectionism as it would be for a serious oncologist to try to cure a patient of cancer by bleeding that patient with leeches.
Let me confirm Boudreaux's suspicion that I would indeed be against imposing intra-state trade restrictions in general (or to be more precise, that I would have a strong presumption against them). So the question he asks is an important one. Why then do I not take an equally strong position against trade restrictions in international trade?
The answer is that the parallel is misleading in this context. The two situations are alike only in the limiting (and counterfactual) case where government-imposed tariffs are the only transaction costs blocking economic exchange across international borders. In reality, national borders demarcate political and legal jurisdictions, which means that there remain plenty of transaction costs which block economic convergence. Capital flows are hindered by sovereign risk and the absence of international regulation and lender-of-last resort functions, which create the kind of syndromes that I often discuss in this blog. Labor mobility is severely restricted. And differences in regulatory regimes impose severe transaction costs (estimated by Jim Anderson and Eric van Wincoop to be of the order of 40% in tariff equivalents) on international trade. In the presence of these transaction costs, free trade in goods (in the sense of zero import tariffs) is in general incapable of achieving rapid economic growth and economic convergence in poorer nations of the world. If you do not believe this, just ask the Mexicans.
Within this U.S., economic convergence is achieved because there is a common constitution, a federal judiciary, nation-wide financial regulation, and free flow of labor. This ensures that a lagging region (such as the South until recently) catches up by a combination of capital coming in and labor moving out. Neither of these channels are operative in a world economy that is divided into nation-states. Removing restrictions on international trade in goods, services, and capital simply does not do it. Trade ends up being too small, and capital flows in the wrong direction (from poor to rich countries).
Integration normally proceeds (e.g. the European Union) from free trade area or customs union, to monetary union, to political union. Europe is trying to make the last step, and it's not an easy one. Here in the U.S., the framers of the constitution were smart enough to establish the fledgling country as a customs union and monetary union. This was in order to form a more perfect political union that that Articles of Confederation was unable to deliver.
Unfortunately, this does not stop the states and localities from pursuing other policies (wooing multinational factories, establishing tax-increment financing districts, etc.) that do with a series of knife cuts a bit of what a tariff would do with a hatchet blow.
Such behavior, when done by all, would result in something that resembles the suboptimal outcome of a prisoner's dilemma. There are, however, probably some asymmetries that would lead to some states winning a tariff war with other states. California vs. Hawaii would probably not be a fair fight, for example. Whether Tennessee has that sort of bargaining power is left to the reader to contemplate.
That there are strong arguments and historical precedent for the usual progression of these economic unions plays into Boudreaux's stand. It would be self-defeating to move backward.
And similarly, Rodrik's position is that where the political (and even monetary) unions do not exist, there is no reason to presume the same prohibition on trade protection you would expect in a political union like the U.S. He can't resist one more jab at Boudreaux.
There is of course the option of global federalism (creating a U.S. or an EU at the global level)--but that does not seem a realistic option anytime soon. I doubt that Don Boudreaux would go for it in any case.
Brad DeLong scores it a knockout for Boudreaux.
The knockout punch, of course, is that Dani Rodrik's country whose "labor force that is producing at low levels of productivity" is doing so because it has lousy political institutions: it lacks the "constitution... judiciary, nation-wide financial regulation, and free flow of labor" that have underpinned economic growth in the rich post-industrial core. The poor country is poor because its government is incompetent, and corrupt.
And yet Dani wants--in this situation--to enhance and extend the role and powers of the poor-country government by asking it to implement an active protectionist industrial policy because "there exists a bunch of arguments having to do with learning and (domestic) market failures under which subsidization of tradable activities could speed up your economic growth."
Certainly many countries are poor because their government is incompetent and corrupt. But I find it hard to accept this as a blanket statement. So does Rodrik, who responds:
No, the argument that poor countries are and remain poor because their governments are incompetent and corrupt is one of the absurd reductionisms of the day which I do not believe in and have written against. The point I made was that a poor country would have the real prospect of converging in living standards with rich countries if international economic integration were near-total (involving free labor mobility, truly integrated capital markets, and a transnational set of regulatory, legal and political institutions that underpin this integration). In the absence of these, trade liberalization does not get you there. You are in a second-best world and you need to think appropriately. The idea that developing countries cannot employ industrial policy in such a world to good effect is downright silly.
Here is a thought experiment: does anyone really believe that China would have grown as fast as it did if it had removed all its tariffs and trade restrictions in 1978, instead of liberalizing strategically and sequentially--first in agriculture, than in industry, then on the export side, and only later in the 1990s on the import liberalization side? There are many reasons why the Chinese strategy worked, but one of them is that it protected employment while industrial capabilities were being built up.
Rodrik's notion of the second-best is one that I employ frequently when I think about reasons why protection has been implemented by countries. While I understand the concept and am sympathetic to the notion, my mind always comes back to the fact that it is difficult to remove protection once it is in place--vested interests being what they are. Hence, I am much more disposed to allowing a country to go slowly and deliberately (like China is in many respects) toward a goal of more openness than I am to allowing a country to take steps backward and impose additional protection.
Completely free trade is neither necessary nor sufficient for fast growth initially. Free trade is necessary but not sufficient for broader and deeper economic integration. Those are two different goals, but they are both goals that many countries have for themselves. And both are reasons why free trade should be pursued, but not necessarily in its extreme form right away. Rodrik is correct that unilateral free trade does not make integration happen (not sufficient), but I would have liked him to address the desirability of multilateral free trade for further integration down the road.
Or, to answer Boudreaux (albeit in a way that he probably didn't intend), Tennessee would not do itself any favors by unilaterally abstaining from offering incentives to companies to locate there. But reducing state level competition of that sort would benefit everyone.
There's a big difference between China going slowly towards reform and the U.S. wanting to punish China for going too slowly.
UPDATE: Brad DeLong responds to Dani Rodrik. He's responding to Rodrik's comment about China.
...The fact that China's government had been so bad before 1978 meant that there were enormous problems and enormous opportunities for a government that was both competent and (relatively) benevolent--in the sense of wanting to see the economy grow rapidly if that could be accomplished without endangering its hold on power.
and...
...A lot of human resources and infrastructural capital was wasted becuase Deng Xiaoping did not dare risk the political consequences of the economic process of shifting resources out of the old Soviet-style industrial sector. I think he was right--from his perspective at least--to initially limit market manufacturing to the south. However, the logic is not economic but political.
No and yes. There is no disputing that the political motivation was real. But DeLong also notes the mass unemployment that would have occurred had the breaking down of the Soviet-style industrial sector happened more quickly. I think even a benevolent social planner who was immune to revolution would hesitate to put that in motion. Ultimately, I think this one is both economic and political, but if DeLong's point is that it was the political constraint and not the economic one that was binding I would not disagree. Furthermore, there is no doubt that the second best argument can be abused by governments that don't want to face up to their own shortcomings. DeLong's example of Peron's Argentina would qualify.
Of course now we've moved so far away from Boudreaux's original challenge that anything further deserves a new post. The lesson is that second best considerations can be legitimate, but because industrial policy--whether in five year plans or "temporary" tariffs--are hard to remove once in place, it is important to resist the temptation to apply the argument everywhere. Yet it seems to me that had China gone full-tilt for liberalization in 1978, they would look different today, and I'm not convinced it would be better.
Posted by William Polley at July 14, 2007 03:01 AM
Trackback Pings
TrackBack URL for this entry:
http://www.williampolley.com/cgi-bin/mt-tb.cgi/772
Comments
I think it is misleading to construe proposed actions as punishment rather than nullification of the effects of existing distorting policy, one which could easily be designed to be phased out as exchange adjustments are increased and allowed to float.
Posted by: Lord at July 15, 2007 03:26 PM
Well, I'm neither the first, nor the only one to use the word "punish". Do a Google search for "schumer china punish" and pretty much every news agency from here to Shanghai is represented.
Isn't the point of the legislation to get them to take corrective action on our terms or suffer a consequence?
Your point about the existing distortion is well taken, but that does not assuage my concern that this proposed action by us could touch off a trade war. My primary reason for this concern is grounded in a long study of the Chinese situation and my conclusion that a tariff by us is not likely to get them to revalue any faster. Our political systems are different. They are very patient, and so this legislation would probably not have the desired effect of correcting their behavior.
In assessing congressional attitude towards this situation, the word "hubris" comes to mind.
Posted by: William Polley at July 15, 2007 06:04 PM